362 Technical Requirement of 800 KV Substation and Above

362
TECHNICAL REQUIREMENTS
FOR
SUBSTATION EQUIPMENT EXCEEDING 800 kV
Field experience and technical specifications of Substation equipment up to 1200 kV
Working Group
A3.22
December 2008
WG A3.22
Technical Requirements
for
Substation Equipment exceeding 800kV
Field experience and technical specifications of Substation Equipment up to 1200 kV
Members:
H. Ito (Convenor), A. Janssen (Secretary),
J. Amon F., S-W. Bahng, M. C. Bhatnagar, P. Boss, J. Brunke, E. Colombo, R. Diaz, D. Dufournet, Y.
Filion, P. C. Fernandez, R. Gorur, A. Gilboulet, J. Jäger, A. Keri, T. Kobayashi, M. Kosakada, E. Kynast,
A. Lokhanin, C. van der Merwe, M. de Nigris, V. Rashkes, D. Peelo, B. Richter, H-D. Schlemper,
B. Shperling, Y. Shirasaka, R. Smeets, L. Stenström, M. Waldron, A. Wiersma, J-W. Woo, Y. Yamagata,
Yao Sili, R. Yeckley, T. Yokota, L. van der Zel
Copyright © 2008
“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 circulation 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 85873-049-0
1
1 Introduction ............................................................................................................................11
1.1 Introductory Remarks and Tasks of WG A3.22............................................................ 11
1.2 Specific phenomena peculiar to UHV AC systems ...................................................... 12
1.3 Summary of future tasks by equipment type (SC A3 only) .......................................... 12
1.3.1 Circuit-breakers ..........................................................................................................13
1.3.2 Disconnectors (DS), earthing switches (ES) and high speed grounding switches
(HSGS) ...................................................................................................................................13
1.3.3 Metal oxide surge arrester (MOSA)...........................................................................13
1.3.4 Instrument transformers .............................................................................................14
2 Field Experience.....................................................................................................................15
2.1 Overview of 800 kV and UHV AC projects exceeding 800 kV ................................... 15
2.2 AEP project ................................................................................................................... 16
2.2.1 Features of American Electric Power’s 800 kV transmission system .......................16
2.2.2 Overview of field experiences and the evolution of AEP’s 800 kV transmission
network...................................................................................................................................16
2.2.3 Specific issues and technical challenges for the operation of AEP projects ..............17
2.3 IREQ/Hydro Québec project ......................................................................................... 18
2.3.1 Features of Hydro Québec 765 kV transmission system ...........................................18
2.3.2 Evolution of Hydro Québec 765 kV transmission network .......................................19
2.3.2.1 Actual 765 kV transmission network .................................................................20
2.3.2.2 765 kV substation layout....................................................................................20
2.3.3 Overview of field experience and the evolution of Hydro-Québec’s 765 kV
transmission network..............................................................................................................20
2.3.4 Specific issues and technical challenges for the operation of Hydro Québec projects
21
2.4 Korea’s project .............................................................................................................. 22
2.4.1 Features of KEPCO 800 kV transmission system......................................................22
2.4.2 800kV GIS substation ................................................................................................23
2.4.3 Insulation level ...........................................................................................................24
2.4.4 Technical specifications of 765 kV power transformer .............................................25
2.4.5 800 kV Gas insulated switchgear ...............................................................................25
2.5 Brazil’s project .............................................................................................................. 26
2.5.1 Features of Furnas’ 800 kV transmission system.......................................................26
2.5.2 Overview of field experience of 800 kV long distance transmission lines................27
2.5.3 Evolution of Furnas’s 800 kV AC Transmission Systems.........................................28
2.5.3.1 Application philosophy for 800 kV Shunt reactor (SR).....................................30
2.5.3.2 Application philosophy for 800 kV Series Capacitor (SC)................................31
2.5.3.3 Other notable aspects of 800 kV transmission system .......................................34
2.5.4 Notable aspects of the ± 600 kV DC Transmission System ......................................35
2.5.5 Specific issues and technical challenges for the operation of Furnas’ long
transmission systems ..............................................................................................................36
2.5.6 Future challenges regarding UHV transmission in Brazil .........................................37
2.5.7 Summary ....................................................................................................................39
2.6 Italy’s project................................................................................................................. 40
2.6.1 Features of Italy’s 1000kV transmission system........................................................40
2.6.2 Overviews of field experiences of the projects ..........................................................41
2
2.6.3 Specific issues and technical challenges for the operation of projects ......................42
2.6.4 Overviews of field experiences of the projects ..........................................................44
2.7 BPA 1200 kV Lyons R&D project ............................................................................... 45
2.7.1 Features of BPA’s 1200 kV research project .............................................................45
2.7.1.1 Background ........................................................................................................45
2.7.1.2 Design of the Lyons facility ...............................................................................45
2.7.1.3 Line configuration ..............................................................................................45
2.7.1.4 Air gap clearances ..............................................................................................46
2.7.1.5 Line insulation....................................................................................................46
2.7.1.6 Line conductors ..................................................................................................46
2.7.1.7 UHV substation design.......................................................................................47
2.7.2 Overview of field experience of BPA’s 1200 kV project -Electrical test and
development program.............................................................................................................47
2.7.2.1 Corona Studies ...................................................................................................47
2.7.2.2 Electric Field Measurements..............................................................................49
2.7.2.3 Insulation studies................................................................................................50
2.7.3 Specific issues and technical challenges for the operations of BPA projects ............51
2.7.3.1 Transformer design ............................................................................................51
2.7.3.2 Surge Arrester design .........................................................................................51
2.8 Russia’s 787-1200 kV interconnected grid ................................................................... 52
2.8.1 General characteristic of Russia’s interconnected grid ..............................................52
2.8.1.1 Selection of nominal and maximum operating voltages ....................................54
2.8.1.2 Maximum operating voltage ..............................................................................55
2.8.1.3 Shunt reactor positioning and connection - Degree of line capacitance
compensation.........................................................................................................................55
2.8.1.4 System of limitation of temporary, switching and lightning overvoltages ........55
2.8.1.5 Levels of overvoltage limitation and arrester characteristics.............................56
2.8.1.6 Line parameters ..................................................................................................57
2.8.1.7 Auto-reclosing application .................................................................................58
2.8.2 Overviews of field experience and evolution of Russia’s 787 and 1200 kV
transmissions ..........................................................................................................................58
2.8.2.1 Organization of research, designing and testing ................................................58
2.8.2.2 Pilot program for the next possible UHV level..................................................59
2.8.2.3 Operating experience..........................................................................................59
2.8.2.4 Radio and audio interference, corona losses ......................................................60
2.9 TEPCO 1100 kV project ............................................................................................... 61
2.9.1 Features of TEPCO’s UHV transmission system ......................................................61
2.9.1.1 Considerations for UHV systems and equipment ..............................................62
2.9.1.2 Concept of insulation coordination and specifications ......................................62
2.9.2 Specific issues and technical challenges for substation equipment - Specifications for
substation equipment..............................................................................................................64
2.9.3 Specific issues and technical challenges for the operation of TEPCO’s 1100 kV
projects - Field tests................................................................................................................66
2.9.4 Summary ....................................................................................................................66
2.10 China’s project .............................................................................................................. 67
2.10.1 Brief summary of China’s 1100 kV projects .............................................................67
2.11 India’s project................................................................................................................ 69
3
2.11.1 Features of India’s transmission projects - Power generation and demand scenario in
India 69
2.11.1.1 Present transmission network of India ...............................................................69
2.11.1.2 Future transmission network of India (with time frame) ...................................70
2.11.2 Consideration for transmission development.............................................................70
2.11.3 Electrical design considerations for transmission lines..............................................71
2.11.4 Preliminary study results for 1200 kV UHV AC systems .........................................72
2.11.5 Overvoltage studies ....................................................................................................72
2.12 Summary ....................................................................................................................... 74
3 Specific topics ........................................................................................................................77

3.1 Insulation coordination.................................................................................................. 77
3.1.1 Power frequency voltage............................................................................................78
3.1.2 Temporary overvoltage (TOV) ..................................................................................79
3.1.3 Slow front overvoltage (SFO) ....................................................................................80
3.1.4 Fast front overvoltage (FFO)......................................................................................84
3.1.5 Insulation levels and arrester protection levels ..........................................................86
3.1.6 Comparison with insulation withstand voltages calculated as per IEC procedure ....87
3.1.6.1 Slow front overvoltages .....................................................................................87
3.1.6.2 Fast front overvoltages .......................................................................................87
3.1.6.3 Comparison ........................................................................................................88
3.1.7 Very fast transient overvoltage (VFTO) ....................................................................89
3.1.8 Clearances ..................................................................................................................89
3.2 Transformers ................................................................................................................. 91
3.2.1 Insulation level ...........................................................................................................91
3.2.2 Equivalent surge capacitance .....................................................................................92
3.3 Circuit Breakers............................................................................................................. 93
3.3.1 TRV for terminal fault test duties ..............................................................................93
3.3.1.1 General ...............................................................................................................93
3.3.1.2 Inherent TRV calculated based on Japan’s UHV system ..................................94
3.3.1.3 First pole-to-clear factor (kpp) for terminal fault test duties..............................96
3.3.1.4 Amplitude factor (kaf) for terminal fault test duties ..........................................98
3.3.1.5 Rate of rise of recovery voltage (RRRV) for terminal fault test duties .............99
3.3.1.6 First reference voltage (U1) for terminal fault test duties..................................99
3.3.1.7 Effect of MOSAs with different characteristics on TRV.................................100
3.3.1.8 Calculation of TLF fault current ......................................................................101
3.3.2 TRV of compensated/uncompensated lines .............................................................101
3.3.2.1 Series compensated lines and their effect on circuit breakers recovery voltage
and switching surges ...........................................................................................................101
3.3.2.2 Effects on the lines CB TRVs during line fault interruption............................102
3.3.2.3 Turkey’s network .............................................................................................103
3.3.2.4 Hydro-Québec experience................................................................................103
3.3.2.5 BC Hydro experience .......................................................................................104
3.3.2.6 Solutions to reduce TRVs ................................................................................104
3.3.2.7 Effects on switching surges..............................................................................106
3.3.2.8 Shunt compensated lines and their effect on circuit breakers recovery voltage
and switching surges ...........................................................................................................106
3.3.3 Characteristics of short-line faults............................................................................106
4
3.3.3.1 General .............................................................................................................106
3.3.3.2 Single-phase faults as the basis for rating ........................................................107
3.3.3.3 Line surge impedance and bundle contraction .................................................107
3.3.3.4 Peak factor “k” ................................................................................................110
3.3.4 DC time constant......................................................................................................110
3.3.4.1 General .............................................................................................................110
3.3.4.2 DC time constant in UHV system in Japan......................................................111
3.3.4.3 Time constant for T100a and UHV applications .............................................112
3.3.5 TRV with/without opening resistors ........................................................................114
3.3.5.1 General .............................................................................................................114
3.3.5.2 Effect of opening resistors on TRV..................................................................114
3.3.5.3 Category of fault current and modes ................................................................115
3.3.5.4 TRV for main interrupters with/without opening resistors ..............................115
3.3.5.5 TRV for resistor interrupter..............................................................................119
3.3.5.6 Evaluation of switching duties with opening resistors.....................................120
3.3.6 Phenomena related to out-of-phase ..........................................................................121
3.3.6.1 General .............................................................................................................121
3.3.6.2 Effect of opening resistors under out-of-phase conditions...............................125
3.3.6.2.1 Phase shift ...................................................................................................126
3.3.6.2.2 Amplitude of line side oscillation...............................................................126
3.3.6.2.3 Other parameters.........................................................................................126
3.3.6.2.4 Simplified calculation .................................................................................126
3.3.6.2.5 Summary.....................................................................................................127
3.3.7 High frequency components imposed on fault currents...........................................128
3.3.8 Capacitive current ....................................................................................................130
3.3.8.1 General .............................................................................................................130
3.3.8.2 Voltage factor for UHV....................................................................................131
3.3.8.2.1 Switching overhead-lines under normal conditions ...................................131
3.3.8.2.2 Switching overhead-lines under fault conditions .......................................131
3.3.8.3 Shunt compensated lines ..................................................................................132
3.3.8.4 Current..............................................................................................................132
3.3.9 UHV reactor current switching ................................................................................132
3.3.9.1 General considerations .....................................................................................132
3.3.9.2 TRV limitation .................................................................................................134
3.3.9.3 UHV shunt reactor switching circuit breaker requirements.............................135
3.3.9.4 Discussion and conclusion ...............................................................................136
3.3.10 Controlled switching ................................................................................................136
3.3.10.1 General .............................................................................................................136
3.3.10.2 Capacitive switching applications....................................................................138
3.3.10.3 Reactor switching applications.........................................................................139
3.3.10.4 Unloaded transformer energization..................................................................139
3.3.10.5 Uncompensated and compensated line switching............................................142
3.3.10.6 Summary ..........................................................................................................145
3.4 DS, ES ......................................................................................................................... 147
3.4.1 DS capacitive switching ...........................................................................................147
3.4.1.1 Introduction ......................................................................................................147
3.4.1.3 UHV air-insulated disconnectors .....................................................................150
5
3.4.1.4 UHV GIS disconnectors...................................................................................150
3.4.1.5 VFTO and its countermeasure..........................................................................151
3.4.1.6 Analysis of VFTO in full GIS ..........................................................................151
3.4.1.7 Analysis of VFTO in hybrid GIS .....................................................................154
3.4.1.8 Requirement for the switching resistor of DS ..................................................156
3.4.1.9 Considerations for structure with switching resistor........................................156
3.4.1.10 Discussion ........................................................................................................157
3.4.2 Bus transfer current switching..................................................................................158
3.4.2.1 Introduction ......................................................................................................158
3.4.2.3 Discussion ........................................................................................................160
3.4.2.4 Bus-transfer switching in Japanese UHV Project ............................................160
3.4.3 Induced current switching for earthing switches (ES) .............................................161
3.4.3.1 General .............................................................................................................161
3.4.3.2 Duty for 1100kV Earthing switch (ES)............................................................162
3.5 Secondary arc current extinction, single-phase auto reclosing ................................... 164
3.5.1 General .....................................................................................................................164
3.5.1 Operating shunt reactors, four-legged shunt reactors...............................................168
3.5.2 HSGS........................................................................................................................169
3.5.2.1 Developments of HSGS for UHV system in Japan..........................................169
3.5.2.2 Delayed zero phenomena caused by successive faults.....................................171
3.5.2.3 Field experiences of HSGS in KEPCO 765kV systems ..................................172
3.6 Surge Arrester ............................................................................................................. 173
3.6.1 Continuous operating voltage Uc, rated voltage Ur and TOV capability ................173
3.6.2 Residual voltages.........................................................................................................175
3.6.3 Long duration current impulse withstand....................................................................177
3.6.4 Lightning impulse discharge capability ......................................................................179
3.6.5 High current impulse withstand capability .................................................................179
3.7 Radio Interference, Corona Losses ............................................................................. 179
3.7.1 General Aspects........................................................................................................179
3.7.2 Eskom 765 kV transmission line and impact of RIV, CL and AN on the conductor
bundle design........................................................................................................................181
3.7.3 Italy’s 1050 kV experience ......................................................................................182
3.7.3.1 Radio Interference ............................................................................................182
3.7.3.2 Audible noise....................................................................................................184
3.7.4 TEPCO 1100 kV experience ....................................................................................185
3.7.4.1 Basic concept of corona noise reduction..........................................................185
3.7.4.2 550kV substation..............................................................................................186
3.7.4.3 UHV (1100kV) substation ...............................................................................186
3.7.4.3.1 Conductor design of UHV transmission line..............................................186
3.7.4.3.2 Practical design of substation equipment ...................................................188
3.7.5 BPA 1200 kV experience.........................................................................................189
3.7.5.1 Radio Interference ............................................................................................189
3.7.5.2 Audible noise....................................................................................................190
3.7.5.3 Electric Field ....................................................................................................190
3.7.5.4 TV interference ................................................................................................190
3.7.6 Russian 1150 kV experience ....................................................................................190
6
4 Technical Specifications of 800 kV Substation Equipment.................................................196
4.1 800 kV Power Transformers ....................................................................................... 196
4.2 800 kV Voltage Transformers (VT), Current Transformers (CT) .............................. 197
4.2.1 800 kV VTs ..............................................................................................................197
4.2.2 800 kV CTs ..............................................................................................................197
4.3 Circuit Breakers........................................................................................................... 198
4.3.1 Dielectric requirements ............................................................................................198
4.3.2 Making, breaking and switching test requirements..................................................200
4.3.2.1 TRV for terminal faults ....................................................................................200
4.3.2.2 DC time constant..............................................................................................201
4.3.2.3 Single-phase fault tests.....................................................................................201
4.3.2.4 Short-line-fault .................................................................................................201
4.3.2.5 Out-of-phase.....................................................................................................201
4.3.2.6 Capacitive current switching............................................................................201
4.3.3 Requirements for Inductive Load switching ............................................................203
4.4 DS, ES ......................................................................................................................... 203
4.5 High-Speed Grounding Switch (HSGS) ..................................................................... 204
4.5.1 General .....................................................................................................................204
4.5.2 Design considerations of HSGS...............................................................................204
4.6 Surge Arresters ............................................................................................................ 205
4.7 Shunt Reactors............................................................................................................. 207
4.7.1 General .....................................................................................................................207
4.7.2 Additional information related to Russia’s 787-1200 kV shunt reactors.................208
5 Technical Specifications of Substation Equipment 1000 kV and above .............................211

5.1 UHV Power Transformers .......................................................................................... 211
5.2 UHV Voltage Transformers (VT), Current Transformers (CT) ................................. 212
5.2.1 UHV VTs .................................................................................................................212
5.2.2 UHV CTs..................................................................................................................213
5.3 Circuit Breakers........................................................................................................... 214
5.4 DS, ES ......................................................................................................................... 215
5.5 High-Speed Grounding Switches (HSGS) .................................................................. 216
5.6 Surge Arresters ............................................................................................................ 217
5.6.1 Continuous operating voltage, Uc, rated voltage, Ur, and TOV capability .............217
5.6.2 External insulation strength......................................................................................217
5.6.3 Residual voltages and classifying currents...............................................................217
5.6.4 Energy requirements ................................................................................................218
5.6.5 Lightning impulse discharge capability ...................................................................218
5.6.6 High current impulse withstand capability...............................................................219
5.6.7 Pollution performance ..............................................................................................219
5.6.8 Mechanical strength .................................................................................................219
5.6.9 Short-circuit rating ...................................................................................................219
5.7 Shunt Reactor .............................................................................................................. 219
6 Conclusions and Future tasks...............................................................................................222
7
Abbreviations
1LG Single-phase line fault to ground
2LG Two-phase line faults to ground
3LG Three-phase line faults to ground
ACSR Aluminium Conductor Steel Reinforced

AEP American Electric Power (USA)
AIS Air Insulated Switchgear
AN Audible-Noise
BPA Bonneville Power Administration (USA)
CCGT Combined Cycle Gas Turbine

CEPRI China Electric Power Research Institute (China)
CESI Centro Elettrotecnico Sperimentale Italiano (Electrotechnical Testing
Centre, Italy)
CISPR International Special Committee on Radio Interference
CRIEPI Central Research Institute of Electric Power Industry (Japan)
CSS Controlled Switching System
CT Current Transformer
DS Disconnector
EHV Extra High Voltage

EMI Electromagnetic Immunity
EMTP Electro Magnetic Transient Program
ENEL Ente Nazionale Per L'energia Elettrica (National Electrical Energy,
Italy)
ES Earthing Switch
ESDD Equivalent Salt Deposit Density
FFO Fast Front Overvoltage

FPD Fast Protective Device
8
GCB Gas Circuit Breaker
GIS Gas Insulated Switchgear
HSGS High Speed Grounding Switch
IEC International Electrotechnical Commission

IREQ Hydro-Québec's Research Institute (Canada)
ITRV Initial Transient Recovery Voltage
ITU International Telecommunication Union
KEPCO Korea Electric Power Corporation (Korea)
LDC Line Discharge Class

LI Lightning Impulse
LIPL Lightning Impulse Protection Level
LIWV Lightning Impulse Withstand Voltage
LLF Long Line Fault
MAIS Manœuvres Automatiques d’Inductances Shunt

(Measure of automatically shunt reactor switching)
MOSA Metal Oxide Surge Arrester
MOV Metal Oxide Varistor
NCI Non-Ceramic Insulator
PD Optical potential devices

PIR Pre-Insertion Resistor
RI Radio Interference
ROW Rights-of-way
RPTC Rejet de Production et Télédélestage de Charge (Scheme to reject a
large amount of generation and to shed large block of load)
RRRV Rate of Rise of Recovery Voltage
RIV Radio Interference Voltage
9
SA Surge Arresters
SC Series Capacitor
SFO Slow-Front Overvoltage
SiC Silicon Carbide
SIPL Switching Impulse Protection Level
SIWV Switching Impulse Withstand Voltage
SN ratio Signal to Noise ratio
SPAR Single-Phase Rapid Auto-Reclosing
SPSR Solutions aux Problèmes de Séparation de Réseaux (Scheme is a last-
resort solution in the case of a risk of total collapse)
SR Shunt Reactor
SVC Static Var compensator
TCSC Thyristor-Controlled series Capacitor

TEPCO Tokyo Electric Power Company (Japan)
TLF Transformer Limited Fault
TOV Temporary Overvoltage
TPAR Three-Phase Rapid Auto-Reclosing
TRV Transient Recovery Voltage
UCTE Union for the Coordination of Transmission of Electricity in continental

Europe
UHV Ultra High Voltage (exceeding 800 kV)
URW Required Withstand Voltage
VEI Russian Electrotechnical Research Institute (Russia)

VFT Very Fast Transient
VFTO Very Fast Transient Overvoltage
VNIIE Russian Electric Power Research Institute (Russia)
VT Voltage Transformer
10
1 Introduction
1.1 Introductory Remarks and Tasks of WG A3.22
In recent years there has been a renewed interest, both economically and technically, in the
development of high voltage ac transmission systems rated in excess of 800kV (known as Ultra
High Voltage exceeding 800 kV, UHV). Plans for new commercial developments in countries
such as China & India, in conjunction with plans to extend existing trials in countries such as
Japan have created a real demand for standardisation in the field of UHV technology. Effective
standardisation of UHV requires that experience gathered to date, both from the planned
installations and from the major work carried out in Japan, Italy and Russia in the past, should be
drawn together and “best practise” identified. In 2006, CIGRE undertook to carry out this pre-
standardisation activity and this Technical Brochure in the first stage in completing this task.
UHV pre-standardisation involves, and impacts upon, many different CIGRE Study Committees
as shown in Table 1.1.1. In order to progress the early stages of the pre-standardisation activities
as fast as possible, CIGRE Working Group A3.22 took a CIGRE-wide lead in surveying the
current state of the art in the field of UHV technology. The resulting findings and
recommendations, developed in cooperation with other Study Committees and Working Groups
within CIGRE, are presented here and form the basis of future work in a number of different
technical areas.
Table 1.1.1 Standardisation tasks and related Study Committee
Standardisation tasks Related Study Committee
Review of international experience A2, A3, B2, B3, C4, D1
Chinese projects, state-of-the-art A2, A3, B2, B3, C4, D1
Insulation coordination and clearance in air / SF6 C4, (D1, A2, A3, B3)
Long transmission lines including series compensation, etc A3, C4, (B2, A2,), WG A3.13
Switching phenomena and transients A3, (C4)
Special requirements and designs for UHV substation equipment A3, (A2, B2)
(TRV, high DC component, line faults, etc)
Measurement challenges: Testing and service D1, (A3)
VT and CT designs A3, WG A3.15
Composite insulators: Substation equipment and overhead lines WG A3.21, B2, (A2, B3, D1)
Environmental impact: Focus on overhead lines B2, (C3)
Operating environment: Impact of pollution, wind, ice, C4, (A2, A3, B2, B3)
earthquake, etc, on performance
Electric Magnetic Field Influence B3, B2
Transformer design and materials A2, (D1)
Surge arresters and optimum switching practises A3, (C4), WG A3.17
Optimum substation and equipment layouts B3, WG B3.22
Designing overhead lines for optimum reliability B2, (C4)
Considering the document in more detail, it has been prepared by 39 international experts from
17 countries based on more than 300 written technical contributions. It is divided into three main
sections dealing with:
• the present state-of-the-art of technical specifications for substation equipment at 800kV
and above including general field experience and technical challenges in the operation of
various national projects.
11
• phenomena unique to transmission systems rated 800 kV and above including
extrapolations and considerations for future standards for substation equipment applied to
UHV systems.
• recommendations regarding specific aspects of technical specifications for systems rated
>800kV.
Future work within the various affected CIGRE Study Committees will look to expand upon
these topics as necessary in order to provide the necessary basis for standardisation. In particular,
WG A3.22 will continue to consider aspects related to switching devices and switching
phenomena.
1.2 Specific phenomena peculiar to UHV AC systems

Table 1.1.2 shows several specific issues of UHV AC systems exceeding 800 kV that may have
great impact on substation equipment and which are analysed in this Technical Brochure. In
particular, the use of multi-bundle conductors with large-diameter as well as large-capacity power
transformers provides distinctive phenomena for UHV systems.
Table 1.1.2 Specific issues of UHV AC systems exceeding 800 kV
Phenomena peculiar to UHV Equipment CIGRE SC IEC TC
Prominent Ferranti effect and TOV due to large Surge Arrester A2, A3, B3, C4 14, 17, 37
capacitance of overhead lines Shunt Reactor
Possibly reduced corona onset voltage with Line
increased corona losses and audibel noise Substation (AIS) B2, B3 11
Prolonged secondary arc extinction time due to 4-Legged Reactor A2, A3, B3, C4 14, 17, 28
higher induced voltage HSGS, GCB
Higher slow front overvoltage at grounding fault Circuit Breaker A3, B3, C4 17, 37
occurence due to low damping of traveling waves Surge Arrester
High amplitude factor in TRVs due to low losses of Circuit Breaker
power transformers and transmission lines Surge Arrester A3, B3, C4 17, 37
High TRV peak value for out-of-phase due to low Circuit Breaker A3, B3, C4 17, 37
damping of traveling waves Surge Arrester
Reduced line surge impedances due to multi- Circuit Breaker A3, B3 17
bundle conductors with large diameter
Large time constant of DC component in fault
current due to low losses of transformers and lines Circuit Breaker A3, B3 17
Reduced first-pole-to-clear factor due to small Circuit Breaker A3, B3 17
zero-sequence impedance in the UHV systems
Severe VFTO due to geometry and topology of Circuit Breaker A3, B3 17
UHV substation such as MTS GIS, MTS
MTS: Mixed Technologies Substation
In addition to those items listed above, the use of high performance surge arresters to effectively
limit overvoltages forms an integral part of the design of some UHV systems. The impact of this
design approach and its impact on specifications are also discussed herein.
1.3 Summary of future tasks by equipment type (SC A3 only)

The following sub-sections summarise the main technical topics identified as being of importance
for UHV equipment within the scope of Study Committee A3 and which will be developed
12
further within appropriate Working Groups. It is expected that, on the basis of the information
presented here other Study Committees will be able to draw up similar summaries and develop
the necessary work plans to complete pre-standardisation activities.
1.3.1 Circuit-breakers
• Short-circuit currents
DC time constants, auto-reclosing time, rated peak withstand current, fault current
with high frequency component, etc
• Transient Recovery Voltages (TRV) for terminal faults
First-pole-to-clear factor, Rate of Rise of Recovery Voltage (RRRV), Transformer
Limited Faults (TLF), Transformer secondary faults, Single phase and three phase
faults, effect of MOSA, etc
• Parameters for line fault interruptions
Surge impedance, amplitude factor, line side time delay, 1LG/2LG/3LG, compact
line, ITRV characteristics, TRV for Short-Line Fault (SLF), Long-Line-Fault (LLF),
series compensated line and by pass switch, etc
• Out-of-Phase switching (making and breaking)
Amplitude factor, RRRV, breaking current, probability of occurrence, out-of-phase
angles, synchronization, etc
• Parameters for no-load line-charging current switching
Capacitive current, shunt compensated lines, voltage factor, induced-voltages,
healthy phase switching, Temporary Overvoltage (TOV), etc
• Parameters for line energization and line reclosing
With/without MOSA, pre-insertion resistor (PIR), controlled switching, shunt
reactors, Metal Oxide Varistor (MOV) across interrupters, etc
• Reactor switching
• Unloaded transformer energization
• Effect of closing and opening resistor on the duties
TRV for terminal faults and Long-Line Fault (LLF) interruption, etc
1.3.2 Disconnectors (DS), earthing switches (ES) and high speed

grounding switches (HSGS)
• Disconnector switching parameters (capacitive, bus-transfer etc)
Amplitude factor, capacitive current, DS for GIS including Very Fast Transient
(VFT), transformer energization, etc
• Earthing switch induced current switching parameters
Peak voltage of TRV, breaking current, etc
• HSGS parameters
Peak voltage of TRV, secondary arc current, comparison with a four-legged shunt
reactor
1.3.3 Metal oxide surge arrester (MOSA)

• V-I characteristics
• Protection level of surge arrester, energy absorbing capability
13
• MOSA layout
Temporary Overvoltage (TOV), energy/currents due to lightning and switching
events, different location of MOSA, coordination current, etc
1.3.4 Instrument transformers

• Very fast transient (VFT) for Air Insulated Switchgear (AIS)/ Gas Insulated Switchgear
(GIS), etc
14
2 Field Experience
2.1 Overview of 800 kV and UHV AC projects exceeding 800 kV
The rapid growth in electrical power demand, especially in China, Brazil and India, creates some
urgency for the development and construction of UHV transmission systems exceeding 800 kV.
Figure 2.2.1 indicates the highest voltage of AC power transmission.
American Electric Power (AEP) began their installation of 345 kV transmission systems in 1952
at the same time that Sweden was building 380 kV lines. Hydro-Québec developed the world’s
first 735 kV transmission system in 1965, and AEP installed 765 kV lines in 1969. In Brazil, the
operation of the first EHV transmission systems, with the voltage levels of 345 kV, 440 kV, 550
kV and 800 kV, started in 1963, 1971, 1975 and 1982, respectively. Tokyo Electric Power
Company (TEPCO) started 550 kV transmission in 1973. At the beginning of the 1970s, ENEL
(in cooperation with CESI SpA) and TEPCO each launched UHV projects and conducted various
field demonstrations of UHV substation equipment in the 1990s.
TEPCO started construction on an 1100 kV double-circuit transmission line in 1988, completing
the first section (190 km) in 1993 and the second section (240 km) in 1999. These transmission
lines are now operated at 550 kV and will be upgraded to 1100 kV in the mid-2010s.
The voltages in the former USSR (Russia) were upgraded to 420 kV in 1957, 525 kV in 1959,
787 kV in 1967, and 1200 kV in 1985. Their UHV system is now temporarily operating at 525
kV.
In China, the first 550 kV projects were commissioned in 1981 and 765 kV transmission was
started in 2005. An 1100 kV pilot project is now being constructed and is scheduled to start
transmission in 2008.
In India, a “supergrid” high-capacity 1200 kV AC system along with a ±800 kV HVDC system is
being planned.
Highest voltage of AC power transmission （kV）
）
1500
1200kV 1200kV
(1985-91,USSR) 1100kV field tests (2012-,India)
787kV (1996-,Japan) 1100kV
1000
(1967-,USSR) (2008-,China)
765kV 800kV
420kV (1965-,Canada) (USA, South Africa, Brazil, Korea, China)
(1957-,USSR)
500
380kV
(1952-,Sweden)
year
1950 1960 1970 1980 1990 2000 2010 2020
Fig. 2.1.1 Highest voltage of AC power transmission
15
2.2 AEP project
2.2.1 Features of American Electric Power’s 800 kV transmission

system
American Electric Power (AEP) started conducting studies on Extra High Voltage (EHV) in 1946
and continued through to 1961. Their 345 kV transmission systems were overlaid on existing HV
transmission systems. AEP was one of the leaders in the deployment of 345 kV. AEP started
800 kV transmission in 1969 and presently owns more than 3400 km of nominal 765 kV lines
and 24 major substations, integrating key generating plants with load centres throughout the
eastern AEP system and providing interconnections with neighbouring utilities. AEP has learned
a great deal about 800 kV technologies through first-hand experience. [1]
2.2.2 Overview of field experiences and the evolution of AEP’s 800 kV

transmission network
Over the years, AEP has accumulated abundant experience in planning, designing, engineering,
constructing and operating 800 kV transmission networks. This experience involved numerous
technical challenges and lessons learned in the field including improvements to existing
technologies and development of new technologies.
During the 1980s, several 765 kV auto-transformers failed and seven newly installed single-phase
EHV transformers faulted at one substation during a period of three years. Most failures occurred
under normal operating conditions and were not accompanied by a major external disturbance.
An investigation revealed specific dielectric failure mechanisms. In response to these events,
AEP changed the dielectric specifications for EHV transformers in four key areas. [2] Since the
modification, no failures have been recorded for 765 kV transformers incorporating the new
dielectric specifications.
Four-legged shunt reactor banks are often used to reduce the secondary arc current on 365-
550 kV-transposed lines. Since AEP’s 800 kV transmission lines are not transposed, a
conventional four-legged shunt reactor bank cannot sufficiently reduce the secondary arc current
on a long line because of the difference in inter-phase capacitance. Therefore, AEP developed a
modified four-legged shunt reactor bank [3] to secure single-phase auto-reclosing of their 800 kV
transmission lines. Figure 2.2.1 shows a modified four-legged shunt reactor scheme.
Conventional 4 Legged Modified 4 Legged

Shunt Reactor Shunt Reactor
Figure 2.2.1 Modified reactor schemes
16
Field tests demonstrated that successful single-phase reclosing could be achieved if the secondary
arc current were limited to less than 40A and extinguished within 0.5 seconds. Since the mid-
1980s, single-phase auto-reclosing has been successfully operated once a year per line on average
and has greatly enhanced system performance.
The majority of the AEP’s 800 kV circuit breakers are live-tank type with closing resistors in
order to limit switching overvoltages to 2.0 p.u. In the recent years, AEP installed dead-tank
(DT) breakers with no closing resistors. However, arresters were installed along the associated
lines in order to limit switching overvoltages to 2.0 p.u. The first transmission line arresters
designed for vertical mounting were installed on the 240 km Marysville-Orange-Kammer 800 kV
line. EMTP studies showed that a three-phase line arrester set at the one-third and two-thirds
locations was required to keep the magnitude of switching surges below the line insulation
withstand of 2.0 p.u. The first 800 kV SF6 dead-tank (DT) circuit breakers were also installed at
Oregon Station. AEP currently has 16 DT GCBs with closing resistors to provide improved
overvoltage control.
Table 2.2.1 AEP specifications
Specifications
Items Circuit breakers Disconnect switches
Highest voltage 800 kV 800 kV
LIWV (kV) 2050 kV-close, 2255 kV-open 2050 kV
SIWV (kV) 1425 close/1550 kV-open (wet), 1400 kV, 1750 kV-open
1700 close/ 1870 kV-open (dry)
AC voltage 960 kV-open and close 850 kV phase to ground
Metal Oxide Surge Arresters (MOSA) main characteristics
Maximum continuous 476 kV
operating voltage
Protective levels 1420 kV (lighting surge at 20 kA, 8 µs/20 µs)
1197 kV (switching surge at 3 kA)
Circuit breakers main characteristics
First-pole-to-clear factor 1.5
TRV peak value 1649 kV for T10
1408 kV for T100
DC time constant X/R=17 for 63 kA and X/R=50 for 50 kA
2.2.3 Specific issues and technical challenges for the operation of

AEP projects
Throughout the 1970s and 80s, AEP continued their research and development on UHV
transmission. The UHV Research Centre located in India was a single-phase facility energized
from 1110/ 3 to 2255/ 3 kV line-to-ground using two transformers connected in cascade. The
facility included a single-phase test line consisting of three spans (each 305 m long), a station
with a 420/835/1785 kV 333 MVA transformer and two test cages with an independent voltage
source for testing short lengths of conductor. The maximum test voltage was 900 kV.
AEP conducted extensive research on conductor corona performance with various configurations.
Corona performance was evaluated in terms of environmental impact, including audible and radio
noise, television interference and corona loss under expected adverse weather conditions. [4]
17
Their UHV research showed that the 1.6 p.u. switching overvoltage level was achievable using
techniques such as closing and opening resistors, controlled switching, surge arresters, shunt
reactors, etc. It was concluded that there were no major impediments to the application of UHV
technology up to the 1500 kV level. However, slower load growth prevented AEP from pursuing
further UHV investigation.
2.3 IREQ/Hydro Québec project
2.3.1 Features of Hydro Québec 765 kV transmission system

The power system of Hydro-Québec is characterized by a large number of high-capacity
hydroelectric power plants located at great distances from the major load centres. This situation
has basically determined the design of the transmission system. Approximately 85% of the total
installed generation feeding the Hydro-Québec system is located at distances varying from about
400 to 1300 km from the closest major load centres (Montréal and Québec City).
Consequently, the transmission system is essentially of a radial nature with two major
transmission axes, northeast and northwest, which connect at a point located about 200 km north
of Québec City at the Saguenay Substation. The northeast axis delivers power generated from the
Manicouagan-Outardes and Bersimis Complexes (installed generation capacity ≈15,000 MW)
while the northwest axis delivers power generated by the La Grande River Complex with a total
of 15,500 MW of installed generation capacity.
Fig.2.3.1 Hydro-Québec 735kV series-compensated transmission systems
18
With this type of configuration, the transmission design essentially depends on the performance
criteria, which specify how the power system should react when subjected to different
disturbances. Basically, the system should be designed to maintain full synchronous operation
without any loss of load under so-called normal contingencies. It became clear from the very
beginning (early 1960s) that in view of the power transmission requirements expected from the
HQ power system, an extra high voltage (EHV) transmission system would be necessary to cope
with stability problems and in fact, this led to the selection of 765 kV AC technology. Using this
EHV system allowed a reduction in series impedance of the lines (relative to the impedance of
the generators) and consequently allowed the transmission of large amounts of power over long
distances and the ability to maintain the stability criteria.
Since the series impedance between generators is the most important factor for stability issues,
the number of switching substations has been of prime importance in view of its impact on post-
contingency system impedance since normal contingency results in the loss of a line section
between adjacent substations. This stability study resulted in the selection of line length varying
from 170 to 400 km on the 765 kV systems.
Following investigative studies and the decisions taken in 1962 to develop a 765 kV system for
the integration of the Manicouagan and Churchill Falls Complex, the first sections of the 765 kV
system were commissioned in August 1965, allowing the transmission of power over a distance
of about 650 km.
2.3.2 Evolution of Hydro Québec 765 kV transmission network

As mentioned earlier, harnessing of the rivers on the eastern side of James Bay could generate up
to 15,000 MW that would have to be transmitted over a distance of 800-1200 km to reach the
main load centres. Therefore, at least four possibilities were studied: AC at 765 kV, AC at
1100 kV and ±500 and ±800 kV DC. From these extensive studies, the following conclusions
were drawn:
• Experience acquired from the 765 kV systems at that time, and system studies combined
with numerous discussions with major equipment manufacturers resulted in a high level
of confidence about the technical feasibility of operating an 1100 kV system for the
integration of the James Bay Complex. Detailed studies on the transmission lines at
1100 kV have been performed. These studies essentially focused on a new type of tower
design and also on the possibility of using expanded conductors. For dynamic stability
issues, it was deemed necessary to add at least 1000 Mvar of voltage support
(synchronous condensers were considered at the time) depending on the scenario (number
of circuits used to transmit the entire 16,000 MW of power from the James Bay
Complex). Corona loss, radio interference and audible noise related to a UHV system
were issues that required more detailed study. Four conductor bundle configurations
(6×1823", 6×1998", 8×1630" and 8×1863") were considered and tested at IREQ in
outdoor test cages under heavy artificial rain and fair weather conditions at the beginning
of the 1970s. The results revealed that of the four bundles tested, the 8×1630" conductor
gave the best corona performance. However, ice loading conditions may lead to selecting
the 6×1863" bundle.
• For DC current transmission scenarios, the selection of voltage level and rated current are
of prime importance. This technology does not involve system dynamic stability; indeed,
a compromise in the number of circuits had to be found in order to deliver the power with
optimum flexibility and reliability while minimizing investment costs. Using a minimum
19
numbers of circuits would certainly reduce the investment costs but on the other hand,
since losses are proportional to the square of the current, the best compromise had to be
made between acceptable loss and capital investment. Different scenarios at voltage
ranging between ±450 kV and ±800 kV have been studied. The rated current varied from
1600 to 3500 A
• The outline on investment costs for the different scenarios studied for the transmission of
power generated by the James Bay Complex indicates an increase in the order of 20 to
30% for 1100 kV compared with 765 kV and a difference of +35% for the scenario with
DC current at ±600 kV. Of course, not only the investment costs should be considered.
Finally, it was decided to apply the system at 765 kV.
2.3.2.1 Actual 765 kV transmission network

As shown in Fig. 2.3.1, the actual HQ 765 kV transmission network comprises:
• 11,000 km of lines;
• Nine synchronous condensers (250 MVA each) for fast voltage control following a
disturbance;
• Eleven 300 Mvar static Var Compensators (SVC);
• 11,000 Mvars of series capacitors installed along 32 line sections providing a degree of
compensation between 16 and 44%, which allows transmission of between 2,200 and
2,700 MW per circuit (increased by at least 50% from the original network);
• 165 and 330 Mvar shunt reactor units to compensate for line charging (55 to 65%
compensation degree). At initial commissioning, the shunt reactors were not switchable.
However, due to the difficulty in controlling the voltage during normal operation with
greatly variable line-loading levels, all shunt reactors were retrofitted with circuit
breakers.
2.3.2.2 765 kV substation layout

To allow for better operating flexibility and to facilitate maintenance work without impacting the
availability of the installation, the 765 kV substations were designed with double busbars, with
two circuit breakers per line.
2.3.3 Overview of field experience and the evolution of Hydro-

Québec’s 765 kV transmission network
Two decades of operational experience revealed the necessity to improve the transmission system
reliability (following major outages by the end of the 1970s and beginning of the 1980s). Major
revision of the system design and operation performance standards essentially focused on
strengthening the system, enhancing voltage control and revising the philosophy on countering
extreme contingencies. This was basically achieved by installing series compensation on 32 line
sections of the 765 kV system in addition to the use of system-wide fast-acting automatic
schemes to switch equipment, reject generation and shed load in order to improve system stability
and avoid the loss of synchronism between generators. The most important system-wide
protection schemes are listed below:
20
• Automatism MAIS (Manœuvres Automatiques d’Inductances Shunt): This scheme is used
to automatically switch-in or -out the shunt reactors following system disturbances.
Special inductive potential transformers with very high-precision requirements are used to
measure the voltage on the 765 kV systems. This scheme was implemented in twenty-two
765 kV substations to control a total of 15,000 Mvars.
• Automatism RPTC (Rejet de Production et Télédélestage de Charge): This scheme is used
to reject a large amount of generation and to shed large blocks of load in order to prevent
a total grid collapse when the system conditions are met. It is implemented in fifteen
765 kV substations.
• Automatism SPSR (Solutions aux Problèmes de Séparation de Réseaux): This scheme is a
last-resort solution used when a risk of total collapse is identified that would involve
network separation and excessive overvoltage levels. When the system separation
conditions are met, the SPSR mainly involves switching of so-called “sacrificial” surge
arresters to allow controlled and safe dismantling of the network.

Hydro Québec projects
Throughout the development of its vast 765 kV network, HQ has faced a number of technical
challenges including those mentioned below:
• The major geomagnetic storm that hit the James Bay transmission system tripped all the
SVCs, resulting in a total blackout. This storm produced circulation of DC current into the
765 kV system and caused saturation of the step-up generating transformers and
consequently, excessive voltage distortion. Series compensation on these lines greatly
reduced the risks associated with a geomagnetic storm given its inherent capacity to block
DC current.
• Due to massive installation of series compensation, line circuit breakers faced more
stringent constraints (increased TRVs, increased switching surges, delayed zero-current,
etc.). Performance assessment of existing line circuit breakers (in terms of TRVs) and
development of special circuit breakers was necessary in order to meet the specific
requirements. [5], [6]
• The exceptional ice storm in the winter of 1998, which left up to 75 mm of radial
accumulated ice on the conductors, resulted in the collapse of 150 towers on the 735 kV
network. A major program of mechanical reinforcement combined with the use of de-
icing equipment for some strategic lines will prevent or minimize further damage in the
case of such an exceptional event. [7]
• Energization of unloaded power transformers on a weak network (during restoration of
the 765 kV system) may lead to significant TOV if no mitigation measures are taken.
With air-blast circuit breaker technology, closing resistors with large pre-insertion time
(12–14 ms) have been used to reduce the inrush current and associated TOV. With actual
SF6 technology (live tank), closing resistors with such pre-insertion time would have
involved major development and significant cost. Therefore, HQ has made major
advances in a special controlled switching system (CSS) taking into account the residual
flux of the transformer. This special CSS has thus far been used to energize step-up
transformers in hydroelectric power plants where frequent energization was foreseen.
21
Reference [8] shows the detailed solution that was applied.
• More recently, a new fast protective device (FPD) was installed to rapidly by-pass the
series capacitors on a 330 kV line for any internal fault, thus eliminating the trapped
charge on the series capacitor before the fault is cleared by line circuit breakers. This new
device has actually been tested in the field since 2003. Application of this solution to the
765 kV lines will be studied after having verified its performance and reliability on the
pilot project at 330 kV. Reference [9] provides full details on this new series-capacitor
protection scheme.
Table 2.3.1 Hydro Québec specifications

Highest Voltage 765kV
Items Transformers CBs (AIS), DS PT and CT
LIWV (kV) 1950 kV standard 2100 (-455 kV) 2100 kV standard
wave across terminals wave
2145 kV for 2100 kV to ground 2310 kV chopped
chopped waves waves
SIWV (kV) 1550 kV 1175 (-650) kV 1550 kV
across terminals
1425 kV to ground
AC voltage 850 kV (2 min) 1150 kV across 830 kV
750 kV (1 hour) terminal
830 kV to ground
Protective levels 1620 kV (Steep front 1 µs, 20 kA)
1425 kV (Lighting surge 20 kA, 8 µs/20 µs)
1140 kV (Switching surge, 2 kA)
Amplitude factor 1.53 for T10 1.5 for T60
1.54 for T30 1.4 for T100
DC time constant 75 ms
2.4 Korea’s project
2.4.1 Features of KEPCO 800 kV transmission system

Korea’s electricity demand has been steadily expanding in line with an average economic growth
rate of approximately 10%. The world's first double-circuit 765 kV transmission system was
constructed in Korea in 2002 to transmit large-capacity energy generated at a remote seaside
facility to urban areas. The maximum transmission capacity of this system is 14,500 MW (double
circuits with 7,250 MW/circuit). The transmission towers, which are 100-150 m high, are
equipped with special suspensions using 6×480 mm2 conductors.
22
Fig.2.4.1 KEPCO 765 kV transmission systems
2.4.2 800kV GIS substation

A full 800 kV GIS with double bus and one and a half CB configuration (800 kV 50 kA, 8000A
2-break GCB) was installed to minimize the installation space as well as to assure safety and
environmental friendliness. KEPCO will expand their 800 kV project to 8 circuits of 765 kV
transmission lines, 12 circuits of 345 kV lines including 5 units of 2000 MVA transformers and
7 units of 200 MVA shunt reactor banks. The maximum power supply capacity will reach
10,000 MW at the final stage. 5 substations (21,110 MVA) have been operating since 2005. High
speed grounding switches (HSGS) can secure multiphase reclosing of the 765 kV lines. [10]
Fig.2.4.2 KEPCO 800 kV GIS substation
23
2.4.3 Insulation level
Power-frequency temporary overvoltage of the sound phase due to phase-to-ground fault or load
rejection overvoltage was analysed using GCB equipped with closing resistors as shown in
Table 2.4.1. Switching overvoltages are shown in Table 2.4.2.
Table 2.4.1 Temporary overvoltage (1p.u.=800/ 3 kV)
Classification Overvoltage P.U
Phase conductor (Bus, line) 1.2 (555 kV)
Temporary overvoltage
Neutral point (Floating) 0.3 (139 kV)
Table 2.4.2 Switching overvoltage (1p.u.=800 2 / 3 kV)

Location Phenomena Phase to ground Phase to phase Remark
Ground fault, Fault clearing, Closing resistor
Tower / line 1.9 (p.u.) 3.5 (p.u.)
Energizing/Re-energizing, 1000 ohm
Ground fault 1.6 2.3
Fault clearing 1.7 2.9 Without closing
Substation
Energizing/Re-energizing 1.8 3.5 resistor
Summary 1.8 3.5
Neutral point
Single phase ground fault 0.3 –
of transformer
Lightning overvoltage was also analysed when flashovers occurred in the substation as well as at
the first tower for outgoing transmission lines, as shown in Table 2.4.3.
Table 2.4.3 Lightning overvoltage
Equipment Transformer Switchgear/Bus HSGS
Lightning overvoltage 1,710 kV peak 2,015 kV peak 2,025 kV peak
Air clearance between phase-to-ground and phase-to-phase of the 765 kV bus shall be in
accordance with Table 2.4.4.
Table 2.4.4 Air clearance of outdoor bus
Outdoor
Highest voltage (kV) Air clearance (mm)
Phase to phase Phase to ground
11,000 7,000
800 Minimum 8,500 5,000
Fig.2.4.3 KEPCO insulation distance of substation outgoing part
Table 2.4.3 summarizes the KEPCO insulation level of 800 kV substation equipment.
24
Table 2.4.3 KEPCO specifications
Highest Voltage 800kV
Items Transformers CBs (GIS) PT and CT
LIWV (kV) 2050 kV standard 2250 kV across 2250 kV standard
wave terminal wave
2255 kV for 2250 kV to ground 2250 kV chopped
chopped waves waves
SIWV (kV) 1500 kV 1100 kV across 1425 kV
terminals
1425 kV to ground
AC voltage 800 kV (2 min) 1100 kV across 975 kV
690 kV (1 hour) terminals
830 kV to ground
Protective levels 1500 kV (steep front 1/9 µs, 20 kA)
1400 kV (lighting surge 20 kA, 8 µs/20 µs)
1300 kV (switching surge 2 kA, 30/60 µs)
1.5 for T30 1.4 for T100
2.4.4 Technical specifications of 765 kV power transformer

The main power transformer is a single-phase auto-transformer. Each single phase has two
divided tanks with primary, secondary and tertiary windings for transportation.
Table 2.4.5 Technical specifications for power transformers
Transformer
Rated voltage 765/√3 kV (Primary)
345/√3 kV (Secondary) 23 kV (Tertiary)
Rated capacity 2000MVA/3
Lighting Impulse Voltage 2050 kV
Switching Impulse Voltage 1500 kV
2.4.5 800 kV Gas insulated switchgear

The 800 kV outdoor-type full GIS is composed of CB, DS, ES, HSGS, CT, PT, MOSA and
Bushing, etc. The main bus is split into a three-phase separation tank with double bus. The
800 kV 50 kA dead-tank GCB with hydraulic operating mechanism has two interrupters in series
per phase. The 800 kV disconnecting switch adopts a resistor-insertion scheme to suppress very
fast transient overvoltage (VFTO). The dead-tank gapless-type MOSA is applied in accordance
with Line Discharge Class 4.
Table 2.4.6 Technical specifications for other equipment
GIS
Rated voltage 800 kV
Rated nominal current 8000 A (Line and Bus)
Lighting Impulse Voltage 2250 kV
Switching Impulse Voltage 1425 kV
Rated short withstand current 50 kA, 2s
25
2.5 Brazil’s project
2.5.1 Features of Furnas’ 800 kV transmission system

This section basically comprises two parts. The first part describes Brazil’s experience with long-
distance transmission systems (800 kV AC and ±600 kV DC), and also covers the problems and
challenges encountered along with some aspects of related technical solutions from the
perspective of planning and operating experience. The second part summarizes the trends and
future technical challenges regarding the potential use and need of UHV transmission system
technologies, such as for solutions associated with major hydro-plants foreseen for the Amazon
region, requiring the transfer of huge amounts of power (6,000 to 10,000 MW) across distances
as long as 3,000 km. [11]
Consequential impact on the existing network is also considered by pointing out the need to
reinforce the receiving network, as well as to investigate procedures for limiting short-circuit
current levels that exceed existing equipment withstand ratings. These new challenges may also
require a thorough review of IEC standards due to special UHV equipment requirements.
Fig.2.5.1 FURNAS main assets

Furnas Bulk Generation and Transmission Company is a Brazilian federal government utility.
Furnas is responsible for supplying electric energy to distribution companies, mostly in
southeastern and central western Brazil, who in turn supply electricity to 8 states and the Federal
District. This region has a population of 90 million people and generates 65% of Brazil’s GDP
26
and 55% of the nation’s total electric energy consumption. Furnas plays another important role
by providing the main interconnections between the southeastern region (the most developed
region of the country) and other regional grids, such as the southern regional transmission grid
and the northern and northeastern regional grids.
GENERATION: Furnas’ present bulk generation assets total about 10 GW of installed power
capacity, of which about 9,000 MW is available from 11 hydro power plants and 1,000 MW
from 2 thermal power plants. Furnas is presently constructing 6 new hydro power plants at the
same time, adding about 1700 MW of installed power capacity. For comparison purposes, the
growth of Brazil’s total power system load, on average, demands an increase of about 4,000 MW
annually in order to cope with the load increase.
TRANSMISSION: The bulk transmission assets of the company total 20,000 km of transmission
lines, from the voltage level of 138 kV up to 800 kV, interconnecting 44 substations, where the
company has a total rated power transformer capacity of 95 GVA. The transmission lines, and
related power equipment operate at rated voltage levels of 138, 230, 345, 500 and 800 kV AC,
and ±600 kV DC.
2.5.2 Overview of field experience of 800 kV long distance

transmission lines
Bulk electric power supply is typically provided by long EHV transmission lines belonging to
the national transmission grid, not only for the purpose of connecting huge-capacity
hydroelectric power plants to the main load centres, but also to interconnect power systems of
different regions, and from distinct geographical areas. The vast expanse of Brazil means that
several transmission lines and substations of 550 kV up to 800 kV had to be built shortly after
these voltage levels were introduced in other parts of the world, due to the long distances
between the load centres and major generation power plants.
The Itaipu Hydroelectric Power Plant was built on the Paraná River, at the border between Brazil
and Paraguay and belongs to the company ‘Itaipu Binacional’. The power plant has 20
generating units of 700 MW each, resulting in 14,000 MW of installed power capacity, which
enables the production of around 100 TWh of energy per year. As the nominal power frequency
differs between the two countries, the frequency for Brazil’s generating units is 60 Hz, while that
for Paraguay’s is 50 Hz. However, Brazil has contracted to buy all of the 50 Hz energy, except
for that consumed by Paraguay, which means that 75% of the 50 Hz energy, on average, was
consumed by Brazilians last year.
Furnas Centrais Elétricas was responsible for developing the transmission solution for integrating
Itaipu Power Plant’s generation into Brazil’s national transmission network. The company
conducted the planning studies, equipment specification, factory and laboratory tests, on-site
construction, commissioning tests. Thus, Furnas started operation of the 800 kV AC and ±600 kV
DC systems in 1982 and 1984, respectively, integrating the Itaipu Power Plant generation into the
national network.
The accumulation of knowledge and experience related to UHV transmission issues started with
the initial investigation and planning studies on a system associated with the Itaipu Power Plant,
which consisted of an 800 kV AC five-line transmission corridor.
After the Paraguay government’s decision to retain its network nominal frequency at 50 Hz
instead of accepting Brazil’s offer to convert the entire network to 60 Hz, it became necessary to
27
review those first investigation studies in order to define a hybrid transmission system of 3 × 800
kV AC transmission lines and 2 bipole ±600 kV DC transmission system.
The planned hybrid UHV AC/DC transmission system, shown in Fig. 2.5.2, is one of the most
important systems in the western world due to its nominal voltage levels, rated power capacity
and significance in Brazil’s electric industry. It not only uses some of the highest voltage levels in
commercial operation worldwide, but it also has some of the highest capacity of transmitted
power (rated at 12,600 MW) over long distances (about 1,000 km).
2.5.3 Evolution of Furnas’s 800 kV AC Transmission Systems

The HVAC transmission system interconnects the Foz do Iguaçu Substation (near the Itaipu
Power Plant) to the Tijuco Preto Substation in the São Paulo area, one of the main load centres
of Brazil’s electrical network. The Foz do Iguaçu Substation contains both the UHV 800 kV AC
switchyard and the ±600 kV DC converter substation (rectifier side). Together, they form one of
the largest substations in operation, in terms of power and size.
Fig.2.5.2 Foz do Iguaçu substation contains both the HVAC switchyard for three 800kV circuits
and the HVDC converter substation equipment/switchyard (HVDC link rectifier side)
The three 800 KV AC circuits, each about 895 km long, as shown in Fig. 2.5.3, are able to
transmit to the Brazilian interconnected network the rated generated power from the 10
generating units at the Itaipu Power Plant, operating at a nominal frequency of 60 Hz. The
planning criteria of such a transmission system considered the possibility of dispatching this
amount of power even in an “n-1” element outage configuration. This 800 kV AC system is also
responsible for energy interchange between the southern and southeastern geographic regions of
Brazil’s interconnected national grid (refer to Fig. 2.5.4).
The electrical requirements for the 800 kV AC equipment were defined in the late 1970s and
early 1980s. At the time, Brazil’s network consisted of several weakly connected subnetworks.
Although each subnetwork could be considered as solidly grounded, the network as a whole
could not. Furthermore, the 800 kV network would not be a meshed network but instead a radial
one consisting of 3 parallel circuits. Thus, in consideration of the new and unfamiliar voltage
levels, the planning engineers at that time decided to adopt conservative requirements for the
equipment, derived from system simulations under severe operation and emergency conditions.
28
Figure 2.5.3 General overview of the Furnas transmission system for
integrating the Itaipu Power Plant into Brazil’s national electrical network
Fig.2.5.4 Detailed view of the Furnas transmission system for

integrating the Itaipu Power Plant into Brazil’s national electrical network
In the mid-1980s, the full power of the Itaipu Power Plant corresponded to more than 35% of the
total installed power in Brazil’s power network (today, it corresponds to less than 14%, even
considering the expansion of the Itaipu Power Plant from 12,600 to 14,000 MW). In the initial
stages of the Itaipu Power Plant operation, during the light load period in the late 1980s, its
generated power corresponded to more than 50% of the dispatched power in Brazil’s
interconnected network.
Therefore, in this context, the Itaipu transmission system was of major importance to Brazil’s
electric bulk network. Its design had to take into consideration, on one hand, Brazil’s first-time
29
use of such high voltage levels, equipment rated power and all related technological aspects and,
on the other hand, the absolute necessity to assure that the equipment withstand levels would
cope with the system requirements, as reliably as possible.
The shunt reactors (fixed at line-ends and switchable at busbars and transformer tertiary), series
capacitors and synchronous condensers provide, respectively, transient switching overvoltages
mitigation (fixed SR at line-ends), voltage regulation (switchable SR at busbars and transformer
tertiary) and system dynamic stability (series capacitors and synchronous condensers).
2.5.3.1 Application philosophy for 800 kV Shunt reactor (SR)

Furnas use SRs in system transmission OH lines for two main purposes: (a) switchable SRs at
busbars or tertiary windings of power transformers for steady-state voltage control issues; and (b)
fixed SRs at OH line ends to deal with switching transient overvoltage requirements related to
line closing and/or opening operation (including fast auto-reclosing). Therefore, the same
philosophy is followed for the SRs installed in this 800 kV transmission system, giving 70%
shunt compensation for line charging. Figure 2.5.5 shows the existing SR application philosophy
installed in such transmission system.
For transient switching overvoltage mitigation, fixed SRs have the aid of arresters installed at the
line ends. Over the last 30 years, two very distinct functions relied on such duty, depending on
the evolution of technology and materials applied by arrester manufacturers: when the only
arresters used were those with active gaps and nonlinear resistance made of silicon carbonate
(SiC), the first arrester that sparked its active gaps was the only one responsible for draining all
the surge energy as well as controlling the switching overvoltage in conjunction with the fixed
SR at the line end.
Fig.2.5.5 Fixed shunt reactor application for transient switching overvoltage mitigation on OH line
Regarding gapless MOSAs, those installed to protect SRs against lightning impulses could also
be used in conjugation with MOAs installed at the OH line ends. Thus, by using this set of surge
arresters (not only the ones installed at the OH line ends, but also those protecting the fixed SR
against lightning surges), control of switching transient overvoltage as well as draining of
switching surge energy, could be achieved more efficiently and reliably, thus helping with
mitigation/control of TOV as mentioned.
For switchable SRs, used to mitigate/control switching transient overvoltage generated when
switching off the SR, it is now possible to use CBs equipped with opening resistors or controlled
switching. In the case of using CBs equipped with opening resistors, for optimal transient
switching overvoltage control, the ohmic value in each phase should be of the same magnitude as
30
the characteristic impedance of the SR (surge impedance). At the 800 kV level, the normal range
of surge impedance values for SRs is 2.000 to 4.000 ohms.
In the 1980s, the 800 kV switchable SRs in the Furnas transmission system had CBs equipped
with opening resistors for transient switching overvoltage mitigation, as defined at the planning
and design stages, developed in the late 1970s and early 1980s. The design criterion for
dimensioning the CB opening resistors was to use the same ohmic value range for SR surge
impedance, meaning it was necessary to provide CBs with opening resistors in the range of 3.000
ohms. In this case, there was no control over the arcing time of the SR current in the CB
chambers during switching off procedures. Consequently, this could lead to reignition within the
CB chambers when the arcing time was too small, causing a new overvoltage with very high
front steepness.
During the first decade of switchable SR operation, this phenomenon caused some major damage
to SRs and related CBs, even leading to the explosion of a switchable SR CB at the Itaberá
Substation. After this accident, the Furnas decided to replace the CBs equipped with opening
resistors with controlled switching. The controlled switching technology demonstrated very good
performance in mitigating transient switching overvoltage, completely avoiding possible
reignition during the switching off procedures. Extensive field tests/measurements confirmed the
expected performance during several switching off operations of the existing 800 kV busbar SR.
On the other hand, the longer arcing times, provided by the CB equipped with a controlled
switching device, would lead to more prominent ‘current chopping’ overvoltage that should be
investigated, in principle, as a significant concern, in order to avoid undesirable damage to SRs,
or even to CBs.
This transient phenomenon is related to the fact that, during SR switching off, the nominal
current is suddenly forced to zero prior to the natural power-frequency-zero-crossing within the
CB chambers, leading to the phenomenon known as ‘current chopping’. Thus, current chopping,
meaning a sudden change to zero at the magnitude of the SR nominal current, forces the
‘magnetic field’ energy stored in the reactor coil to be converted to ‘electric field’ energy stored
within the stray capacitance, thereby causing a sudden change in the load-side transient voltage.
This energy transfer phenomenon is known as ‘chopping overvoltage’. The lower the stray
capacitance is, the higher the chopping overvoltage will be during the occurrence of the current
chopping.
However, the field experience of Furnas showed that it was much more reliable and secure to
have complete control over avoiding reignition during SR switching off, by means of using CBs
equipped with controlled switching, compared to the increase in current chopping overvoltage. In
the case of 800 kV switchable SRs, the increase in chopping overvoltage, derived from the
increase in minimum arcing time (given by CBs with controlled switching), is very small, due to
the fact that the stray capacitance of the SR windings is quite significant.
2.5.3.2 Application philosophy for 800 kV Series Capacitor (SC)

800 kV series capacitors (SCs) provide system dynamic stability when dispatching huge amounts
of power through the 800 kV transmission system from the Itaipu Power Plant and/or high
interchange between the two interconnected regional grids. The existing SCs were designed to
offer compensation of 40 to 50% related to the OH line longitudinal inductive reactance,
depending on the location in the 800 kV transmission system. Under these dispatch conditions,
31
the voltage profile along the transmission system must be as close as possible to the maximum
continuous operating voltage level (at nominal frequency), or 800 kV rms between phases.
The SCs were intended to operate with two series segments, each segment corresponding to 20 or
25% of the longitudinal reactance of the transmission lines. SC segments can operate together or
separate from each other, thus increasing the reliability of the SCs themselves and of the 800 kV
transmission systems as a whole. The maximum operative voltage level (800 kV rms between
phases) of this transmission system is measured right after the SC (at the line side) in the Ivaiporã
and Itaberá substations, since a sudden increase in voltage profile occurs in the presence of the
capacitive reactance in series with the line inductive longitudinal reactance. (refer to Fig. 2.5.6)
Equivalent Source 1 Line flow Equivalent Source 2
Zeq1 Zeq2
800 kV rms
Line length
Fig.2.5.6 Voltage profile caused by Series-Capacitor placed in an OH line
During the planning stages, SCs having capacitive reactance to compensate for part of the line
inductance, should be designed in such a way that some main technical and economic
requirements regarding the SC equipment itself could be achieved: (a) SC self-overvoltage-
protection devices (in the presence of short circuits); (b) protection coordination of parallel lines
with series-capacitor compensation; and (c) sub-synchronous resonance.
In the presence of SCs connected to the transmission lines, as they compensate the line inductive
reactance, short-circuit currents tend to increase (due to the lower values of short-circuit
equivalent impedance of the system), and transient overvoltage caused by short circuit flowing
through the SCs also tends to be greater. In order to control such short-circuit currents as well as
transient overvoltage during the flow of short-circuit currents in the compensated lines, SCs have
a ‘self-overvoltage-protection device’, as shown in schematic detail in Fig. 2.5.7.
Regarding the transmission line voltage profile, the higher the short-circuit current flowing
through the SC, the greater the rise in transient voltage at the SC terminal. The design of such
self-overvoltage-protection device level is driven mostly by economic reasons regarding
insulation withstand requirement costs. Thus, the nonlinear varistor and gap-protection-circuit
operate whenever necessary, bypassing the SC, in order to keep the transient overvoltage values
within the designed withstand levels of the SC components.
32
SEGMENT 1 SEGMENT 2
Fig.2.5.7 (a) (b) Detailed view of the 800 kV AC series-capacitor at Itaberá Substation
The main role of the varistor is to operate as an electronic switch, bypassing the SC while short-
circuit current flows through it. The purpose of the parallel gap and the SC-segment-bypass
breaker is to protect the varistor when draining high amounts of energy during severe transient
overvoltage in the presence of larger short-circuit currents.
In addition to this self-overvoltage-protection scheme, the main conventional protection features
of the SC are as follows:
• Gap operation
• Platform failure
• Current imbalance (alarm or activation of SC bypass)
• Fibber optics failure
• Bypass breaker failure
• SC bank overload
33
• Air-compressed central failure (the gap operates with air-compressed blast, provided by a
dedicated central SC)
Protection coordination of parallel lines with series-capacitor compensation, also an issue of
significant concern, has to do with the distribution of short-circuit current among parallel circuits,
which may cause undesirable protection operation of such parallel (and not faulted) lines. This
may happen due to overvoltage protection or SC bypass malfunction.
‘Sub-synchronous resonance’ may occur when thermal turbines are in close electrical proximity
to series-compensated line. Depending on the compensation degree provided by the SCs, they
might activate a natural oscillation mode having typical values of resonance frequency below the
rated power frequency in conjunction with line inductance. Normally, thermal turbines also have
mechanical natural oscillation modes in that frequency range, thus possibly leading to sub-
synchronous resonance.
2.5.3.3 Other notable aspects of 800 kV transmission system

Figure 2.5.8 shows the 800 kV line-phase transposition design criterion.
ΦC ΦB ΦA
ΦA
ΦA ΦC ΦB
ΦB
ΦB ΦA ΦC
ΦC
1/6 1/3 5/6 6/6
LENGTH
Fig.2.5.8 800 kV line phase transposition criterion

Regarding fast auto-reclosing of the Itaipu 800 kV transmission system, the main related data is
that only 3-phase fast auto-reclosing is in operation. Dead time is 600 ms for all line segments
(all 3 circuits: Foz do Iguaçu – Ivaiporã; Ivaiporã – Itaberá and Itaberá – Tijuco Preto). The
leader terminal for auto-reclosing is Ivaiporã for all lines between Foz do Iguaçu / Ivaiporã
substations, and Itaberá for all lines between Ivaiporã / Itaberá and Itaberá / Tijuco Preto
substations. The determination of leader terminal is related to the minimum power-accelerating
impact in hydro generators at the Itaipu Power Plant and other hydroelectric plants located in the
southern regional grid (refer to Fig. 2.5.9).
For equipment requirements, the 800 kV transmission system should take into consideration the
worst possible conditions regarding switching transients. Thus, it was decided at the beginning of
the design stage to take into account, for instance, the criteria for switching transients derived
from total 'opposition-of-phase' - OP (i.e., 180 instead of 120°, foreseen in international standards
at that time). This was considered reasonable, since the 800 kV AC system would be, in practical
terms, not only connecting the Itaipu Power Plant to Brazil’s interconnected grid, but also acting
as the main interconnection tie between Brazil’s southern and southeastern regional transmission
grids, the two major subsystems of the country at that time. Before the Itaipu transmission system
existed, the two regional transmission grids were weakly interconnected with only a few tie-lines
at voltage levels of 88, 138 and 230 kV at the border between the state of São Paulo (southern
34
border of the southeastern regional grid) and the state of Paraná (northern border of the southern
regional grid).
Auto-reclosing leader terminal
Auto-reclosing leader terminal
Fig.2.5.9 Auto-reclosing terminal

Considering the same reason for the choice of OP, a value of 2.0 was adopted for the out-of-
phase factor for all TRV parameters. The reason for the use of an amplitude factor of 1.64 for Uc
of TRV, instead of 1.25 as stipulated by IEC, is the adopted 'grounding criteria.' Although each
regional grid could be considered as solidly grounded, the 800 kV system as a whole could not,
as it was radial and not a meshed network like the two regional grids interconnected by means of
this AC transmission link.
Following this idea of adopting conservative criteria for defining the equipment withstand
requirements, derived from system simulations under severe operation and emergency conditions,
it was also found necessary to define other requirements based on strict premises, such as the
maximum over-speed during total load rejection in the 800 kV AC system as 1.1 of nominal
frequency (66 Hz), instead of adopting the standardized value of 1.05 fn (63 Hz). This strict over-
speed criteria was used to verify the possibility of self-excitation regarding the Itaipu generators,
and also to define the withstand requirements of switching equipment due to overvoltage
transients caused by full load rejection.
Finally, as other example, the value of 800 kV was defined as the 'maximum operating voltage'
for the Itaipu transmission system, which means only 1.045 p.u. of the nominal operating voltage
(765 kV). In all other voltage levels in Brazil’s power system, the maximum operating voltage
(non-continuous operating voltage limit, for emergency operating conditions) is 1.10 p.u. of the
rated or nominal voltage value (138/152 kV; 230/253 kV; 345/380 kV; 440/484 kV; 500/550
kV).
2.5.4 Notable aspects of the ± 600 kV DC Transmission System

The HVDC transmission system consists of two ±600 kV AC bipoles, each about 900 km long,
one AC/DC rectifier side at the Foz do Iguaçu Substation and one DC/AC inverter side at the
Ibiúna Substation, located very close to São Paulo city, as shown in Fig. 2.5.10.
The system is rated to transmit 6.300 MW from the power produced by the 10 other 50 Hz
generating units owned by the Paraguayan side of the bi-national Itaipu Power Plant. In order to
35
increase Brazil’s know-how on HVDC transmission, being used for the first time in the country,
the following constraints were established: Services, equipment and products should be
nationalized; Brazilian enterprises/engineers should take part in the planning, designing, building
and operation of the HVDC transmission system; and technology transfer through “on-the-job
training” should be practised.
Fig.2.5.10 ± 600 KV DC transmission corridor subsystem

The main characteristics of the ±600 kV DC transmission system are as follows:
• Two substations: Foz do Iguaçu (rectifier side) and Ibiúna (inverter side);
• Nominal power: 6,300 MW;
• Nominal voltage: ±600 kV (DC);
• Nominal current: 2,625 A (DC);
• Two bipoles (±600 kV);
• Each bipole composed of two poles;
• Each pole composed of two 300 kV series-connected converters.
FURNAS CENTRAIS ELÉTRICAS S.A.
Sistema de Transmissão HVDC

+ 600kV
Polo 4
Bipolo 2
Polo 3
Itaipu 50 Hz
- 600kV
Ibiúna
+ 600kV
Polo 2
Bipolo 1
Polo 1
- 600kV
Fig. 2.5.11 FURNAS +/-600 kV HV DC main features (AC in-feed lines, rectier AC busbar,
converter transformers, AC filters, 12-pulse valves, DC lines)

Furnas’ long transmission systems
It is important to point out that, at that time, there were no existing IEC standards for such a
huge HVDC transmission system. In this sense, from the perspective of planning and operating
experience, some of the aspects and problems related to the hybrid HVAC and HVDC features
36
and equipment required thorough investigation right from the start. For new long-distance
transmission systems, some important features and concerns related to equipment withstand
requirements should also be thoroughly investigated, such as:
A. Comparison between ‘equivalent’ solutions for HVAC and HVDC systems and between
‘equivalent’ GIS and AIS equipment, in terms of engineering and economics issues (considering
environmental and regulatory issues);
B. Characteristics of HVDC system facilities, such as the so-called ‘dynamic performance
feature’ (concerning the impact on AC systems connected to the converter substations, both
‘rectifier’ and ‘inverter’ HVDC system sides, in terms of mitigation/elimination of dynamic
electromechanical transients/oscillations); HVDC transmission system ‘forced isolation
protection scheme’ (implemented to protect against electrical transients derived from partial or
total load shedding); ‘high MVar consumption’ operating mode; ‘automatic fast switching’ from
one faulted DC line to another under unrecoverable short-circuit conditions in the first line;
C. Specification of HVDC reliability features (ability of the system to transmit the rated power
under contingency conditions and outages) including operation mode (bipolar/monopolar) and
related engineering and environmental issues; temporary overloading/overcurrent capacity; AC
system faults/configurations that may create commutation failures mainly at the inverter station;
D. Importance/advantages of analogue/digital simulation facilities for HVDC/HVAC
transmission systems (necessity/usefulness of simulation tools for HVDC transmission systems
in terms of planning/specification/operational optimization and economics);
E. DC switchyard bushing isolation problems and implemented mitigation solutions;
F. Issues related to filtering requirements (short-circuit level criteria, filter overloading, use of
“active filters” since they are not a sink for harmonics other than their own, which can be
decisive in filtering performance, etc);
G. HVAC shunt and series compensation devices (and the related inherent aspects in terms of
voltage profile control, overvoltage transients and protection issues);
H. Importance of defining equipment withstand requirements in the planning phase, taking into
account the possible/foreseen evolution/expansion of the AC grid/network (upgrading in short-
circuit level requirements);
I. Enterprise staging definition, i.e., intermediate stages of transmitting power capacity
considering cost constraints, necessity of scaling the power transmission, time interval between
stages, etc. (series connection of converter groups per pole easily allows the staging of 25, 50, 75
and 100% of the total rated power);
J. Oil chemistry issues related to HVDC converter transformers;
K. Possibility of controlled switching usage.
2.5.6 Future challenges regarding UHV transmission in Brazil

Stepping into the future and trying to foresee the possible needs, trends and challenges related to
the use of UHV transmission in Brazil, [12] we have, on one hand, 180 GW corresponding to the
unused portion of Brazil’s hydroelectric potential. According to official estimates, a high
37
percentage (around 70%) is located in the Amazon region. On the other hand, the consumption of
energy is spread throughout the country according to the following geographical distribution:
Isolated systems in Amazon region (2% of consumption)
Interconnected system in north/northeast (19% of consumption)
Interconnected system in south/southeast/central-west (79% of consumption)
In terms of growth, it is reasonable to say that Brazil will have to transport huge amounts of
energy produced mainly in the Amazon region to the main load centres of the country, covering
distances of more than 2,500 km, as shown in Fig. 2.5.12. Under such circumstances, the use of
UHV transmission systems (DC links and/or AC overhead lines) appears to be a competitive and
suitable alternative, both economically and environmentally. In this case, it would not be
necessary to change the frequency from the rectifier AC side to the inverter AC side, since it is
simply a matter of a set of issues: stability, loss reduction, compactness, environmental
friendliness and transmission cost over the long distance from the power plants in the Amazon to
the main load centres.
Figure 2.5.12 Geographical distribution of load centres in Brazil (left) and average distance
between the main load centres and future Madeira river complex (right)
The Amazon region has abundant water resources in quite a plain geographic area without
waterfalls, and the rivers are huge with sustained water flow. Thus, generation of electricity can
be obtained by water flow rather than by potential energy. In order to minimize the
environmental and social impact, preliminary prospective studies of the region suggest the
construction of hydroelectric power plant dams with very low height.
The next project in the Amazon region, under consideration by the Brazilian federal government,
is the ‘Madeira River Project’. It consists of a business complex including two hydroelectric
power plants on the Madeira River: ‘Santo Antonio’ near Porto Velho city, the capital of the state
of Rondônia, and ‘Jirau’ near the border between Brazil and Bolivia, both businesses bringing a
total installed power of 6,450 MW. The complexity inherent to the project requires considerable
effort on the part of the government to obtain the environmental licenses needed to start the
installation process. According to Brazil’s regulatory rules for the electric sector, the business
owners, who will be responsible for the construction and operation of the hydroelectric power
plants, as well as for the transmission system regarding the plants’ integration into the national
electrical network, will be defined by public bidding. This procedure will be announced in the near
future by the Brazilian Regulatory Agency for the electric sector - ANEEL. [13] Furthermore, the
long distances to be covered and access difficulties to resolve bring challenges to new
38
transmission corridors in the Amazon region, concerning equipment size, logistics and
transportation.
The following topics must be analysed in detail in order to establish the reference UHV
transmission alternatives (DC, AC or hybrid transmission systems):
• Line configuration /insulation/ clearances;
• Corona and field effects line performance;
• Level of power to be transmitted;
• Weight and size of equipment for shipping and transport (mainly the transformers);
• Power loss;
• Spare parts;
• Series connection of converter groups per pole (DC alternative);
• Overload /stability requirements;
• Need for reinforcing the receiving network;
• Need for investigating procedures for limitation of short-circuit current levels exceeding
existing equipment ratings.
Despite Brazil’s experience with hybrid parallel AC 800 kV and DC ±600 kV systems, the country
faces huge challenges related to the foreseen Amazon transmission system requirements, and
depending on the outcome of the UHV transmission alternative references, the existing IEC
standards may not cover all of the equipment special withstand requirements. This would lead to
the necessity of developing new specification standards possibly concerning planning and design.
2.5.7 Summary
Brazil’s electric power is mainly supplied through long EHV transmission lines. Consequently,
the building knowledge and experience on long-distance transmission systems were
accumulated. The initial investigation studies for integration of the Itaipu Power Plant into the
national interconnected grid took part in this role as well. The Itaipu transmission system,
planned, built and operated by Furnas, is one of the most important in the western world due to
its nominal voltage levels, rated power capacity and significance in Brazil’s electric industry. It is
important to point out that, at the time, there were no existing IEC standards covering all
necessary technical-related aspects for such a huge long-distance transmission system,
particularly for the HVDC link.
Around 70% of Brazil’s unused hydroelectric potential is located in the Amazon region, while
the consumption of energy is spread throughout the country, mostly near the shores. Thus, it will
be necessary in the near future to transport huge amounts of energy covering distances of more
than 2,500 km. Due to the huge challenges related to the Amazon transmission system
requirements, the existing IEC standards may require a review of equipment withstand
requirements, leading to the necessity of developing new standards/specifications related to UHV
concerns.
To meet these challenges, several Brazilian scientific organizations, universities and utilities are
preparing themselves by means of studying new technologies and developing know-how.
39
2.6 Italy’s project
2.6.1 Features of Italy’s 1000kV transmission system

The fundamental features of the 1000 kV system were identified assuming that this future
network would have to connect three or four new, powerful generating centres of the order of
4000 MW each, to the main load areas located at a distance of about 200-250 km.
Fig. 2.6.1 Assumed energy demand trend in Italy Fig. 2.6.2 Assumed energy demand network
The basic rationale for this scenario stems from Italy’s robust growth in demand for electric
energy experienced at the end of the 1960s. At that time, electricity consumption showed a trend
of doubling every ten years, as seen in Fig. 2.6.1. This need was previously met by introducing
larger generating plants and periodic (about every 20 years) increases in network voltage levels
(145-245-420 kV). The new 1000 kV level is considered appropriate to keep the amount of land
occupied by overhead lines within acceptable limits and for dealing economically with the
increasing need for power transmission.
Investigations were conducted on connections from the Po River estuary to the Milan load area;
from the coast between Tuscany and Lazio to the Roma load area; and from Puglia to Naples. In
all cases, the use of two lines at 1000 kV, having SIL (surge impedance loading) of about
4000 MW (doubled in emergency conditions), was confirmed as being an economic and reliable
solution (Fig. 2.6.2).
These 1000 kV lines were considered mainly for the connection of large power plants; thus,
sufficiently high short-circuit power would be ensured and power flow essentially directed one
way from the generation facility to the load centre. This configuration is favorable for voltage
profile control by generators/switchable Var compensation devices and line energization without
the need for enhanced Var compensation equipment such as FACTS devices.
Furthermore, since Italy’s existing EHV transmission grid was highly meshed, the connection of
a large-capacity link at 1000 kV did not require large network reinforcements in the underlying
voltage levels (400 and 220 kV) that would warrant security conditions, such as those stipulated
40
in the UCTE (Union for the Coordination of Transmission of Electricity in continental Europe,
see www.ucte.org) rules.
The feedback from three years of operating the 1000 kV pilot plant (Fig. 2.6.3) was considered
very positive, even though application of this technology across Italy was put on hold due to
changing conditions in the electricity sector, i.e. much lower demand growth compared to what
was foreseen in the 1970s; new generation technology such as CCGT (combined cycle gas
turbine) was based on mid-size plants scattered throughout the country instead of on large power
plants. However, the practical experience gained from operating a 1000 kV UHV line, substation
and associated cable connection provides an important reference for future wider applications of
AC UHV corridors. An extensive presentation of the outcome of Italy’s 1000 kV project was
recently presented in [14].
1. GCB
2. Disconnector
3. Earthing
discunnector
4. Earthing switch
5. Surge Arrester
1050kV Cable
1050kV GIS
1050kV OH Line
420kV Bus
Fig, 2.6.3 1050 kV Pilot Plant configuration
2.6.2 Overviews of field experiences of the projects

Extensive commissioning tests on the GIS substation were performed in 1994 and in early 1995.
During tests to prove the adequacy of the mounting, and during commissioning of the
installation, the following problems were encountered and subsequently solved:
• Solid particles inside the enclosure were revealed by partial discharge measurement using
the acoustic method; the subject compartments were cleaned and from then on they
maintained satisfactory performance;
• Oil leakage due to the lack of sealing from a case door bolt washer that had thermally
aged due to the eddy current circulating in the case frame;
• One discharge occurred on an inner insulator and on an operating rod of the circuit
breaker units. Subsequent analysis revealed the need to reduce the voltage stress on the
rods and modify the quality assurance procedures adopted by the manufacturer of the
41
circuit breaker for acceptance of the rods.
During service, two minor failures were recorded in the cooling system for the cable and the
autotransformer, respectively, and in the LV/MV supply system.
An additional cause of interruption was a scheduled out-of-service due to maintenance on high-
voltage equipment and auxiliary systems, or following a request from the Regional Dispatching
Centre, when the EHV at the Suvereto substation exceeded 420 kV to avoid a situation in which
the additional Var contribution of the UHV would cause excessive steady-state voltage values at
the 400 kV level.
The above incidents revealed, in hindsight, the following:
• GIS: More attention should be paid to compartment cleaning during on-site assembly;
• Autotransformers: Major attention should be paid to tightening and transportation aspects
and to ensuring good local electrical conductivity in order to avoid high eddy-current
density.

projects
Due to the unusual electrical, mechanical and environmental constraints related to this UHV
level, the following aspects have undergone exhaustive study in the development of Italy’s
1000 kV project, as they required an original non-standard attitude in providing the most suitable
solution from a technical, economical and environmental point of view:
• Control of overvoltage;
• Performance of air and surface insulation;
• Electric fields on conductors and accessories and related effects (corona loss, audible
noise, radio interference);
• Effects of the electric field at ground and that of wind and ice on bundled conductors
(vibration, sub-span galloping);
• Mechanical and electrical behaviour of conductors and accessories in the event of short
circuit
• Necessity to limit the extension of rights-of-way and the visual impact on the
environment, within the framework of sustainable development;
• Additional losses in the conducting materials of transformers, cables, etc., due to the
presence of alternating electromagnetic fields of relatively high intensity;
• Electrodynamic problems associated with short circuits;
• Withstand capability against earthquakes;
• Testability of large components;
• Transportability (especially of transformers).
Identification of the solutions to the following specific problems was a challenge:
• Adoption of single-phase reclosure; this operation requires particular consideration with
42
respect to actual arc phenomena in order to ensure the extinction of the secondary arc. In
this respect, fast grounding switches were considered with the relevant
capacitive/inductive switching duties;
• Presence of high and repetitive very fast transient overvoltages (VFTO) and the
consequences on equipment insulation, mainly transformers;
• Trapping charges in the GIS and the corresponding reduction of the lightning impulse (LI)
withstand.
Testing of UHV equipment may present problems basically related to the huge overall
dimensions of air-insulated technologies compared with those of the test laboratories.
Within this context, a number of challenges must be met, including the following:
• Dielectric testing of UHV apparatus when the volume of the unit under test exceeds the
availability of the indoor test room, which should be higher than 30×30×30 m;
• Performance of certain tests as withstand tests under wet conditions and partial discharge
or RIV tests;
• Verification of performance under contaminated conditions and other electrical-
environmental stresses;
• Configuration of the short-circuit withstand test for switching apparatus;
• Configuration of the unit test for circuit breakers equipped with switching resistors, which
cannot be tested as a whole.
In principle, all tests could be performed in accordance with the present standards by investing in
adequate new laboratories and facilities. However, there are some constraints to be considered:
• Return on investment. At this time, the investment may not be reasonable for a third party
interested in the business aspect alone. Different considerations may be applicable to
manufacturers and utilities concerned with the field reliability of the future high-power
system. A large investment in a new laboratory may be negligible when compared to the
cost of a fault in service;
• Time required for creating or updating facilities versus the stringent time schedule of
some projects.
Should the component testability remain an impassable challenge, alternatives may be studied
and recommended with CIGRE IEC support for UHV equipment, such as:
• Revision of test parameters accurately taking into account future system characteristics,
such as the use of continually more efficient surge arresters;
• Adoption of non-standard test procedures to achieve the required stress on the equipment;
• Development of suitable and well-proven “virtual testing” techniques;
• Development of a suitable on-site test program providing sources and instrumentation for
commissioning and field testing;
• Application of suitable monitoring systems, including measurement such as:
o Local temperature measurements in the winding and core of transformers and reactors
43
o Partial discharge detectors in GIS
o Ground current detectors on transformer bushings
A report on the testability of UHV components was recently presented in [15].
2.6.4 Overviews of field experiences of the projects

The technical specifications of UHV substation equipment are summarized in the Table 2.6.1.
Table 2.6.1 Rated electrical characteristics of the 1050 kV pilot plant

System (general) Circuit-breakers
Rated voltage 1050kV Terminal fault Short-line fault Capacitive switching
Short-circuit Surge 300 Breaking
LIWV 2250 kV 63 kA 900 A
breaking current impedance ohm current
Short-circuit Amplitude
current factor Voltage
SIWV 1800 kV 45 ms 1.6 1.2
time constant factor
(ms)
Short-circuit Out-of-phase Switching
PFWL 1050 kV 160 kA (out-of-synchron.: 90°) Uc (1-cos) 2.060 kV
making current
Rated current 8000 A Applied voltage Breaking 25 kA
busbar bay 4000 A during making 606 kV current
operation
1st pole –to-
Earthing Switch 1.3 Voltage factor 1.4 Autotransformer
clear factor
Induced current 50 A Transient recovery Transient recovery Voltage
electrostatic 1000 A 1000/√3/400/√3/12.2
electromagnetic voltage voltage kV
Induced voltage 50 to 95 1640
T10: Uc (kVp) Uc (kVp) 1890 400/400 MVA
electrostatic 50 to 70 11.5 Rated power
electromagnetic dV/dt (kV/µs) dV/dt (kV/µs) 3.0 (single-phase unit)
T60:Uc (kVp) 1520 Applied voltage

Rated
Oil-filled cable 3.3 during making 849 kV 2250/1300/95 kV
dV/dt (kV/µs) operation LI WV
T100:Uc(kVp) 1370
Rated voltage 1100 kV Rated SIWV 1800 kV/consequent
dV/dt (kV/µs) 3.3
Rated power 7 normal Short circuit
15
GVA 9 except. voltage (%)
Table 2.6.2 Italy’s specifications

Highest voltage 1050 kV
Items Transformers CBs (GIS)
LIWV (kV) 2250 kV 2250 kV
SIWV (kV) 1800 kV 1800 kV
AC voltage 1000 kV 1050 kV
1.3 for T100
44
2.7 BPA 1200 kV Lyons R&D project
2.7.1 Features of BPA’s 1200 kV research project
2.7.1.1 Background
Bonneville Power Administration (BPA)’s 1200 kV Transmission Line Test and Development
Program was initiated in the mid-1970s to meet the need for economical transportation of large
amounts of electric power over limited rights-of-way and with minimum power loss. There are
several advantages to a 1200 kV transmission system, the most important of which are better
utilization of available right-of-way, reduced environmental impact (compared to multiple
550 kV lines), and reduced losses. Before committing to a commercial transmission system at
1200 kV, BPA needed to test the proposed design, solve technical problems, eliminate economic
uncertainties, and gain public acceptance. So, they set out in 1974 to build a prototype 1200 kV
line and evaluate its performance.
The Lyons 1200 kV prototype line was the first transmission line operating at the 1100-1200 kV
level utilizing actual 1200 kV construction. As such, the line is a prototype of the commercial
1200 kV line as opposed to being a test facility designed strictly for research. In 1974, BPA
authorized the construction of a three-phase, 1200 kV prototype transmission line in order to
investigate the technical, economical and environmental feasibility of transmitting electric power
at this voltage. After the 1200 kV test facility was completed in 1977, an extensive testing and
evaluation program was launched. This section summarizes the results of electrical
investigations, substation equipment performance, and electric field effect studies. Field
experience includes both transmission line and substation component failures. [16]
2.7.1.2 Design of the Lyons facility

The three-phase 1200 kV line at Lyons was 2.1 km long and was built on an existing 230 kV
right-of-way parallel to the Detroit-Santiam No. 1, 230 kV line. The dead-end tower at the far end
of the line was a self-supporting, lattice type while the substation dead-end tower and the
midpoint angle tower was a guyed tubular steel pole construction.
Fig.2.7.1 Plot plan for the Lyons test facility
2.7.1.3 Line configuration

A triangular configuration was used for the three phases in order to minimize right-of-way
requirements. The design of the self-supporting, lattice type suspension tower for the Lyons 1200
kV line is shown below.
As seen in Fig. 2.7.2, the spacing between the two lower phases will be 22.0 m for the major
45
portion of the line. The location of the OH grounding wires provides a shielding angle of about 5°
at the towers.
Fig.2.7.2 Self-supporting suspension tower for Lyons line
2.7.1.4 Air gap clearances

The line was designed for a switching surge factor of 1.5 per unit. Based on a maximum design
voltage of 1200 kV, this results in a minimum air gap clearance of 620 cm between conductor
and workman. To allow for tower inspection and tightening of bolts on an energized line, another
30 cm of air gap clearance is provided. This results in a minimum conductor-to-steel clearance of
650 cm.
A minimum conductor-above-ground clearance of 23.2 m was adopted. In the electrical test area,
however, the height of all three phases was varied so that the minimum height above ground for
the lower phases could be adjusted to between 17 and 25 m.
2.7.1.5 Line insulation

V-strings were used in all phases of the suspension towers to minimize conductor swing. Glass,
polymer and porcelain cap-and-pin insulators were used with an M&E rating of 535 kN. Each leg
of the V-strings can accommodate 29 units of 240×380-625 mm insulators. This results in a
leakage distance of 17.6 m, which is equivalent to a specific creepage distance of 2.5 cm/kV rms.
Each dead-end insulator consists of four parallel strings, each composed of 29 units of the same
type as used for the suspension towers.
2.7.1.6 Line conductors

The 1200 kV line was initially strung with an 8×40.7 mm (Chukar) conductor per phase and
equipped with two overhead ground wires. This was expected to result in an audible noise level
of approximately 49 dB (A) at 30 m lateral distance from the centre line at 1100 kV.
Preliminary calculations indicated that the optimum bundle diameter should be slightly more
than 75 cm. Due to anticipated mechanical problems associated with sub-conductor oscillation, a
bundle diameter of 107 cm was selected. This will result in a voltage gradient (at the conductor)
of about 13.5 kV/cm, at a voltage of 1100 kV.
46
2.7.1.7 UHV substation design
The 230/1100 kV power transformer consists of three single-phase units, each rated 50 MVA
(150 MVA 3-phase rating) at a 55ºC rise. The transformer was equipped with no-load taps for a
voltage range of 1100 to 1250 kV. Special surge arresters permitted a reduction of the SIWV of
the 1100 kV winding to 1800 kV. The LIWV was specified at 2050 kV.
Table 2.7.1 Technical specifications of power transformer
1200 kV Transformer
MVA rating 150/168 MVA (55C/65°C rise)
LIWV (BIL) 2050 kV
SIWV (BSL) 1800 kV
1250 kV
1212 kV
HV taps 1175 kV
1137 kV
1100 kV
LV taps 241.5 kV
239.0 kV
236.0 kV
Impedance 18 %
Weight 336,000 lbs
Height 47 ft
Table 2.7.2 Technical specifications of surge arrester

Surge arresters
Duty cycle rating 770 kV
MCOV 692.8 kV
Protective levels
LIWV 1860 kV 26 kA, 8×20 µs
SIWV 1470 kV 1 kA, 45×90 µs
2.7.2 Overview of field experience of BPA’s 1200 kV project -Electrical

test and development program
2.7.2.1 Corona Studies

Long-term corona measurements were conducted on the 8×41 mm and 7×41 mm configuration at
the Lyons facility. For both configurations, the horizontal phase spacing was 22 m, vertical phase
spacing was 18.3 m, and the bundle diameter was 1.07 m. The vertical height of the outside
phases was 16.8 and 22.9 m for the 7 and 8 conductor bundle configurations, respectively. [17],
[18], [19]
Table 2.7.3 Conductor corona performance
Lyons prototype line BPA single circuit, delta 550 kV
Location
(Oper. voltage 1150 kV) lines (Oper. voltage 525 kV)
Electrical effect 8×4lmm 7×4lmm 2×4lmm 3×33mm
AN: (dBA) 53 56 56 46
RI: Fair/Foul weather,
52/70 46/68
QP 0.5 MHz (dB/uV/m) 46/65 58/77
TVI: QP 75 MHz (dB/uV/m) 13 12 26.5 24
Ozone: (ppb) ND* ND* NM* NM*
Corona loss: (kW/km) NM* 43 NM* NM*
The above table shows the corona performance of typical BPA 550 kV lines, and clearly
47
demonstrates that with a suitable selection of conductor design and ground clearances, the corona
performance of 1200 kV transmission lines can be made comparable with the already accepted
performance of 550 kV lines.
The results indicate that decreasing the number of subconductors from 8 to 7 increases AN by
3 dBA and RI by 12 dB. However, TVI is essentially unaffected by the conductor configuration.
Results also indicate that TVI is caused primarily by insulator and hardware corona rather than
conductor corona. No increase was found in mean ozone concentrations at line height or at
ground level, which could be attributed to conductor corona.
Corona loss levels are reported only for the 7-bundle configuration. Since only half the line was
conductored with this configuration, and since corona sources from the entire line contribute to
the measured losses, the losses from a line strung with a 7×41 mm bundle can be expected to
differ slightly. Studies were also carried out on 1200 kV tower and insulator hardware in a
laboratory environment. Single-phase corona tests for suspension-type tower hardware conducted
in the laboratory agreed satisfactorily with 3-phase corona tests performed at Lyons.
However, single-phase corona tests carried out on dead-end hardware in the laboratory did not
agree with 3-phase tests performed at Lyons. It is suspected that the primary cause of this
disagreement is the influence of other phases on the electric field environment. Over a period of
years, additional tests were conducted to evaluate the impact of mid-span clearance and phase-to-
phase spacing on RIV, AN, TVI, ozone generation and corona loss.
Table 2.7.4 Line configurations used at Lyons 1200 kV facility
Line Number and Energized phases Mid-span Phase A-C
configuration diameter of subcon and arrangement clearance (m) spacing (m)
I
B
Nov. 1977 to Sept. 1978
8×41 mm C • A 24.4 22
• •
II
Grnd.
Jan-May 1979
7×41 mm B ο A 22.9 22
• •
III
Grnd.
7×41 mm B ο A 19.8 22
June-Nov. 1979
• •
IV
B
7×41 mm C • A 16.8 22
Jan-April 1980
• •
V
B
May- 1980
7×41 mm C • A 16.2 13
• •
Table 2.7.5 Corona effects for different line configurations at Lyons 1200 kV test facility
Operating voltage=1150kV
Line configuration I II III IV V
AN*- A-wt. L50 level during rain (dBA), 15 m
from outside phase
53 47 50.5 56.5 61.0
RI*- QP L50 level (dB,μV/m), 15 m from outside phase 42/61 43/62 46/64 54/73 50/71
a) 1 MHz. ANSI, Fair/Rain
b) 0.5MHz. CISPR, Fair/Rain 46/65 47/66 50/68 58/77 54/75
TVI – 75MHz. QP L50 level during rain
13 12 13 12 9
(dB,μV/m), 40m from outside phase
Ozone --- (ppb) None None None None
―
detect. detect. detect. detect.
Corona Loss – L50 level. kw/km, during rain
― 22 24 43 45
48
Table 2.7.6 Comparison of predicted AN .RI. and TVI levels for 1100 kV
with 735 and 500 kV transmission lines used in the USA and Canada
Bad weather Fair weather Bad weather
Voltage (kV) Line
AN (dBA) μV/m)
RI (dBμ μV/m)
TVI (dBμ
1150 8 × 41.7 mm 53.0 38.6 13.9
1150 7 × 41.7 mm 56.3 43.4 18.7
735 4 × 35.2 mm 51.9 44.4 21.9
735 4 × 30.4 mm 55.8 48.9 26.4
525 2 × 41.7 mm 55.3 49.7 27.3
525 3 × 33.1 mm 47.0 42.6 20.3
525 4 × 24.1 mm 43.0 37.5 15.1
525 4 × 21.6 mm 45.2 30.4 18.0
Table 2.7.7 Line configurations – Reduced and normal phase spacing

Normal phase spacing Reduced phase spacing
Line parameter testing period
Jan. April 1980 Jan. April 1981
Line length 2.1 km 2.1 km
Sub-conductor number 7 7
Sub-conductor diameter 40.7 cm 40.7 cm
Outside phase
16.8 m 17.8
Conductor height
Centre phase conductor height 35 m 35 m
Horiz. phase spacing (Outside phases) 22 m 12.4 m
Conductor surface grad EA.C =15.7 kV/cm EA.C =17.1 kV/cm
(based on 1150kV) EB=14.8 kV/cm EB=15.2 kV/cm
Table 2.7.8 Corona effects for different phase spacing

7 × 41 mm Lyons 30, 7 × 41 mm Lyons 30,
Electrical Effect Voltage = 1150kV, Voltage = 1150kV,
Reduced Phase Spacing Normal Phase Spacing
AN*- A-wt. L50 level during rain (dBA), 15 m from
60 56.5
outside phase
RI*-QP L50 level (dB,μV/m), 15 m from outside phase 49/71 54/73
a) 1MHz, ANSI, Fair / Rain 53/75 58/77
b) 0.5MHz, CISPR, Fair / Rain
TVI – 75MHz. QP L50 level during rain
14 13
(dB,μV/m), 40m from outside phase
Electric field – Maximum at mid-span.
12 14
(kV/m). 1m above ground
Corona Loss – L50 level. kw/km, during rain 47 43
2.7.2.2 Electric Field Measurements

The vertical clearance used for the 8×41 mm configuration was 22.9 m (outside phases) to give a
maximum electric field comparable with one of BPA's typical 550 kV transmission lines. The
maximum measured electric field (1 m above ground) was 7.5 kV/m.
To investigate higher field strengths, the outside conductors were lowered to 16.8 m by
introducing additional conductor sag. The maximum measured electric field was 14 kV/m.
Measurements of induced current and voltage (for various vehicles and fences) under the Lyons
test line indicate that, even for the largest vehicles, the maximum induced current at road
49
crossings will not exceed the 5 mA criteria established by the National Electric Safety Code.
To investigate the effects of 1200 kV transmission on wood poles, a simulated 115 kV wood pole
distribution line was built under the Lyons Test Line. Testing included long-term monitoring of
surface and internal pole leakage current, electric field and space potential measurements, and
evaluation of the various wood pole hardware installations. It was found that, with the wood pole
line conductors grounded, four of the nine poles showed signs of burning in areas exposed to the
highest electrical stresses. Ungrounding the wood pole line significantly increased burning. The
electric field measurements were obtained on the configurations tabled on Table 2.7.8.
Table 2.7.8 Electric Field Measurements (1150 kV)
Line Configuration Electric Field Maximum at mid-span (kV/m 1m above ground)
I 7.3
II 9.9
III 12.5
IV 14.1
V 11.5
2.7.2.3 Insulation studies

The primary objectives of the insulation studies were to:
(1) Evaluate the 60-Hz performance of insulation systems;
(2) Evaluate the lightning and switching impulse performance of insulation systems;
(3) Evaluate the performance of non-ceramic insulators; and
(4) Obtain data on the dielectric strength of air gaps and insulators necessary to optimise
1200 kV insulation systems.
The corona performance of different types of porcelain, glass, and non-ceramic insulators was
studied at Lyons by means of night-viewing devices and high-speed photography. The ceramic
insulators performed as expected. However, some non-ceramic insulators (NCIs) experienced
surface erosion due to corona and arcing, pointing out the need for adequate grading of NCIs.
One NCI failed dielectrically and separated after about 6 months of operation due to severe
internal discharges. [20]
The Lyons 1200 kV line insulators that were exposed to natural contamination and washing have
performed well, with one exception. During a snowstorm, after a heavy layer of wet snow had
built up on the insulators, two flashovers occurred on dead-end insulators. No problems have
occurred with the 1200 kV substation insulators.
A long-term test program was conducted to investigate the 60 Hz performance of artificially
contaminated insulator strings. The insulators on the two outside phases of the line were
artificially contaminated at equivalent salt deposit densities (ESDD) found on BPA's 550 kV
system. Preliminary results indicated that insulator contamination may be a more serious problem
at 1200 kV than at lower voltages.
Lightning and switching impulse testing on 1200 kV air gap configurations was conducted at
BPA's laboratory in Vancouver, where there is an outdoor test tower and a 5.6 MV, 448 kJ
outdoor impulse generator. Impulse tests were performed on basic conductor-to-tower air gaps
with and without insulators, as well as on the Lyons 1200 kV prototype suspension and dead-end
tower designs. The effects of geometrical and electrical conditions were also investigated.
50
A simulated Lyons suspension-type tower without insulators and a positive 250/2500 µs
switching impulse waveform provided a dry, minimum critical flashover voltage of
approximately 1800 kV for both centre and outside phase configurations. For a positive 1.2/50 µs
lightning impulse waveform, the critical flashover voltage was 3600 kV. The Lyons dead-end
tower was about 5% greater in dielectric strength than the suspension tower.
2.7.3 Specific issues and technical challenges for the operations of

BPA projects
2.7.3.1 Transformer design

Lyons had three 230/1200 kV autotransformer units with a rating of 150/168 MVA at a 55/65ºC
rise, respectively. These can be considered power transformers for practical insulation purposes
with no-load taps from 1100 to 1250 kV. The BIL of the 1200 kV windings is 2050 kV, the
lowest possible in the present state-of-the-art. The SIL is 1800 kV, the impedance 18%.
Extensive factory acceptance testing, including a one-hour partial discharge test, was performed
at 960 kV on each unit.
BPA's transformer operating experience has been good, except for the failure of one unit. The
failure was caused by a manufacturing defect: during assembly, an excessive amount of glue (or
voids in the glue) was used to bond the inner surface of the static plate between coils 25 and 26.
Also, magnetic particles were found on the surface of the insulating washers between these coils.
The flashover arc path was between the washers. After repair, all factory tests were successfully
passed. The transformer was reinstalled and has since performed satisfactorily.
2.7.3.2 Surge Arrester design

Based on the sustained dynamic overvoltages at Lyons under fault conditions and line
energization, special conventional current-limiting gapped surge arresters were necessary. The
arresters’ control gaps and blocks needed to be specifically designed and carefully graded so that
both the switching surge and lightning impulse protective levels would be maintained as low as
practical. Manufacturers indicated that a protective level of 1530 kV could be obtained with an
arrester rated at 768 kV. This corresponds to a protective level of 1.56.
Following the construction of the test facilities, field tests were conducted to assess the
overvoltage magnitude and waveform of the switching transients. Surge arrester sparkover
occurred at Lyons in practically all switching performed by the 230 kV circuit switches, limiting
the line overvoltage to 1450 kV (1.5 per unit) and below.
A total of four arrester failures have occurred since the initial installation of the current-limiting
gap arresters. These failures are associated with improper grading of the units, causing external
flashovers as well as the overheating of the internal grading circuits.
Metal oxide surge arresters were installed at Lyons in 1980 and 1981. Field tests conducted in
1981 confirmed the switching surge protective level and the surge currents under various
switching modes. The arresters maintain a protective level of 1.5 per unit.
Thermal stability of the metal oxide arresters was a concern since the maximum 60-Hz voltage is
about 10% higher than the stress on typical commercially available arresters. Since that time,
BPA has purchased and successfully applied 1.5 p.u. arresters to the 230 and 500 kV systems.
Table 2.7.4 BPA specifications
51
Highest voltage 1200 kV
Items Power Transformer
LIWV (kV) 2050 kV
SIWV (kV) 1800 kV
AC voltage 1250 kV
Maximum continuous 770 kV
operating voltage
Protective levels 1860 kV (Lighting surge at 26 kA, 8 µs/20 µs)
1470 kV (Switching surge at 1 kA, 45 µs/90 µs)
2.8 Russia’s 787-1200 kV interconnected grid
2.8.1 General characteristic of Russia’s interconnected grid

The former USSR was 22.4 million km2 in area with a population of 286 million people; after
dissolution of the USSR, the remaining territory of Russia equalled 16.9 million km2 and the
slowly declining population is now about 140 million. Russia is extremely rich in natural
resources (45% of the world’s natural gas, 23% of the coal, 6% of the oil), but its main sources of
energy, including hydro, were concentrated in the Asian part of the USSR, whereas the
population was concentrated mainly in the European and southern parts, leaving vast areas in the
polar region and Siberia almost completely uninhabited. As a result, the distance required to
transport bulk electric energy was generally quite long, requiring higher voltages.
Following several decades of continually increasing demand for electric energy, Russian
electrical engineers designed and built a huge EHV AC interconnected grid with 363-525-765-
1200 kV nominal voltage (maximum operating voltage, respectively, 363-525-787-1200 kV).
EHV and UHV voltages were attained as follows: 420 kV in 1957, 525 kV in 1959, 787 kV in
1967, and 1200 kV in 1985. In 1990, just before the dissolution of the USSR, the EHV/UHV
interconnected grid consisted of:
• 363 kV lines: 3110 km
• 420–525 kV lines: 4330 km (420 kV lines applied to international transmission only)
• 787 kV lines: 7300 km
• 1200 kV lines: 1900 km (900 km operated at 1150 kV, the rest at 500 kV).
That year, total electricity production in the USSR was 1850×103 GWh and the total capacity of
power plants was 344 GW, with 290 GW being connected to the interconnected grid. Together
with the interconnected grid of East European countries, 11 time zones were covered. The entire
grid was equipped solely with Russian electrical equipment.
The USSR’s power industry was state owned, government ruled and highly centralized, which
was favorable for implementing large-capacity units and large power plants as the most
economical. Maximum unit capacity at that time was 1200 MW at steam power plants, 950 MW
at hydro power plants and 1500 MW at nuclear power plants, with the maximum capacity of a
single power plant reaching 6400 MW for steam or hydro and 6000 MW for nuclear. About 100
regional power companies operated power plants and transmission and distribution networks
52
under the auspices of the Ministry of Electric Power and Electrification (over the years the name
of this Ministry changed several times). These networks were united in 11 regional power pools
with a capacity ranging from 5 to 45 GW each. In total, the USSR’s interconnected grid produced
and transmitted up to 89% of the entire Soviet electricity production (in 1990) leaving 11% to
relatively small local power utilities in remote areas. With the East European countries, this
interconnected grid stretched from east to west over a distance of almost 9000 km. Industry
centralization was favorable for unifying technical specifications on the equipment, standards,
design and operational norms on a state-wide scale, as well as for analysis of operational data.
The dissolution of the USSR into 15 independent states in 1991 partially destroyed the integrity
of the grid, and the economic crisis that followed the dissolution drastically reduced the demand
for electric power (in 1993, Russia’s GDP dropped to about 50% of that in 1990 and electric
energy consumption decreased to 75%). This situation made the full-voltage operation of the
1200 kV transmission system economically ineffective, and it was turned into 500 kV. Only in
the last few years has Russia’s electric power industry recovered from this crisis.
After the disintegration of the Soviet Union, the lion’s share of natural resources remained in
Russia, as well as the major part of the EHV grid. As of 1995, Russia had 3690 km of 500 kV
lines, 2800 km of 750 kV lines and 950 km of 1150 kV lines. Russia’s interconnected grid still
extends to 7000 km, and is the second largest grid in the world. It includes 5 regional power
pools, and its share of Russia’s total power plant capacity reached 93% (205 and 193 GW,
respectively) in 2001 at a total electric energy production of 8880 GWh (in the same year). The
entire interconnected grid divided between 15 newly independent republics still exists, but its
operation and power exchange suffer from numerous intergovernmental contracts and pricing
problems and it is far below the technical capability of 9-10 GW estimated in 1990.
Although some privatization has touched Russia’s power industry, it still is mainly owned by the
state, especially in terms of major power plants and EHV/UHV transmission systems; electric
energy and heat tariffs are still controlled by the government. As of 2005, the Russian state joint
stock company “Interconnected Grid of Russia owned:
• Total shares of fossil plants with a capacity of 1 GW and over, and of hydro power plants
with a capacity of 0.3 GW and over;
• 49% shares of all local electric utilities owning smaller power plants and local electric
grids;
• All EHV transmission systems and substations forming the interconnected grid;
• Central and all regional boards of the power pool;
• Design and research organizations serving electric utilities.
This stock company has started the process of radical changes aimed at further decentralization
and privatization that is scheduled for completion in 2008.
When the implementation of 1200 kV transmission was initially decided, the transmission
systems had to bring bulk power from Siberian coal fields (where huge power plants were
planned) to the central part of the Soviet Union. The maximum power transmitted through
1200 kV lines was estimated as 5.5 GW and required the implementation of FACTS technology.
The total length of prospective 1200 kV transmission lines at that time was estimated as
10,000 km with individual line lengths ranging from 300 to 600 km. The transport had to be
supplemented with +750 kV, 6 GW HVDC transmission lines 2400 km long. The first four
53
1200 kV lines were built and put into operation, with two of them (total length about 900 km)
operating under full voltage for testing purposes. However, the planned power plants were not
constructed in time and the disintegration of the USSR led to a drop in consumer demand for
electric energy; therefore, after two successful years of operating under full voltage, these two
1200 kV lines, as well as all newly built 1200 kV lines were changed to 500 kV to save on losses
and maintenance expenses.
Russia’s GDP and electric energy production has shown steady growth during the last several
years. In the period of downturn, the construction of new transmission lines almost came to a
complete standstill, but with the start of economic recovery, the construction of EHV lines
resumed, including one 1150 kV line 445 km long temporarily operated at 500 kV. The designing
of new 1150 kV lines resumed, with the goal of reshaping the grid according to the changed state
boundaries. The main energy transit is now expected to be between Siberian power sources and
the Urals area. The total length of already designed 1150 kV lines exceeds 1600 km. The
construction of +500 kV DC transmission lines, 1800 km long, is under discussion, and DC back-
to-back substations will be used to increase the export of Russia’s electric energy to UCPTE that
has more stringent requirements on the quality of electric energy compared to Russia. Initial
predictions published in 2001 were for an increase to 1400-1600×103 GWh in Russia’s electric
energy production by the year 2010; the total capacity of 100 GW must be refurbished or built;
new nuclear plants of about 25 GW and new gas turbine and combined cycle fossil power plants
of about 30 GW must be built. In 2006, the Russian government approved a more moderate
program stating that newly built power capacities will reach 21.8 GW in 2006-2010 and at the
same time, 4.2 GW of aged equipment would be taken out of operation. The construction of
10,000 km of new HV lines at 220 kV and above is expected, with an investment of about $6
billion in 2006 and $20 billion annually starting in 2008.
2.8.1.1 Selection of nominal and maximum operating voltages

In the past, the USSR had two systems of nominal transmission voltages,
• 35-150-330 kV in the western and southern parts of the country (400 kV was retained on
several connections to adjacent countries);
• 35-110-220-500 kV in the central and eastern parts of the country.
As industry and power consumption grew, the need arose for an overlaid EHV ‘supergrid’ in both
systems. Economic analysis showed that every next nominal voltage had to exceed 2–2.5 times
the previously existing highest voltage in the system. So, a nominal voltage of 750 kV was
selected for upgrading the first system and 1150 kV for the second one. In the European part of
the USSR, the two systems became entangled, which required, in addition to obvious 330/750
and 500/1150 kV connections, the use of 220/330, 330/500 and 500/750 kV autotransformers.
The following areas of application of 750 and 1150 kV were economically determined for
maximum load duration of 5000-7000 hours a year:
• 750 kV compared to the existing 500 kV transmission: at a distance of 700 km if
transmitted power exceeds approximately 1.2 GW, or at a distance of 400 km if power
exceeds 2.2 GW;
1150 kV compared to the existing 750 kV transmission: at a distance of 700 km if power exceeds
approximately 2.8-3 GW, at a distance of 400 km if power exceeds 3.6-3.8 GW.
54
2.8.1.2 Maximum operating voltage
Taking into account problems with electrical equipment insulation and corona effects on
overhead lines, the following margins between the nominal and highest permissible operating
voltages were approved in the USSR:
• 500 and 750 kV: 5% (i.e., 525 and 787 kV, respectively)
• 1150 kV: 4.3% (i.e., 1200 kV).
Line length, number of circuits, transposition: In the erected 750 kV grid, the line length
varies between 300 and 550 km. In both the constructed and designed 1150 kV grids, the line
length varies from 400 to 700 km.
In accordance with Soviet practise, all lines are single-circuit lines; analysis showed that double-
circuit lines are more prone to the loss of both circuits compared to two separate single-circuit
lines, especially if routed via different rights-of-way. For this reason, if the capacity of one line is
not sufficient for carrying the required power, two lines may be constructed leading to different
areas of consumption, creating “loops” within the grid.
All constructed and planned 750-1150 kV lines have a full cycle of transposition, thus reducing
asymmetry and improving the conditions for applying single-pole auto-reclosing.
2.8.1.3 Shunt reactor positioning and connection - Degree of line

capacitance compensation
In all cases 787 and 1200 kV shunt reactors with their surge arresters are placed on line ends and
made switchable. Their circuit breakers are equipped with air-gaps for instant connection of
switched-off reactors to the line. That permits to use the reactors effectively not only to balance
reactive power in operational modes, but also for limiting temporary, switching and even
lightning overvoltages.
Only one design of 787 kV single-phase shunt reactor is in usage with the capacity of the group
330 MVAr, so the degree of shunt compensation varies with the length line being usually 60-
90%. For 1150 kV lines also only one design of shunt reactor was adopted with group capacity of
900 MVAr, and desired shunt compensation was estimated as 90-100%.
Neutral point insulation in shunt reactors permits to use them in four-legged schemes to suppress
the secondary arc at single pole auto-reclosings.
Series compensation didn’t receive practical application in the existing 787-1200 kV Grid,
although may be considered for some future projects.
2.8.1.4 System of limitation of temporary, switching and lightning

overvoltages
The system for limiting temporary overvoltages and switching surges in the Soviet/Russian
750 kV transmission system that inherited the positive experience of 500 kV transmission
included:
• Positioning shunt reactors on the ends of EHV/UHV lines;
• Equipping their circuit breakers with air gaps for instant connection of reactors to the line
if a reactor is switched off in the pre-failure operational mode;
55
• Automatic mechanism for closing, without time delay, the reactor circuit breaker in all
poles if shunt reactor current appears in at least one pole, or in the event of an automatic
command to switch off at least one pole of the line;
• Automatic mechanism for connecting switched-off shunt reactors to the line and then, if
necessary, switching off the line in the event of a temporary overvoltage. This automatic
mechanism checks the peak value of overvoltage and delays switching off the line in
accordance with permissible TOV characteristics;
• Automated operation of planned and failure switching of the line;
• Application of switching surge arresters on transformers and shunt reactors.
The 1150 kV transmission system also includes pre-inserted resistors in line circuit breakers. A
view of Russia’s 1150 kV shunt reactor circuit breakers with air gaps is shown below. [21]
Fig. 2.8.1 1200 kV circuit breakers applied to shunt reactor

As Russia’s EHV/UHV lines are generally long, and at the first stages of grid development some
of the connected systems may have increased reactance, the limitation of temporary overvoltages
plays a decisive role in overvoltage protection of UHV transmission. Russian standards for
EHV/UHV equipment include norms for permissible TOV levels versus their duration and
frequency of repetition. These norms are jointly determined by manufacturers and customers and
are based on “natural” insulation capabilities of different types of equipment for which the
insulation is traditionally designed based on operational voltage and permitted switching and
lightning overvoltages.
2.8.1.5 Levels of overvoltage limitation and arrester characteristics

Switching surge (and lightning overvoltage) limitation levels were initially selected for zinc-
oxide gapped arresters available in the USSR, and all existing equipment in the Soviet’s 787-
1150 kV transmission systems corresponds to those levels. Later, when MOVs became available,
lower limitation and withstand voltage norms were adopted for equipment that may be developed
and installed in the future (see Table 2.8.1).
Table 2.8.1. Overvoltage limitation and arrester currents in the Soviet EHV/UHV transmissions
Parameter Nominal voltage, kV 750 1150
Arrester type
Switching surge limitation, p.u. of crest value GSOA* 2.1 1.8
56
of L-G rated voltage MOV** 1.85 1.6
Switching surge current, kA peak GSOA* 1.8 1.5
MOV** 1.8 2.2
Lightning impulse limitation, p.u. of crest GSOA* 2.57 2.05
value of L-G rated voltage MOV** 2.05 (2.15x) 1.8
Lightning impulse current, kA peak GSOA* 10 14
MOV** 10 (15x) 15
Notes: * Overvoltage protection system with gapped zinc-oxide surge arresters (GSOA). All installed
equipment corresponds to this row.
** Overvoltage protection system with MOVs can be used with newly developed equipment having lower
x
insulation levels (see below); In rare cases of increased current impulse.
Insulation levels: Russia’s standards prescribe mandatory levels of withstand test voltage for
EHV/UHV equipment with special attention paid to IEC recommendations. Some extracts are
provided in Table 2.8.2. These norms are valid for a maximum location altitude of 500 m above
sea level. In order to use the 4-legged reactor scheme to suppress secondary arc current at single-
pole rapid auto-reclosing, shunt reactors must have their neutrals insulated from the ground.
Insulation class of the neutral is 35 kV for 750 kV reactors (1-min test voltage of 85 kV); for the
neutral of 1150 kV shunt reactors, the 1-min test voltage is 120 kV.
Table 2.8.2 Technical specifications of 787 and 1200 kV Equipment

Applied test voltage 50 Hz ( kV rms) SI (kV peak) LI (kV peak)
Rated voltage and type of equipment 1 min Long applied Smooth rise FW FW CW
750 kV transformers/ shunt reactors 800/900 1550/1675 2100/2250 2250/2400
635
(two values if different), (L-G) (750) (1425) (1800/1950) (1950/2100)
750 kV equipment, CT, capacitive 950 1550 2100
PT, (L-G) (830) (1425) (1950)
750 kV circuit-breaker, between the 1550/ 2250
1400 2100 2550
contacts of the pole 1350x (2000)
750 kV insulator string, (L-G) 530^ 1500
1550
750 kV air gap, (L-G)
(1425)
2550
750 kV air gap, (L-L)
(2400)
1150 kV transformers/ shunt reactors, 1100 900 2100 2550 2800/3200
1300*
(two values if different), (L-G) (1000) (900) (1800) (2250) (2550)
1150 kV apparatus, CT, capacitive PT, 1150 900** 2100 2900
1300*
(L-G) (1100) (900**) (1800) (2400)
1150 kV circuit-breaker, internal and
external insulation between contacts 2000 2000 3100 2900 3200
when open
x
Notes: * For equipment developed prior to 1985; ** CT only; ^ Under contamination conditions; Dry and
xx
rainy conditions, respectively; For transformers and shunt reactors, respectively.
(SI: switching impulse; LI: lightning impulse; FW: full wave; CW: chopped wave. Without brackets:
equipment now used in transmission systems and protected by gapped zinc-oxide surge arresters; With
brackets: values for newly developed equipment protected with MOV)
2.8.1.6 Line parameters

To facilitate the construction of lines and to reduce corona noise and losses at moderate current
loads, Russia’s bundled phases have an increased number of smaller-diameter Aluminium
Conductor Steel Reinforced (ACSR) sub-conductors. The typical line parameters are shown in
Table 2.8.3. All lines of these voltages are transposed, single-circuit, and protected with two
shielding wires that are also used for high-frequency communication transmission. Guyed
support towers are widely used.
Table 2.8.3 Parameters of Russian 787 and 1200 kV lines
Nominal voltage 750 kV 1150 kV
Typical length, km 300-550 400-700
57
Maximum transmitting capacity, GW 1.0-2.1 4-6*
Phase conductor (AL only) cross section, mm2 1200-2000 2600-3200
Typical bundling of the phase, conductor diameter and 5xAS-240/56; 2.24cm, 30cm 8xAS-330/43; 2.75cm; 40cm
spacing in the bundle 4xAS-400/93; 2.91cm; 60cm 8xAS-400/51; 2.75cm; 40cm
4xAS-500/64; 3.06cm; 60cm
Phase-to-phase distance, m 17.5-19 21.5-25
Surge impedance in positive sequence ZW, Ohm ~265 ~250
Same in zero sequence ZW0 , Ohm ** 455/520 435/505
Inductance in positive sequence X1, Ohm/km 0.29 0.27
Same in zero sequence X0, Ohm/km ** 0.65/0.85 0.63/0.85
Capacitive conductivity in positive sequence b1, µS/km 4.02 4.36
“Natural surge load” of the line, GW ~2100 ~5300
Capacitive charging current of the line I1, A rms/100km ~180 ~290
Notes: * Using controllable shunt compensation (FACTS) and automatic regulation in the adjacent grid.
** Prior to the slash, shielding wires are grounded at both ends of each anchor span; after the slash,
shielding wires are grounded only from one side of each anchor span to reduce power losses.
2.8.1.7 Auto-reclosing application

In 787–1200 kV lines, single-pole faults are almost the only kind of faults observed in operation.
Most occur through arcing and can potentially be resolved using rapid auto-reclosing.
Accordingly, Russia employs single-pole auto-reclosing as a mandatory method, and three-pole
auto-reclosing is usually used as the second line of defence if during the dead time at single-pole
opening of the line, the secondary arc did not extinguish. Using single-pole auto-reclosing is
particularly important for single-circuit UHV lines, as it minimizes the disturbance experienced
by the grid at line faults. As the lines are transposed, the secondary arc current is less than that in
untransposed lines, and the application of four-legged reactors effectively reduces it. In
symmetrical mode of line operation, the neutral reactor is shunted by a circuit breaker, and to
protect the neutral insulation when the neutral reactor is de-shunted, an arrester is placed in
parallel to the neutral reactor. Additional measures can be used if the 1200 kV lines are heavily
loaded and HSGS fully resolves the problem of secondary arc extinction.
2.8.2 Overviews of field experience and evolution of Russia’s 787 and

1200 kV transmissions
2.8.2.1 Organization of research, designing and testing

EHV/UHV equipment development, and transmission system design and testing were executed in
the USSR based on state programs with government funding. Particularly, the development of
787 kV equipment and prototype testing started in the 1960s. In 1965, the experimental 787 kV
commercial transmission system, 87 km long, with two 787 kV substations at its ends (Bely Rast
and Konakovo), was put into operation, and various tests were performed jointly by a group of
researchers from equipment manufacturers and the power industry, represented, respectively, by
the All-Union Electrotechnical Institute (VEI) and the All-Union Electric Power Research
Institute (VNIIE). Numerous improvements were made to the equipment following the tests.
From 1967 until about 1988, each newly constructed 787 kV transmission line was subjected to
similar tests performed in accordance with national programs developed annually and approved
by the Ministry of Electric Power and Electrification and the Interconnected Grid Pool. The tests
were executed by VNIIE and VEI with the participation of utilities as the owners of the
respective transmission systems.
A similar process was adopted for the creation of the 1200 kV transmission system. Field tests on
prototypes of the 1200 kV equipment were conducted by VEI at the manufacturer’s test site, and
58
starting in 1973, were jointly executed by VNIIE and VEI at the 1200 kV test site built as an
extension to the 787 kV Bely Rast substation. Tests were conducted on the first 1200 kV
transmission line commissioned in 1985, and then on the second 1200 kV transmission line.
The designing of all 787 kV and 1200 kV transmission systems as well as of the entire grid was
performed by the state-owned Design and Research Institute, Energoset project, which more or
less monopolized this field of engineering activity. Overall, the monopoly led to high quality and
uniformity of projects, as well as to prompt improvements when operating experience showed the
need for a particular change.
2.8.2.2 Pilot program for the next possible UHV level

A state-funded pilot program was adopted in the USSR in the 1980s to determine the highest
possible AC transmission voltage. The rated voltage was determined as 1800 kV, with L-G
switching surge limitation of 1.45 p.u. The bundled line phase had 11 ACSR conductors 240 mm2
positioned in a circle 3 m in diameter. A number of equipment prototypes were tested, but
development halted when it became clear that such transmission would not be in demand in the
near future, and further technical progress would likely make today’s technical solutions
obsolete.
2.8.2.3 Operating experience

The USSR runs an annual operational analysis on the entire grid under the auspices of the
Ministry of Electric Power and Electrification. A portion of the most significant data is presented
in Table 2.8.4. Generally, the operational experience of the 787 kV lines and the first 1200 kV
lines was satisfactory. The increased level of 1200 kV transformer failures and especially the
lengthy restoration time are clearly signs of “growing pains” for the first equipment and
transmission samples.
As the operational duration for full-voltage 1200 kV transmission was relatively short, not all
parameters in Table 2.8.4 are determined; they are marked with notes. Further long-term
experience in operating 1200 kV lines at 550 kV made it possible to identify some mechanical
problems in the lines.
Table 2.8.4 Operational data for Russia’s 787-1200 kV transmission systems
Equipment Parameter 787 kV 1200 kV
Line Failure rate, per 100 km per year 0.2 0.12
Single –pole failures, % 98 100*
Successful single-pole reclosures, % 52 **
Total successful single- and triple-pole reclosure, % 52 **
Average restoration time, in hours, if reclosing was 5.8 10.5
unsuccessful
Transformer Outages, per phase per year 0.107 0.13
Average restoration time, hours 112 5,090x
Shunt Outages per phase per year 0.068 0.25x
reactor Average restoration time 198 23x
Notes: * Insufficient observation time; ** Devices for automatic reclosure and schemes of 4-legged
x
reactors were not properly established during the relatively short period of tests and operation; Initial
period of “growing pains”
Extinction of secondary arcs at single-pole auto-reclosing application: In very long arcs

typical of EHV/UHV insulation, non-uniformity along the arc length and internal exchange of
energy between sections of the arc channel play a more important role in secondary arc extinction
59
compared to elongation, creating intermittent arcing characteristics. [22] It was often possible to
observe several sequential unsuccessful attempts to extinguish the arc followed by re-strikes
along the arc that would not be possible to extinguish in a uniform elongating arc. In this
situation, it was decided to estimate the required extinction time using exclusivity tests and
failure recording data for 330–750 kV lines. The total number of test points approached 720.
Strong wind reduces the extinction time, but this factor was impossible to control in field tests,
which increased the scattering of test data.
All extinction times were statistically analysed as a function of the crest value of a steady-state
secondary arc current. Initially, the data was separated into two groups: lines without shunt
reactors and lines with shunt reactors (including 4-legged reactor schemes). Lines without shunt
reactors provided a small recovery voltage (0.1–0.2 of L-G voltage), but with a short time to
peak. Lines with shunt reactors and 4-legged reactor schemes showed a longer voltage recovery
time, and the peak value ranged from small to very large (in schemes close to resonant conditions
up to 1.5 of L-G voltage). For this reason, the lines were subdivided into two subgroups: lines
with recovery voltage below 0.5 L-G voltage, and those above. Statistical analysis showed,
however, that the influence of the steady-state magnitude of secondary arc current is statistically
much stronger than the influence of the presence or absence of shunt reactors. Therefore, all data
was presented as a function of only the peak value of steady-state current. For 90% reliability in
extinguishing the secondary arc current, the following extinction times were established for EHV
lines:
• At 40 Apeak - 0.7 s,
• At 60 Apeak – 1.2 s,
• At 90 Apeak – 2.6 s.
The single-phase rapid auto-reclosing (SPAR) dead time can be determined as the time needed
for extinguishing the secondary arc plus approximately 0.5 s for dissipation of the initial power
arc cloud and attenuation of initial transients in the secondary current. In addition, for dynamic
stability of Russia’s interconnected grid, the maximum SPAR dead time usually had to be limited
to 3.0-3.5 s. For these reasons, the secondary current had to be limited to below 90 A in the 787
kV lines. At currents above 90 A peak, extinction becomes more problematic: it is still possible,
but the percentage of successful extinction drops.
As the insulation length is usually designed to be directly proportional to the maximum operating
voltage, and the non-uniformity in longer arcs is more pronounced, the extinction of secondary
current arcs in tests on 1000-1200 kV lines occurred more quickly and at higher currents in
comparison to that observed in the operation of 330-787 kV lines. [23]
2.8.2.4 Radio and audio interference, corona losses

Designing the 787 kV and 1200 kV lines was preceded by measuring corona losses and TV, radio
and audible noises on experimental short prototype lines and then was followed by field
measurements on lines in operation. [24], [25], [26] The maximum gradient on the conductor
surface in 1150 kV bundles must not exceed 90% of the starting gradient of total corona, thus
providing some margin for a possible increase over the rated voltage, lower position of the
conductors in the midspan and reduced air density at altitudes up to 500 m. Radio noise 100 m
from the line was prescribed at levels of 48 to 30 dB for frequencies, respectively, from 0.15
MHz up to 1000 MHz. Acoustic noise at a distance of 300 m must not exceed 55 dB (A) during
daytime and 45 dB (A) at night.
60
On bundled phases of 787 and especially 1200 kV lines, extremely high corona loss is possible in
bad weather. As an effective measure for reducing corona loss, a temporary reduction of
operating voltage was recommended. To assist interconnected grid dispatchers, a system for
monitoring on-line corona loss was developed for trial operation on several 787 kV lines.
Taking into account strong electrical fields at UHV substations and near UHV lines, standards
were developed for electric field exposure of both utility personnel and general population:
• Limited duration of unprotected personnel staying in electric substations depending on the
field gradient;
• Limited permissible gradients for the general population depending on the land
designation (city/town, agricultural use, non-populated areas);
2.9 TEPCO 1100 kV project
2.9.1 Features of TEPCO’s UHV transmission system

Tokyo Electric Power Company (TEPCO) has been working to expand their 550 kV network
since the mid-1970s, but it is difficult to obtain multiple power transmission routes in Japan.
Countermeasures against the increase in short-circuit current due to network expansion were also
required. To cope with these problems, TEPCO decided to introduce UHV (1100 kV)
transmission system with a capacity of 3 to 4 times that of the 550 kV network.
By 1999, TEPCO had already constructed UHV-designed double-circuit transmission lines that
ran 240 km from east to west and l90 km from north to south, totally 430 km, as shown in
Fig. 2.9.1. These lines are now operated at 550 kV and they will be upgraded to UHV in the mid-
2010s.
UHV systems are required to deliver very large power transmission (approximately 13GW/route)
and to ensure high reliability. Additionally, various new technologies had been developed for
substations. To establish carefully these technologies towards UHV upgrading, field-testing of
substation equipment has been carried out since 1996 connecting to the actual grid. [27]-[30]
Tokyo
Fig.2.9.1 TEPCO trunk transmission systems
61
2.9.1.1 Considerations for UHV systems and equipment
The UHV (1100 kV) system has a number of distinctive features such as large line-charging
MVA (nearly 4 times that of the 550 kV system), low transmission loss due to the adoption of 8-
bundle conductors with 810 mm2 ACSR (partially 8-bundle conductors with 610 mm2 ACSR in
mountainous areas) for transmission lines to suppress corona noise (radio interference and
audible noise).
Secondary arc at a line fault lasts longer due to electrostatic induction from other healthy phases.
To quickly extinguish the secondary arc and successfully secure high-speed multi-phase reclosing
within one second, the high-speed grounding switch (HSGS) is introduced. AC temporary
overvoltage (ACTOV) tends to be higher due to the Ferranti effect and load rejection. Attention
must be paid to these tendencies.
Low transmission loss also affects the performance of substation equipment. The DC time
constant of a fault current slowly decays and goes up to nearly 150 ms. Zero-miss current may
occur depending on the load conditions and fault timing. Attenuation of high-frequency
component in a fault current becomes slow, which may increase di/dt at around current zero.
These phenomena have a significant effect on the circuit breakers. Magnetic saturation of CT
applied to busbar protection requires an air-core CT and/or some type of measures in the
protection relay system.
As the system voltage increases, surge voltages induced in the secondary circuit also tend to
increase in severity. Optical potential devices (PD), which are hardly affected by surge voltages,
are adopted.
2.9.1.2 Concept of insulation coordination and specifications

Double-circuit vertical line configuration is adopted for the transmission lines. Suppressing
switching overvoltages as much as possible is important for making the transmission lines
compact, which leads to a reduction of construction costs as well as environmental impact.
For substations, full GIS is adopted as the main circuit from the transmission line entrance to the
transformer terminal taking into account space constraints, earthquake, pollution, maintenance,
etc. The determination of rational LIWV is important, because lightning overvoltages dominate
the non-self-restoring internal insulation design of GIS and transformers.
Designs for insulation coordination throughout transmission lines and substations were carefully
studied based on the development of a high-performance metal oxide surge arrester (MOSA) with
the protection level of 1620 kV (at 20 kA) by precise computer-aided calculations.
To suppress lightning overvoltages effectively, MOSAs are installed at adequate locations of
substation circuit. With respect to very fast front overvoltages (VFTOs), these levels are
suppressed to be 1.3 p.u. or below with application of resistor fitted disconnectors with a 500
ohm resistor. This scheme is also effective to suppress electromagnetic interference in the
secondary circuit of CT, VT, and protection/control system.
Switching overvoltages are suppressed to the same level as that of ground fault overvoltages (1.6
p.u. in the north-south route and 1.7 p.u. in the east-west route) with MOSAs installed at both
ends of the transmission lines and circuit breakers with closing and opening resistors. The
required resistance value is 700 ohm for closing, and 1000 ohm for opening. Thus, 700 ohm was
chosen for common use for both operations.
62
High performance metal oxide surge arrester
Reliable CB with closing/opening Resister

Ω , common use for both operation)
(700Ω
LIWV (Substation) Switching Overvoltage TOV

(Transmission line)
1.5pu
・Transformer: 1,950kV
1.6 -1.7pu
・ GIS : 2,250kV
TRV of CB
Fig.2.9.2 Insulation coordination based on high performance surge arrestors
The suppression of switching overvoltages to 1.6-1.7 p.u. from the conventional level, (which is
2.0 p.u. in 550 kV system), reduces the distance between the line and the tower to be approximate
6 m from 9 m and above. This insulation design reduces the tower height, which would be 143 m
in accordance with the 550 kV technologies, to 110 m as shown in Fig. 2.9.3.
Extrapolating the application of
550 kV technology
Transmission line constructed by

the reduced insulation design
Fig. 2.9.3 Compact design of UHV Tower
Table 2.9.1 Insulation levels of substation equipments

Highest voltages 1100kV
Item Transformer GIS
Phase to earth: 2250 kV
LIWV (kV) 1950 kV
Across terminal: 2250 kV + 1100x ( 2 / 3 ) kV (For CB and DS)
SIWV (kV) 1425 kV* 1550 kV*
PFWV (kV rms) 1.5E (30 min) – 3 E (1 min) - 1.5E (30 min)
* Transformer SIWV was specified to check the internal oscillation caused by slow front overvoltages
although LIWV is a decisive factor for the design. GIS SIWV is for the gas bushings at the line entrance.
63
2.9.2 Specific issues and technical challenges for substation
equipment - Specifications for substation equipment
Various UHV substation equipment was developed for field testing. Tables 2.9.2 and 3 show the
general specifications and main characteristics of this equipment.
• The transformer is a forced-oil circulating, single-phase, autotransformer (3000 MVA
bank). Single-phase transformers are transported in two units and assembled on-site due
to transportation constraints. The divided unit was designed to meet the same capacity as
that of a typical 500 kV transformer (1500 MVA bank), while it must withstand twice the
voltage. Insulation technologies to ensure high reliability under long-term operation were
developed. On-site assembly and connection technologies were also developed.
• MOSA plays a key role in insulation coordination for the UHV system. In order to
improve the protection characteristics and heavy discharge duties, 4-MO columns are
electrically connected in parallel, the elements of which were developed for UHV
arresters and are well proven in the field as 550 kV high-performance arresters. During
development, focusing on mainly severe ACTOV duty of 55 MJ specified in
consideration of load rejection, various tests were performed, such as energy injection
destructive tests for many of the elements to ensure quality stability and current
uniformity tests between each MO column, etc.
• 1100 kV / 50 kA two-break GCB with resistor closing and opening scheme was
developed. The main and resistor contacts are operated separately by their own hydraulic
mechanism, because the resistor contact precedes the main contact by 10 ms in closing
and lags by 30 ms in opening. The delay time of 30 ms was specified considering
interruption failure by multiple lightning discharges and possible missing current zero.
Reliability of the resistor closing and opening scheme was confirmed by the following aspects
and has been verified by long-term field testing.
1. 550 kV CB with resistor closing scheme has been standardized and is well proven in field
operation in Japan.
2. Refined and proven hydraulic operating mechanisms are applied to both the main and
resistor contacts.
3. Two operating mechanisms are controlled by a single valve to secure synchronization.
4. FMEA studies applied to various failure modes of the operating mechanism was carried
out. (Example 1: Hydraulic pipe broken between the main and resistor cylinders should
not be driven in a closing operation. Example 2: Orifice between main and resistor
cylinder shut by trouble shall be opened by both the main and resistor contacts.)
5. Quality control in cooperative actions between manufacturers and utilities using
appropriate standardized specifications.
• HSGS was developed based on CB technology such as a hydraulic operating mechanism
that enables high-speed switching and a puffer cylinder for the interrupter. The
specifications of HSGS include extended current interruption duration up to 80 ms
considering current zero phenomena that occurs when another ground fault (second fault)
takes place during the interruption of electromagnetic induced current.
• Resistor-fitted disconnectors are adopted to suppress VFTOs. In addition to ordinary tests
such as the capacitive current interruption test, multiple discharge tests by simulating
64
restriking surges were performed to ensure that resistor withstand multiple discharges and
restriking currents commutate to the resistor without bypassing.
• Gas bushings at the entrance of substations employ porcelain housing to ensure long-term
reliability. For seismic design, porcelain housing was developed by enlarging the diameter
with improved shed shapes. Contamination withstand voltage tests were carried out with
non-uniform contamination along vertical and diametrical directions to confirm the
performance of large bushings based on data from exposure tests in the field, where a
typhoon occurred. Switching impulse withstand voltage tests were performed not only
under dry conditions but also wet conditions considering the tendency towards lower
performance under wet conditions.
• Optical PD is adopted considering equipment compactness, cost reduction, surge-proof
characteristics, etc. It was developed focusing on the accuracy of voltage control
throughout the entire network.
Table 2.9.2 Specifications and main characteristics of UHV transformer

Primary 1050/ 3 kV
Rated voltage Secondary 525/ 3 kV
Tertiary 147 kV
Rated capacity 3000 MVA/3
Tertiary capacity 1200 MVA/3
Primary tapping range ± 7% (27 taps)
Impedance 18 %
LIWV Primary 1950 kV
Test voltage Primary and
AC 1.5E (1 hour) - 3 E (5 min) - 1.5E (1 hour)
Secondary
Surge capacitance 10,000pF (Design characteristic)
Table 2.9.3 Specifications and main characteristics of 1100kV GIS
Rated continuous current 8000A (Line and Bus)
GIS
Rated short-time withstand current 50kA, 2 s
(Common)
Test voltage LIWV 2250 kV
AC 1.5E (30 min) – 3 E (1 min) - 1.5E (30 min)
Breaking current 50 kA
Making/Breaking resistor 700 Ω (used for both making and breaking)
GCB
Standard operating duty O- θ -CO-1 min-CO
Thermal duty of resistor BTF ’O’ + Out-of-phase ’CO’
First-pole-to-clear Factor 1.1
GCB
Amplitude Factor 1.4 (with opening resistor and MOSA)
(System
characteristic)
DC time constant 150ms
Surge impedance for SLF 450 ohm
Surge suppression resistor 500 Ω
Thermal duty: CO (Continuous one operation)
Small capacitive current switching Recovery voltage: 635 kVrms (= 1,100/ 3 )
Disconnector
Load side capacitance: 2000 pF
Loop current switching Recovery voltage: 300 Vrms
Loop Current: 8000 A
Electromagnetic induction current Interrupting current: 5800Arms
HSGS
interruption Recovery voltage: 640kVp
Electrostatic induction current Interrupting current: 1200Ap
interruption Recovery voltage: 900kVp
65
Standard operating duty C - 0.5s - O (Synchronized with CB operation)
Continuous operating voltage 1100/ 3 kV
ZnO surge Residual voltage 1550 kV (at 10 kA), 1620kVp at 20kA
arrester Steep wave response V(1 µ s)/V(8 µ s) at 10kA ≤ 1.1
ACTOV duty 55MJ
SIWV 1550 kV
Contaminated withstand voltage 762 kV (1.2E)
Gas bushing
3 cycle sine waves of 0.3G at the bottom of the
Seismic withstand (targeted value)
supporting structure, safety factor of 2
Switching current: 1000A
Duty of interrupting induced current Recovery voltage: 70kVrms
Earthing
Rate of rate of recover voltage; 160kV / µs
switch
Duty of interrupting electrostatic Switching current: 40A
induce current Recovery voltage: 50kVrms

TEPCO’s 1100 kV projects - Field tests
The developed UHV equipment was installed at the UHV test station (550 kV Shin-Haruna
Substation) and full-scale field tests have been carried out since May 1996. The configuration of
the UHV test station is shown in Figs. 2.9.4 and 5, and a photograph is provided in Fig. 2.9.6 The
test station has one three-phase power transformer bank and one GIS bay with the control and
protection systems. Each single-phase transformer comprises two split units each equipped with
one on-load voltage regulator, enabling current to flow through the coil by using the tap-voltage
difference between same-phase tanks. The GIS has an additional loop circuit with a power
current transformer. Test currents up to 10,000 A can be achieved by reverse excitation of the
power current transformer. Tables 2.9.4 and 5 summarize the accumulated duration of long-term
verification tests and the testing items. [31]-[41]
Exi st i ng Fi el d Test Fac i l i t i es

Shi n- Har una
Subst at i on L o adin g Test C ircu it
P o w er C T
B u sh in g
10 00 kV
Tran sform e r
w ith LV R
C ircu it
M M B rea ke r M
27
D iscon n ecting D iscon n ectin g

C ircuit
Sw itch S w itch
B reak er
Su rg e O ptical H igh Speed
A rrester M
PD G rou nding
Sw itch
5 00 kV 1 00 0k V
Fig. 2.9.4 Configuration of UHV Field test Fig. 2.9.5 Single line diagram of UHV field test site
2.9.4 Summary
To secure high reliability and economy of transmission lines and substation, the specifications
for UHV equipment were determined based on detailed network analyses. Additionally, the
developed UHV equipment including the three single-phase transformers, MOSA, CB, HSGS,
disconnectors and gas bushings, etc, have been verified at the UHV test station since May 1996
and their long-term reliability and performance have been confirmed.
66
Table 2.9.4 Operation records (August 2007)
Item Hours
Cumulated operation time of 60456
UHV Test Site
Current carrying time of GIS 9995
Current carrying time of 7918
transformer by using the tap-
voltage difference
Fig. 2.9.6 TEPCO testing station

Table 2.9.5 Field test items
Equipments Testing items
Transformer Common Long-term energizing and current carrying
Periodical Dissolved gas analysis, insulation oil characteristics
inspection measurement
Leakage current measurement at neutral terminal
Noise, vibration measurement
Inrush test
GIS Common Long-term energizing / loop current carrying, temperature and
thermal behaviour measurement
Periodical Partial discharge, surge arrester leakage current
measurement Operating characteristics
Optical PD accuracy
Surge DS operating surge
measurement HSGS making surge
Transient response of grounding system
Residual DC voltage damping characteristics
Surge arrester’s temperature and thermal behaviour
Gas bushing vibration test
2.10 China’s project
2.10.1 Brief summary of China’s 1100 kV projects

China’s UHV grid will be used to interconnect the large coal-fired thermal power plants and
hydro power plants in the northern and western parts of China with the load centres in the east
and central parts of China. Around 2010, the total installed capacity of synchronized power will
be about 514 GW, increasing to about 770 GW in 2020. The UHV AC transmission system will
form the backbone of the synchronized power system, replacing the weak 500 kV connections
and enhancing the transfer of bulk power between regions and improving the system stability.
[42]
67
Research and feasibility studies with the involvement of CEPRI (China Electric Power Research
Institute) have led to the optimum achievable voltage level for bulk transmission and system
performance in China. According to the economic and technological evaluation, 1100 kV AC is a
reasonable voltage level for UHV AC transmission in parallel with an 800 kV DC system. This
hybrid system offers benefits with respect to security and stability performance, as an emergency
control function will be adopted by the UHV DC system.
In system fault simulations, based on the period 2015 to 2020, the hybrid UHV AC/DC system
shows a high capacity to maintain stability under contingencies such as monopole failure (no
overloading and no voltage or frequency problems) and bipolar blocking (stable by means of
tripping some hydro plants) by specifying a 10% long-term overloading capability in the UHV
DC system.
According to China’s “Criteria of electric power system security and stability,” three coordinated
defence lines will be implemented in the UHV transmission system:
• Preventive measures against system collapse through protection and automatic control of
generators and other equipment;
• Security and stability control per area;
• Corrective measures through the coordinated tripping of generators, load shedding, grid
separation, etc.
For the UHV AC system, the fault clearing times can be identical to those applied in the 500 kV
grids. At increasing transmission capacity, thyristor-controlled series compensation will be
needed and further investigation is required to suppress sub-synchronous resonance oscillations.
Besides conventional control techniques, advanced control by means of FACTS and HVDC
should be adopted.
Real-time UHV AC/DC simulations have been performed in order to study the system dynamics.
Commutation failures in one or more converters, faults in the UHV DC lines and coordinated
control actions in the UHV DC lines have been simulated in order to study the impact on the
UHV AC system. The results of this research form the basis for implementing the first phase of
the UHV AC / UHV DC system. Protection and control systems for the UHV AC system have
been simulated and tested separately in relation to the system’s dynamic and transient behaviour,
leading to the protection policy and selected protection relays for the UHV AC system.
Studies with respect to overvoltage and insulation coordination for the UHV AC system revealed
the following possibilities:
• TOVs limited to 1.3 p.u. in substations and to 1.4 on OH-lines
• SFO phase-to-ground limited to 1.6 p.u. in substations and 1.7 on OH-lines
• SFO phase-to-phase limited to 2.6 p.u. in substations and 2.7 on OH-lines
• Proposed methods for overvoltage limitation are MOSA (rated voltage 828 kV) along
OH-lines, neutral reactor applied for shunt reactors giving SPAR time of 1 s, closing
resistor and MOSA to limit switching overvoltages.
68
2.11 India’s project
2.11.1 Features of India’s transmission projects - Power

generation and demand scenario in India
India’s power system is poised for accelerated growth. Peak demand is expected to increase to
more than 450 GW by 2025 and beyond from the present level of about 100 GW for which an
installed capacity of about 600 GW is required.
It is estimated that peak demand by 2012 would be about 157 GW for which an installed capacity
of about 210 GW is envisaged. Additional generation is envisaged at only a few locations for
which the development of a strong transmission network would be required to connect these
generation complexes to various load centres.
In order to ensure optimal utilization of dispersed energy resources, the high-capacity national
grid of 16,600 MW inter-regional capacity, comprising 400 kV AC and HVDC system is already
under operation, and shall be progressively enhanced to more than 37,700 MW by 2012 through a
high-capacity 765 kV AC and +800 kV, 6000 MW HVDC system. To meet the long-term power
transfer requirement by 2025 and beyond, large transmission networks interconnecting the
generating resources with load centres are being planned. Preliminary work on the 1200 kV
UHVAC system has already started and various alternatives are being studied for the selection of
the UHVAC corridor.
NR
NER
ER
WR
SR
LA
ANDAMA
NICOBA
KS
HA
DW
N&
R
EE
P
Fig. 2.11.1 Region of India’s power system
2.11.1.1 Present transmission network of India

The present transmission network in India is a hybrid system comprising an EHVAC system up
to 765 kV and an HVDC long-distance bipole and back-to-back system. The transmission
network consists of about 1700 ckt.km of 765 kV lines, 190,000 ckt.km of 400 kV/220 kV lines,
and about 5870 ckt.km of HVDC lines. Four (4) HVDC back-to-back systems (capacity 500-
1000 MW) are presently in operation. In addition, fixed series compensation, thyristor-controlled
69
series capacitors (TCSC) and static var compensators (SVC) are provided on various 220/400 kV
transmission systems. Other technologies such as upgrading/uprating of transmission lines, high
temperature endurance conductors, tall towers, GIS/compact substations, remote operation &
substation automation, GIS/GPS-based line route survey techniques, etc. are also integrated in the
system.
2.11.1.2 Future transmission network of India (with time frame)

By 2012, in addition to the existing network, India’s system shall comprise a strong 765 kV and
400 kV network along with +500 kV, 2500 MW and +800 kV, 6000 MW long-distance HVDC
links. The addition of about 9000 ckt.km of 765 kV AC lines and fifteen 765 kV substations are
envisaged apart from a large number of 400 kV substations and transmission lines. Preliminary
work on the 1200 kV AC system has already started and the first 1200 kV link is expected to be
operational by 2012-13. The adoption of transmission voltage level in India’s power system is
shown in Fig. 2.11.2 below.
±800 kV HVDC 1200kV UHVAC
765kV AC
±500 kV HVDC
400 kV
1977 1990 2000 2011 2012/13

Fig. 2.11.2 Adoption of transmission voltage level in India’s power system
2.11.2 Consideration for transmission development

Additional generation is envisaged at only a few locations in eastern and northeastern regions of
India. This necessitates the transfer of large amounts of power across the region. Considering the
demand and availability scenario of each region, it is estimated that about 90 GW of power is to
be imported from NER/ER to NR/WR/SR (See the regions in fig.2.11.1). For this, the
development of a large number of transmission corridors is required. However, the development
of a large transmission network for transferring bulk power across the regions poses a number of
challenges, as described below:
Rights-of-way (ROW);
Environmental considerations;
Optimization of cost;
Faster project implementation;
Coordinated development of transmission corridor together with other
infrastructure;
Transmission capacity enhancement through upgrading;
Reduction of loss;
Integration of emerging technologies.
70
Keeping the above in view, a high-capacity transmission system interconnecting the eastern,
western, northern and northeastern regions through a high-capacity +800 kV, 6000 MW HVDC
system and 765 kV ring has already been planned. However, taking into consideration the
increased power transfer requirement to meet the above challenges, an alternative 1200 kV AC
transmission system between various regions is also being explored. The 1200 kV AC technology
would yield benefits in terms of fewer ROW corridors as well as savings in transmission line cost
compared to using a 765 kV transmission system to transfer a similar amount of power.
2.11.3 Electrical design considerations for transmission lines

Preliminary design of the 1200 kV transmission lines has been carried out. Studies included:
a. Clearances and tower configuration analysis;
b. Thermal and surge impedance loading;
c. Conductor surface gradient and corona onset gradients;
d. Insulation/insulator design;
e. Electric field at ground;
f. Interference studies (RIV, AN);
Various bundle configurations were analysed and Octagonal ACSR conductor configurations
named Bersimis, Moose and Lapwing were selected for further study. For finalization of bundle
configuration, corona cage studies have been undertaken.
The phase conductor shall be placed in delta configuration. The insulator shall be in I-V-I
configuration. Power frequency air gap clearance of 2.3 m under 55º swing conditions and
switching surge clearance of 10 m under stationary conditions are being considered. (Tests for air
gap clearances are in progress.)
Insulator string length of 10-11 m is envisaged comprising 55×320 kN or 50×420 kN porcelain
disc insulators providing 21,600 mm creepage distance at 18 mm/kV
Phase-to-phase conductor clearance of 27.5-30 m and ground clearance of 20-23 m are being
considered. The electrical field below the conductor shall not exceed 10-12 kV/m. Some
electrical parameters are given in the following tables.
Table 2.11.1 (a) Corona measurement (1)
S.No. Bundle conductor Conductor surface gradient kV/cm Corona onset gradient KV/cm
Mean Max Fair weather Bad weather
1 Octa Moose 14.34 17.01 21.1 14.89
2 Octa Bersimis 13.02 15. 7 20.89 14.75
3 Octa Lapwing 12.0 14.67 27.73 14.64
Table 2.11.1 (b) Corona measurement (2)
S.No. Bundle conductor RIV (dB) AN(dBA) L5 level AN(dBA) L50 level
Max At 50 M ROW Max At 50 M ROW Max At 50 M ROW
1 Octa Moose 39.1 31.6 61.3 58.6 57.6 54.9
2 Octa Bersimis 36.5 29.1 60.3 57.6 55.9 53.2
3 Octa Lapwing 34.8 27.4 59.4 56.7 54.4 51.6
2 2 2
ACSR conductor size: 528.5 mm for Moose, 689.7mm for Berimis and 807.5 mm for Lapwing
71
2.11.4 Preliminary study results for 1200 kV UHV AC systems
Preliminary studies are being carried out on the 1200 kV AC transmission system to limit
switching and temporary overvoltages under different network conditions and varied source
strength and line length.
For study purposes, 60% line shunt compensation is being considered. In addition, 40% reactive
compensation is provided on the sending as well as receiving end buses. Studies are also being
carried out to observe the impact on reactive power flow on the line in case of an increase in
active power flow with and without line reactive compensation. Study results are shown in Fig.
2.11.3.
3000
2000
Recative Flow ( MVAR)
Without compensation
1000
0
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 7500 8000
-1000 Power Flow (MW)
-2000
With 60% compensation
-3000
Fig.2.11.3 impact on Reactive power flow
2.11.5 Overvoltage studies

Deterministic studies are being carried out to find the optimal pre-insertion resistor (PIR) value to
control switching overvoltages within the stipulated limits. The results of preliminary studies
with and without PIR are listed below.
Table 2.11.3 Switching Over Voltages (SOV)
Switching overvoltage (P.U)
Case
W/o PIR 300Ω/ 10 ms 600Ω/ 10 ms 700Ω/ 10 ms
Source Short circuit-15,000
2.17 1.36 1.55 1.58
MVA; Line length-350 km
Source Short circuit-15,000
2.22 1.45 1.62 1.69
MVA; Line length-450 km
From the above results, it was observed that the optimal size for a PIR could be in the range of
600–700 ohm. Further statistical studies are also being carried out to ascertain the frequency of
occurrence of maximum voltage encountered during SOV studies with the above PIR values, the
results of which are shown below in Figure 2.11.4:
72
700 ohm PIR
%Freq Occurecne
50
40
30
20
10
0
1.35 1.4 1.45 1.5
Voltage (p.u)
Fig. 2.11.4 Switching Over Voltages in case of 700 ohm
600 ohm PIR

% Freq Occurcen
50
40
30
20
10
0
1.3 1.35 1.4 1.45
Voltage (P.u)
Fig. 2.11.5 Switching Over Voltages in case of 600 ohm

Keeping the above in view, the following protection level was selected for the 1200 kV AC
system:
• Switching impulse withstand level (SIWV) (Switchgear/Bushings/Transformer windings)
- 1800 kV
• Lightning impulse withstand level (LIWV) (Switchgear/Bushings) - 2400 kV
• Lightning impulse withstand level (LIWV) (Transformer windings) - 2250 kV
• Transformer rating 3×1000 MVA (Three single-phase units) with 18% impedance.
• Short-circuit levels - 50 kA
Temporary overvoltage studies are also being carried out to observe the level of overvoltage
encountered by the system in the event of sudden load throw-off, etc. The results of TOV studies
with different line lengths and without/with reactive compensation are listed in Table 2.11.4:
Table 2.11.4 TOV with Power flow of 3200 MW at 0.92 pf., 15000 MVA Source strength
W/O Line reactor With Line Reactor
340 km long line 1.48 p.u. 1.31 p.u. (500 MVAR LR)
400 km long Line 1.60 p.u. 1.34 p.u. (660 MVAR LR)
73
In addition to the above, studies are also being carried out to determine the appropriate surge
arrester rating for the optimal insulation level of electrical equipment. Based on the study results,
the following surge arrester is recommended:
Table 2.11.4 Surge Arrester Parameters

SA Class 5
Rated Voltage 826 kVrms
Continuous Operating Voltage(COV) 702 kVrms
Nominal discharge current 20 kA
Lightning Impulse residual voltage 1675 kVp at 10 kA
1790 kVp at 20 kA
Energy level 30 MJ
Typical V-I characteristics of the surge arrester are considered：1460 kV at 1 kA, 1520 kV at 2
kA, 1575 kV at 5 kA, 1675 kV at 10 kA and 1790 kV at 20 kA.
2.12 Summary
Chapter 2 reviewed the state-of-the-art of project and national technical specifications for
substation equipment at voltages of 765kV and above. The rapid growth in electrical power
demand especially in China, Brazil and India is pushing the development and construction of new
UHV transmission systems. The transmission capacity of UHV transmission systems with two
routes has reached about 10GW, which is three times larger than that of 550kV transmission
systems. The reviews identified the areas of commonality and of divergence that will be
discussed further in chapter 3. The operations of 800kV transmissions provided several valuable
field experiences that help new technical challenges for UHV transmissions.
The pioneering investigations of corona noises provided the fundamental designs of the multi-
bundle conductors for UHV transmission lines. The use of the multi-bundle conductors with large
diameters can reduce the line surge impedances but increase the time constant of the DC
component in a fault current. Furthermore, the use of large capacity power transformers reduces
the first-pole-to-clear factor due to the small zero-sequence impedance of the system. Low losses
of power transformers and transmission lines increase the amplitude factor of the TRV. Surge
arresters also play an important role in reducing the TRV values.
The various national technical specifications described in chapter 2, are compared and evaluated
in chapter 3 carefully considering the technical maturities, economic and reliability aspects. Then,
the best practises for UHV specifications are recommended along with some comprehensive
explanations of their technical background.
74
References
[1] T. J. Nagel et al., “Development of the American Electric Power System Transmission
Network: From 345kV to 765kV to UHV”, CIGRE Session 32-13, 1974
[2] L. B. Wagenaar et al., “EHV transformer Dielectric Specification Improvement”, IEEE
Transactions on Power Delivery, Vol.9, No.1, January 1994
[3] A. J. F. Keri et al., “The use of Reactor Switches on Single Phase Switching”, CIGRE
Session 13-06, 1980
[4] H. N. Scherer et al., “AEP-ASEA UHV Research Project-Results to 1983”, CIGRE Session
38-04, 1984
[5] CIGRE WG 13.04 “Specified Time Constants for Testing Asymmetric Current Capability
of Switchgear”, ELECTRA No.173, pp.19-31, 1997
[6] Khodabachian et al., “TRV and the non-zero phenomenon in Hydro-Québec’s projected
series-compensated system”, CIGRE 13-303, Session 1992,
[7] A. Déry, J.-P. Gingras, "Hydro-Québec TransÉnergie De-icing project at Lévis Substation,"
presented at the 11th Int. Workshop on Atmospheric Icing of Structures, Montreal, Canada,
June 2005.
[8] A. Mercier et al. “Transformer Controlled Switching Takinng into account the core residual
flux, a real case study” CIGRE 13-201, Session 2002
[9] J. Redlund et al., “A new fast protective device for high voltage series capacitors”, IEEE
No 1-4244-0493, 2006
[10] S. D. Baik et al., “Introduction of KEPCO’s 765kV AC Transmission System”,
International Conf. of UHV Power Transmission Technology, pp.24-31, 2006
[11] Motta and H. A. M. Fadini “Itaipu Transmission System – Consolidation of FURNAS
Experience” - Book published only in portuguese under the coordination of S. S. G. ISBN
85-85996-01-3 - Rio de Janeiro, 1995
[12] “Achieving its Potential in Electrical, Electronic and Multimedia Technologies” IEC Latin
America Regional Centre Seminar 2007 (IEC-LARC) - May 2nd and 3rd, São Paulo, Brazil
[13] ‘Regulatory Acts’ Library of the Brazilian Regulatory Agency for electric energy
[14] Italian Panel on UHV “Open aspects and possible alternative technologies following the
UHV 1000 kV Italian experience”, IEC/CIGRE UHV Symposium 2007, paper 1-1-2
[15] Italian Panel on UHV “Open Testing challenges for UHV equipment”, IEC/CIGRE UHV
Symposium Beijing 2007, paper 4-1
[16] Annestrand, S. A., Parks G. A., “Bonneville Power Administration Prototype 1100/1200
kV Transmission Line Project,” IEEE Transactions on Power Apparatus and Systems, Vol
96, March/April 1977, pp. 357-366
[17] Klinger, et al., “Experience of a 6 year 1200 kV Transmission Line Test and Development
Program at the Bonneville Power Administration,” CIGRE Conference on Large High
Voltage Electric Systems, Aug. 29- Sept. 6, 1984, Paper 38-08
[18] Perry, D. E., et al., "BPA 1100-kV Transmission System Development Corona and Electric
Field Studies," IEEE Transactions on Power Apparatus and Systems, Vol 98, Sept/Oct.
1979, pp. 1728-1730).
[19] Stearns, R. D., and T. D. Bracken. "Corona and Electric Field Performance of the BPA
Prototype 1200 kV Transmission Line," (CIGRE Symposium 22-81, 1981rPaper 232-06)
[20] Nellis, C. L., et al., “BPA 1100 kV Transmission System Development Insulation Studies,”
IEEE Winter Power Meeting, Feb 9, 1979
[21] V. Kovalev et al, “Prospects for Creation of New Generation of Ultra-High Voltage
Equipment”, IEC-CIGRE UHV Symposium, Beijing, 18-21 July 2007, Paper 2-4-2
75
[22] N.N. Belyakov, et. al., “Results of internal overvoltages and electrical equipment
characteristics measurements in the Soviet 750 kV networks”, CIGRE Session 1978, 33-08
[23] N.N. Belyakov, et. al., “Application of single-phase auto-reclosing in a complex EHV
network containing 1200 kV transmission lines”, CIGRE Session 1990, Paper 34-207
[24] N.P. Emelyanov, et al., “Investigation of corona effects on EHV and UHV transmission
lines”, CIGRE Session 1984, Paper 36-11
[25] L.V. Timashova, Ecological Influence and Electromagnetic Compatibility for Overhead
Electric Transmission Lines, VNIIE Herald (Proceedings), 2004, pp.42-63, in Russian
[26] Methodology of calculating corona losses is described in the book A. Tamazov, Corona on
Conductors of AC Overhead Lines, Sputnik Co Publishing House, Moscow, 2002, 316 pp.,
in Russian
[27] T.Watanabe et al., “Required duty for 1000kV circuit breaker (phase 1)(UHV-1)”,1993
JEEA, Power and Energy Meeting, No.54.
[28] Y.Yamagata et al., “Required duty for 1000kV circuit breaker (phase 2)(UHV-2)”,1993
JEEA, Power and Energy Meeting, No.55.
[29] Y.Yamagata et al.,”Capacitive current switching test method for 1000kV GCB”,1994 JEEA,
Japan Review, No.1455.
[30] Y.Yamagata et al.,”Field test of 1000kV gas insulated switchgear,” CIGRE 2000 Paris
Session SC13-209
[31] T.Tanabe et al., “Interrupting Duties and Test Specifications of 1,100kV Gas Circuit
Breaker and Its Development for Future UHV Power Transmission System in Japan”,
Colloquium of CIGRE SC13, Report 1.2, 1995 Florianopolis.
[32] Y.Yamagata et al., “Development of 1,100kV GIS-Gas Circuit Breakers, Disconnectors
and High-speed Grounding Switches”, CIGRE 1996,13-304
[33] A.Nakamura et al., “1100kV AC Transmission Project in Japan “, IEC/CIGRE UHV
Symposium Beijing 2007, 1-2-1
[34] E.Zaima et al., “System Aspects of 1100kV AC Transmission technologies in Japan:
Solutions for Problems Specific to UHV AC Transmission System and Insulation
Coordination “, IEC/CIGRE UHV Symposium Beijing 2007, 2-1-1
[35] M.Isozaki et al., “Special design and arrangements of UHV AC Conductors and
subconductors taking reduction of environmental impact into consideration“, IEC/CIGRE
UHV Symposium Beijing 2007, 2-2-4
[36] G.Sun et al., “Suitable Configuration of Switchgear for UHV AC substation“, IEC/CIGRE
UHV Symposium Beijing 2007, 2-3-2
[37] T.Kobayashi et al., “Basic design / specifications of GIS for UHV AC and its verification
test at site“, IEC/CIGRE UHV Symposium Beijing 2007, 2-3-3
[38] N.Qiu et al., “Environmental Considerations for UHV substations“, IEC/CIGRE UHV
Symposium Beijing 2007, 2-3-5
[39] Y.Yamagata et al., “Development of 1100kV Gas Circuit Breakers – Background,
Specifications, and Duties - “, IEC/CIGRE UHV Symposium Beijing 2007, 2-4-3
[40] K.Hidaka et al., “Development of 1100kV AC GIS-arrester and the verification Tests“,
IEC/CIGRE UHV Symposium Beijing 2007, 2-5-2
[41] T.Kawamura et al., “Development and long term field tests for UHV, 3000MVA
Transformer in Japan“, IEC/CIGRE UHV Symposium Beijing 2007, 2-6-5
[42] Yin Yong-hua, et al., Research and application of UHVAC transmission technology,
IEC/CIGRE UHV Symposium Beijing 2007
76
3 Specific topics
3.1 Insulation coordination
With higher system voltages, especially UHV exceeding 800 kV, the technical and economic
consequences of the insulation levels become increasingly important. Experts involved in UHV
(pilot) projects therefore pay considerable attention to reducing the overvoltage to as low a level
as possible. Optimal insulation coordination is thus extremely important for UHV systems. Based
on past experience and studies from several places around the world, an overview is given of the
insulation coordination policies applied or foreseen for UHV systems.
Insulation coordination throughout transmission lines and substations is a key factor for realizing
a reliable and economical UHV system. Optimal insulation coordination can be achieved based
on high-performance MOSAs, as demonstrated in UHV projects of the 1990s and later in Japan,
Italy and China. Sophisticated design of insulation coordination by means of accurate computer-
aided calculations and simulations is common practise for such projects, while withstand voltage
can be roughly estimated by IEC’s simplified method as described further in section 3.1.6.
1800
Section Aresster
Steep Current Impulse(1/2.5µｓ)
1700 　　◇ 　　◆
Lightning Impulse(8/20µｓ)
　　○ 　 ●
Switching Impulse(30/60µｓ)
AC(50Hz)
　　△ 1620(1.80pu)
　　□
1600
Volt age(kV)
8/ 20μ s Upper Limit

1500
8/ 20μ s Median
1.56pu
1400
8/ 20μ s Lower Limit
1300
1200
0.01 0.1 1 10 100
Current（ｋA)
Fig. 3.1.1 Example of V-I characteristics of high-performance surge arrester
Suppressing switching overvoltages as much as possible is a prerequisite for air clearance to
insulation in order to reduce the height of transmission towers and the dimensions of open-air
parts in substations. As lightning overvoltages dominate the non-self-restoring internal insulation
design of equipment such as GIS and transformers, here it is important to rationalize LIWV by
means of MOSA arrangement. To effectively suppress lightning overvoltages, single or multiple
high-performance MOSAs must be installed at an adequate number of locations, such as at line
entrances, busbars and transformers. To suppress switching overvoltages, pre-insertion (closing,
closing/opening) resistors are often applied in addition to MOSAs. Other options include the
installation of MOSAs along OH lines and controlled switching, but these technologies are not
widely applied. An example of the V-I characteristics of high-performance surge arresters is
given in Fig. 3.1.1. Furthermore, VFTO caused by GIS disconnectors can be reduced to 1.3 p.u.
77
or less by means of pre-insertion resistors in GIS disconnectors.
Another important aspect of insulation coordination is the utility’s policy regarding withstand
margins for severe lightning or switching conditions with a very low probability of occurrence.
As precise computer simulations accurately show the impact of such severe conditions,
consideration might be given to omitting the additional safety factor, especially when some stress
factors are less severe than those defined in the standards. A good example is lightning discharge
currents, where the front duration tends to be longer with larger amplitude of lightning stroke
current, as shown in Fig. 3.1.2 [1].
Example of observed lightning current waveform
0
Current [kA]
PEAK : -120.0kA
-50 Tf : 6.4 mirco sec
-100 µs]
Time [µ
0 20 40 60 80 100
101
µs]
Maximum Rate of Rise [kA/µ
µs]
102
[µ
8
0
Maximum Rate of Rise
First Peak Amplitude
101
CIGRE: y=6.6x0.38
y=1.40x0.77(r=0.819) 95% confidence interval of data
95% confidence interval of data
1 10-1
1 101 102 1 101 102
Peak Amplitude [kA] First Peak Amplitude [kA]
Fig. 3.1.2 Relationship between lightning stroke current amplitude and rate-of-rise
3.1.1 Power frequency voltage

Table 3.1.1 indicates that the maximum operating voltage does not differ greatly in the UHV
pilots and projects, apart from one or two exceptions [2]. IEC doc 1241/8/CDV proposes 1100
kV and 1200 kV (under consideration) as highest voltage for equipment.
Table 3.1.1 Highest voltage of UHV projects
Project Italy Russia Japan China
Um (kV) 1050 1200 1100 1100
The exceptions are related to pilot projects that have since been discontinued and that included
maximum operating voltages of 1600 kV (USA) and 1800 kV (Russia). The values given in
Table 3.1.1 are used as a reference level for that specific country. For power frequency voltages,
1 p.u. = Um / 3 ; and for transient overvoltages and transient recovery voltages (TRV), 1
p.u.= Um 2 / 3 .
78
3.1.2 Temporary overvoltage (TOV)
TOV is an important power frequency voltage level, which expressed in p.u. must be as low as
possible. TOV in large systems is mainly caused by Ferranti effect, (ferro)-resonance phenomena
and single-phase faults. One special resonance phenomenon in relation to shunt reactors, is that
under SPAR (single-phase rapid auto-reclosing), it might be energized by capacitive coupling
from the healthy phases. [3] If the inductance of shunt reactors connected to the faulted line phase
and the line capacitance are very close to parallel resonance at power frequency, the recovery
voltage on this phase may be limited by corona loss (depending on the corona onset voltage).
Calculations of the voltage level on Russia’s 750 kV lines, confirmed by field tests, showed that
steady-state voltage in this case reached 1.4-1.5 p.u. In a 420 kV case [3] with a low visible
corona onset voltage of 1.35 p.u., the TOV level is limited to 1.6 p.u. In the presence of a neutral
reactor, this voltage is usually less, due to detuning from resonance conditions.
Another special resonance phenomenon is the energization of large power transformers or shunt
reactors in a weak network, as may occur during the initial stage of UHV network or during
system restoration. TOV up to 1.45 and 1.55 has been reported, line to ground and lasting for
seconds. [4],[5] Ferro-resonance overvoltages on second harmonics were observed during one-
side switching of 750 kV lines even from powerful feed networks where they were generated by
nonlinearity of shunt reactor and/or transformer magnetization curves. These TOV could be
especially dangerous for EHV/UHV equipment if protective overvoltage relays operate on
effective, not peak, voltage values and are not able to detect the TOV.
As suitable countermeasures are available, it is assumed that resonance and ferro-resonance
phenomena would be anticipated and prevented or at least limited in magnitude and/or duration.
The two remaining causes of TOV are single-phase faults and Ferranti effect.
In the case of single-phase faults, the relationship between TOV and kpp (first-pole-to-clear
factor) results in an overvoltage of 1.27 p.u. for kpp = 1.3, of 1.16 p.u. for kpp = 1.2 and of 1.06
p.u. for kpp = 1.1. Only in cases of relatively high kpp, the TOV reaches a certain importance.
Single-phase fault clearing will take several tens to several hundreds of ms. The relationship
between the healthy-phase power frequency voltage and the X0/X1 ratio can be seen in Fig. 3.1.3.
When the impedance is indicated in p.u., the impedance of OH lines is nearly in reverse
proportional to the square of the system voltage. In UHV systems, the ratio of line impedance to
total system impedance, seen from the fault location, becomes smaller and, thus, the earth-fault
factor tends to be smaller.
The Ferranti effect is normally controlled by shunt reactors (and, when applied, series capacitor
banks) that are switched on before the OH line is energized and switched off at high loads or,
sometimes, after switching off the OH line. In the case of sudden load rejection at a heavily
loaded long line, the voltage jump due to Ferranti effect will lead to a TOV, which lasts until
either the OH line is tripped at both sides or the shunt reactors are switched on. The amplitude of
the TOV under such circumstances is system dependent, but computer simulations show a typical
maximum TOV of 1.4 to 1.5 p.u. lasting < 0.5 s. This applies to studies performed in Italy,
Russia, Japan and China. [2],[6],[7]
79
Fig. 3.1.3 General tendency of earth-fault factor in Japan
For TOV control during load rejection, it is vital to apply automatic high-speed insertion of shunt
reactors and transfer-tripping to the other side of the OH line. In some countries, for extreme
contingencies that lead to higher TOV than stated above, MOSAs are used to limit the TOV to
1.7-2.0 p.u. [8] At the 800 kV level, in Canada, sacrificial switchable MOSAs are used to control
the TOV to 1.6 p.u., as may be necessary under severe system conditions. Normally, however,
surge arresters are not used to limit TOV, as it would be less than 1.4 p.u. A contingency may be
a case in which a phase-to-earth fault occurs in combination with load rejection, reaching TOV
values higher than stated above for the occurrence of single-phase faults. In Japan, the TOV
during load rejection can reach 1.4 p.u. or higher due to Ferranti effect under self-excitation
phenomena, and when an earth fault follows the load rejection, TOV can reach around 1.5 p.u.
with a trapezoid waveform due to the clipping voltage of MOSAs. The energy capability of
MOSAs is specified as 55 MJ or more to ensure this TOV level. If necessary, an overvoltage
protection relay system to open the no-load OH line is applied to avoid excessive energy
absorption. [9]
It is important to recognize that in EHV/UHV transmission, it is impossible to completely
exclude TOV, but it is possible to reduce its magnitude and duration to acceptable limits. From
this point of view, it would help to ask UHV equipment manufacturers to provide information
about permissible TOV magnitude versus time of application for equipment with insulation,
magnetic core and energy absorption designed to withstand operating voltage and switching and
lightning overvoltages. Such information, based on the “natural” insulation strength, already
exists in Russia’s standards. [5]
3.1.3 Slow front overvoltage (SFO)

Slow-front overvoltage or switching overvoltage is controlled by means of such technologies as
pre-insertion resistors (closing resistors or PIR), opening resistors, controlled switching, and
application of MOSAs and of MOVs across the arcing chambers. Without taking certain
measures, the switching surge at line energization can lead to a maximum SFO of 3 p.u. By
80
means of PIR, the switching surge is reduced to 2.0-2.2 p.u., and by two-step PIR even further to
1.5-1.6 p.u. The application of controlled switching leads to switching surges less than 2.0 p.u.,
while MOSAs lead to SFO of 1.7-2.2 p.u., depending on the SIPL. A combination of measures,
such as controlled switching with PIR and MOSAs, could decrease the SFO to 1.6 p.u. These
SFO levels are along the OH line and are higher than at the line ends and at the substation busbar
side. Phase-to-phase switching overvoltages along the OH line are 2.6 to 2.9 p.u.
Without taking measures, SPAR leads to SFO of less than 2.4 p.u., but by means of the measures
mentioned above, it is possible to achieve SFO similar to that with line energization. TPAR
(three-phase rapid auto-reclosing) gives an SFO between 3.8 and 4.0 p.u. (with trapped charge) or
up to 3.0 p.u. (without trapped charge). The application of PIR, controlled switching and/or
MOSAs decreases the SFO to less than 2.5, 2.0 and 1.7 p.u., respectively. [8] When such low
SFO values are reached, the SFO caused by clearing single-phase faults becomes quite important,
as it may be larger: 1.8-2.1 p.u. Opening resistors will reduce the SFO at clearing short-circuit
currents for single-phase faults as well as three-phase faults. [2]
Note that for systems with series compensation, the given values for SFO may be higher due to
the residual charge on the series capacitor bank. This depends on the network topology in
combination with the location of the capacitor banks, as well as on the protection measures
implemented for the series capacitor bank: MOV, triggered spark gap, bypass breaker and its
protection scheme, etc.
Ⅰ Ⅱ Ⅲ Ⅳ
UN UN UN UN
～～～～～～～
U2 U1 U1 U1
～～～～～～～～～
U2 U2 U5 U2 ～ U5 U4～～～
U3 U3 U3 ～ U3 ～
～～
p.u p.u
1.2 1.2
Voltage on sound phase(pu)
１LG ２LG
× ×
×
△▲ △
▲ 1.1 ▲
1.1 △ × △
▲ ▲
▲ × ▲ ×
× △
▲ ×▲ △
▲ × △ × ▲ ▲
▲
△ △ △ △ △ △ ×
▲ ▲
△
× × △ ×
1.0 △ 1.0 ×
△ ▲
△
U2 UN U3 U2 U5 U4 U1 U2 UN U3 U2 U5 U4 U1
Ⅰ △ Ⅴ For future Pj, Ⅰ △ Ⅴ For future Pj,
Ⅱ × Ⅵ which is not 1100 Ⅱ × Ⅵ which is not 1100
1 pu = kV 1 pu = kV
explained 3 explained
Ⅲ Ⅶ Ⅲ Ⅶ 3
here. here.
▲ Ⅳ ▲ Ⅳ
Resistance at fault point:0ohm Resistance at fault point:0ohm
Fig. 3.1.4 TOV in healthy phases during ground faults (Japan)
At the occurrence of a single-phase-to-ground fault (1LG), the phase voltage changes to zero
potential instantaneously. The voltage jump leads to travelling waves in the faulty phase and
induces travelling waves in the healthy phases. In the process of propagation and reflection in the
healthy phases, transient voltages are superimposed on the AC voltages, leading to overvoltages.
These ground fault overvoltages are regarded as SFO and are highest when the earth fault occurs
near the peak value of the power frequency voltage. Maximum SFO tends to appear in the middle
81
of the OH line, where TOV is highest; see Fig. 3.1.4 where TOVs are simulated for TEPCO’s
future 1100 kV system. A simplified model of the system with simulated fault locations is given
in Fig. 3.1.5 and ground fault overvoltages along the system are shown in Fig. 3.1.6 (i.e.
maximum SFO for any fault location and any phase angle of the earth fault). Figure 3.1.7 gives
an example of the waveforms for the maximum SFO at the location of the highest SFO value
(1.62 p.u.) for each phase. The travelling waves can be clearly distinguished.
Fig. 3.1.5 Simulated future UHV system planned by TEPCO
Fig. 3.1.6 Ground fault overvoltages (maximum SFO at healthy phases) along UHV system
As reported in Ref. [10], the SFO is proportional to the earth-fault factor and may reach values as
high as 1.5 p.u. in systems with an earth-fault factor of 1.25 to 1.3. However, as reported in
References [10] and [11], in the future UHV system in Japan with kpp as low as 1.1 (earth-fault
factor as low as 1.06), the overvoltage at healthy phases may reach 1.6 p.u. in substations and 1.7
on OH lines, as shown in the above figures. In a UHV system, transmission losses are very low
because multi-bundle conductors are applied in order to suppress corona noise. Surge voltage
propagates along the OH lines in line-to-earth and line-to-line modes. The results of measurement
on an actual UHV OH line reveal that line-to-line waves hardly attenuate during propagation
along 200 km.
82
Fig. 3.1.7 Analysed waveform of maximum SFO at single phase to ground fault
(near location U6, SA protection level: V20 kA = 1620 kV)
Moreover, the topology of UHV systems is radial and rather simple, leading to far less refraction
of travelling waves compared to that in meshed EHV systems. Both characteristics of UHV
systems lead to SFO at the occurrence of a phase-to-ground fault as high as 1.6-1.7 p.u., even in
the case of low earth-fault factors.
For UHV systems, the following calculations for SFO have been reported [2],[12],[13]:
Table 3.1.2 Slow front overvoltages (SFO)
SFO (pu) Italy Russia Japan China
Line ph-gr 1.7 1.8 (1.6) 1.6/1.7 1.7
Substation ph-gr 1.8 (1.6) < 1.6 1.6
Line ph-ph 2.7 2.6/2.8 2.9
Substation ph-ph 2.6/2.8 2.8
Note that switching overvoltages in Japan are reduced to the same level as ground-fault
overvoltages by means of closing/opening resistors. Additionally, UHV OH lines in Japan are
rather short in length compared to those in Russia and China. For Russia, the values without
brackets apply in cases of gap-type SiC arresters; the values within brackets are those when
MOSAs would be used and are comparable with the other countries. Brackets are used for the
same reason in other tables.
83
3.1.4 Fast front overvoltage (FFO)
Fast front overvoltage or lightning overvoltage is a less important parameter for UHV OH lines in
comparison to EHV and HV OH lines. The specified TOV and SFO determine the dimensions of
the OH lines. Furthermore, the accepted number of outages due to lightning determines the
shielding angles, tower footing resistance, etc.
For substation equipment, the specified LIWV is of interest. Table 3.1.3 gives the general
specifications for LIWV and SIWV in substations, as required in several countries. [2],[12],[13]
Table 3.1.3 Insulation level for substation equipment

(kVpeak ) Italy Russia Japan China
Max. operating voltage 1050 (kVrms) 1200 (kVrms) 1100 (kVrms) 1100 (kVrms)
SIWV 1675 2100 (1800) 1550 1800
pu 1.95 2.14 (1.84) 1.73 2.00
LIWV 2250 2900 (2400) 2250 2400
pu 2.62 2.96 (2.45) 2.51 2.67
Lower values are specified for transformers, as can be seen in Table 3.1.4, where only the UHV
side is presented.
Table 3.1.4 Insulation level for transformers
(kVpeak) Italy Russia Japan China
SIWV 1800 2100 (1800) 1425 1800
pu 2.10 2.14 (1.84) 1.59 2.00
LIWV 2250 2550 (2250) 1950 2250
pu 2.62 2.60 (2.30) 2.17 2.51
In countries such as Russia and China, the same SIWV is proposed for transformers and reactors
as for other substation equipment, while in Japan the proposed SIWV requirements are roughly
10% lower in comparison to other equipment, and in Italy higher SIWV values are proposed.
This is mainly due to the fact that Japan’s UHV system has been optimised with respect to
limiting the switching surges as much as possible. Furthermore, expressed in p.u., the SIWV
requirements for both transformers and other equipment in Italy, Russia (with MOSAs) and
China are relatively close to each other, but are much higher than those of Japan.
Also, with respect to the LIWV, Italy shows the same LIWV for transformers as for the other
equipment. Japan requires LIWV values roughly 13% lower for transformers than for other
equipment, and Russia/China roughly 6% lower values. Expressed in p.u., LIWV requirements
for transformers in Italy, Russia (with MOSAs) and China are comparable. For other equipment,
the proposed LIWV levels are fairly close.
For all equipment, Japan requires the lowest SIWV and LIWV at each level, as full advantage is
taken of improved technologies to decrease the insulation levels. In figure 3.1.8, a comparison in
p.u. is shown between the SIWV and LIWV applied in different countries.
84
4
Lighting Impulse Withstand Voltage, LIWV (p.u.)
3.21
2.96
3 2.76 2.67 2.62 2.62 2.60
2.51 2.51 2.45
2.30
2.17
Substation equipment
2
Transformer
Transformer
Transformer
Transformer
Transformer
1
0
kV kV
kV 0k
V
0 kV
0k
V 1 200 SA) 1200 A)
C8
00 110 110 105 ssi
a MO
ssi
a OS
an ina y Ru thout Ru ith M
IE Jap Ch Ital i W
(W (
4
Switching Impulse Withstand Voltage, SIWV (p.u.)
2.18 2.10 1.95 2.14 2.14

1.99 2.00 2.00
2 1.84 1.84
1.73
1.59
Substationequipment
Substationequipment
Substationequipment
Transformer
Transformer
Transformer
Transformer
Transformer
0
kV kV
kV 0k
V
0 kV
0k
V 1 200 SA) 1200 A)
C8
00 110 110 105 ssi
a MO
ssi
a OS
an ina y Ru thout Ru ith M
IE Jap Ch Ital i W
(W (
Fig. 3.1.8 LIWV, SIWV
Withstand voltage for UHV equipment were previously discussed within IEC TC 28 WG01.
Through document IEC 28 (Secretariat) 81, a proposal for withstand voltages was made in 1987,
but they were not standardized because there was no market relevance at that time. Table 3.1.5
compares the SIWV and LIWV of the document with that of China and Japan’s recent projects.
For SIWV, the values applied in Japan are well in line with the document, but the values applied
in China are rather high. For LIWV, the values of China and Japan are in agreement. [14] The
discussion will continue in IEC TC 28.
Table 3.1.5 Comparison of LIWV and SIWV of IEC with China and Japan
IEC28 (Secr.)81 China Japan
Highest voltage for
1050-1100 1100 1100
equipment Um (kV)
Standard switching 1425
1425
impulse withstand voltage 1550 1800
1550
phase-to-earth (kV) 1675
1950
Standard lightning impulse 2100 2250 1950
withstand voltage (kV) 2250 2400 2250
2400
In Ref. [15], the specifications for longitudinal withstand strength are given for China. The SIWV
is 1675 + 900 kV and the LIWV is 2400 + 900 kV. Additional information is presented in
Table 5.3.1.
85
3.1.5 Insulation levels and arrester protection levels
The information available covers gapped arresters applied in Russia and MOSAs applied in Italy,
Japan and China. [12],[13] A comparison is shown in Table 3.1.6.
Table 3.1.6 MOSA characteristics
kV Italy Russia Japan China
Rated voltage (rms) 749 800 826 828
Peak at 10 kA 1550 1553
Peak at 14 kA 1850
Peak at 20 kA 1800 1620 1620
LIPL (pu) 2.10 1.89 1.80 1.80
SIPL (pu) 1.62
In the figures 3.1.9 and 3.1.10 IEC insulation levels and commonly used protection levels are
given for system voltages up to and including 800 kV. Insulation levels and arrester protection
levels for the UHV projects in Italy, Russia (USSR), Japan and China have been added as well.
As seen from the diagrams the trend in decreasing relative insulation levels continues. The very
low relative arrester protection levels for the projects in Japan, China and Russia are also striking.
4
SIWV and SIPL (p.u.) SIWV: Switching Impulse Withstand Voltage
3.06
SIPL: Switching Impulse Protection Level
3 2.77 Max.
2.62
2.48
Min. 2.34 Max. 2.37
2.12 2.18 Max.
1.99 2.00 1.95
2 SIPL level Min. Min. 1.84
1.73
SIPLlevel SIPL level
China's SIPLlevel
Russia 1200 kV
Japan 1100 kV
China 1100 kV
Italy 1050 kV
1
0
kV kV kV
20 50 00 UH
V
IE C4 IE C5 IE C8
Fig. 3.1.9 SIWV, SIPL
Apart from Russia’s use of conventional gap-type surge arresters (SiC, ZnO) as well as MOSAs
for new applications, the following information can be given on Russian specifications. ZnO
arresters used for SIPL of 1.8 p.u. have a maximum permissible operating voltage of 694 kV and
permitted voltage of 760 kV for 20 minutes. The extinguishing voltage of the power frequency
current, after a switching impulse current, is 1000 kV and after a lightning impulse current,
800 kV. The residual voltage at a switching impulse current of 1.5 kA (3/8 ms) is no more than
1760 kV and at a lightning impulse current of 14 kA (8 µs) is no more than 1940 kV.
MOSAs, used for SIPL of 1.6 p.u., have the same maximum permissible operating voltage and a
permitted voltage of 765 kV for 60 minutes. Residual voltage at a switching impulse current of
2.8 kA (1.2/2.5 ms) is no more than 1570 kV, and at a lightning impulse current of 15 kA (8 µs),
no more than 1760 kV.
86
3.1.6 Comparison with insulation withstand voltages calculated as per
IEC procedure
The IEC procedure for insulation coordination is outlined in IEC 60071-1. [16] The application
guide can be found in IEC 60071-2. [17] The procedure is applied here to an 1100 kV system,
using arrester protection levels for fast- and slow-front overvoltages, i.e. LIPL and SIPL with
values specified for the UHV systems in Japan and China. The calculation results such as for
required withstand voltage could be compared with the selected insulation levels for existing and
planned UHV systems.
3.1.6.1 Slow front overvoltages

Depending on the degree of limitation of slow-front overvoltage by arresters, the coordination
factor Kcd shall be between 1 and 1.1 as per Ref. [17]. See Fig. 3.1.4.
The coordination withstand voltage, Ucw, is Kcd times the arrester protection level and thus shall
be 1 to 1.1 times the SIPL.
To obtain the required withstand voltage, Urw, the coordination withstand voltage shall be
multiplied by the altitude correction factor, Ka, considering 1000 m and the safety factor, Ks.
The altitude correction factor, applied for external insulation only, is equal to em(H/8150) where H is
the altitude and m the factor according to the Figure 9 of Ref. [16]. For the coordination voltages
expected for the 1100 kV system, the m-factor, for phase-to-earth insulation, is around 0.5, which
results in Ka = 1.06.
The recommended safety factors as per IEC 60071-2 are as follows:
- For external insulation Ks = 1.05
- For internal insulation Ks = 1.15
Therefore, the required withstand voltage will be:
- For external insulation Urw = 1.11 to 1.22 times SIPL
- For internal insulation Urw = 1.15 to 1.27 times SIPL
With an SIPL of approximately 1.6 p.u.(1440 kV), Urw will be:
- For external insulation 1598 to 1757 kV
- For internal insulation 1656 to 1829 kV
3.1.6.2 Fast front overvoltages

The coordination lightning impulse withstand voltage can be directly calculated by means of the
simplified method in IEC 60071-2 using the following empirical formula:
Ucw = LIPL+ 2*S*L/c = LIPL + A/n*L/(Lsp+La), where
LIPL = Lightning impulse protection level of the arrester
La = Ra/Rkm line length with an outage rate equal to Ra (m)
Ra = Acceptable failure rate (1/year)
Rkm = Outage rate for lines connected to the station (1/(m*year)
S = 1/(Kco*(Lsp+La))
Kco = Corona damping factor (µs/(kV*m))
A = 2/(Kco*c)
87
c = Speed of light (300 m/µs)
Lsp = Span length (m)
L = Height of arrester + length of connection leads + distance to protected object (m)
n = Number of lines connected to the station
As an example, assuming the following:
- Ra: Acceptable failure rate 0.0025 (once in 400 years)
- Rkm: 0.00001 = One lightning fault per 100 km per year on the line in front of the
station
- N: One or two lines to the station
- Lsp: Span length 500 m
- Kco = 0.4E-6 (for 6- or 8-conductor bundle)
- L = 30 m
This will result in the following coordination withstand voltage:
- For one line to the station Ucw = LIPL + 667 kV
- For two lines to the station Ucw = LIPL + 333 kV
To obtain the required withstand voltage, Urw, the coordination withstand voltage shall be
multiplied by the altitude correction factor, Ka, considering 1000 m and the safety factor, Ks.
The altitude correction factor, applied for external insulation only, is equal to e(H/8150) where H is
the altitude (m-factor equals to 1). This gives Ka = 1.13.
The same safety factors as for slow-front overvoltage apply.
Therefore, the required withstand voltage, Urw, will be:
- External insulation Urw = 1.19*(LIPL+667) …… one line
- Internal insulation Urw =1.15*(LIPL+667) …… one line
- External insulation Urw =1.19*(LIPL+333) ……. two lines
- Internal insulation Urw =1.15*(LIPL+333) ……. two lines
Finally, with an LIPL of 1620 kV (1.8 p.u.), Urw will be:
- External insulation Urw = 2721 kV …… one line
- Internal insulation Urw = 2630 kV …… one line
- External insulation Urw = 2324 kV ……. two lines
- Internal insulation Urw = 2246 kV ……. two lines
The method described above was developed for AIS, but many of the UHV stations are GIS.
However, it is stated in IEC 60071-2 that the procedure gives a conservative estimate for GIS
even if a generally valid recommendation cannot be made on how conservative it is compared to
AIS.
3.1.6.3 Comparison
For slow-front overvoltage, the selected SIWV for substation equipment in Japan is 1550 kV and
in China 1800 kV. The procedure above indicates a required level for external equipment of 1598
to 1757 kV and for internal equipment 1656 to 1829 kV. China’s requirement is well within the
calculated values, but Japan’s significantly lower level is considered sufficient.
For fast-front overvoltage, the selected LIWV for substation equipment in Japan is 2250 and in
China 2400 kV. For transformers, the corresponding figures are 1950 and 2250 kV. The
88
procedure above, considering two lines, gives 2324 kV for external equipment and 2246 kV for
internal. For China, both substation equipment and transformer LIWV are above these values. In
Japan, LIWV for the substation equipment is above, but LIWV for transformers is below. If only
one line is considered, neither China nor Japan’s LIWV values are above the calculated levels.
It must be pointed out that the IEC procedure is a simplification and, as mentioned above, not
directly applicable to GIS. However, the procedure does reveal that the insulation coordination
for fast-front overvoltages taking into account the preferred low insulation levels for UHV
requires further detailed study.
3.1.7 Very fast transient overvoltage (VFTO)

Very fast transient overvoltage is generated by disconnectors within GIS installations. It can
reach up to 2.6 p.u. or even more (in one specific case reported in Ref. [18], the VFTO reaches
3.05 p.u.), at frequencies in the MHz range. When PIR is used for these disconnectors at opening
and closing, overvoltage can be limited to values such as 1.4 p.u. [19] At the UHV level, in order
to prevent flashovers inside the GIS between conductor and tank, it is crucial to apply such PIR
to the disconnectors. As reported in Ref. [18], in China’s pilot project, PIR will not be installed
for MTS (mixed technology switchgear or HGIS, hybrid GIS).
VFTO is discussed in further detail in Section 3.4.1.5.
3.1.8 Clearances
The required clearances between phase and ground, between phases, between different circuits
and between phase and structure is dependent on altitude, prescribed wind velocity, safety
requirements, risk of flashover, etc. But at the UHV level, it is obvious in all countries that the
switching overvoltage is the most crucial stress factor for determining the required clearance. As
climatic parameters and legal conditions are most dominant in OH lines, it does not make much
sense to compare the clearances in OH lines between the UHV configurations in different
countries. In substations, the conductor position is less influenced by climatic parameters and
therefore clearances in substations are preferable for comparison.
The clearances between phase and ground, between phases and between phase and structure are
given in Table 3.1.7. [2],[13]
Table 3.1.7 Clearances of phase to ground, phases to phases, and phase to structure
(m) Italy Russia Japan China
Phase to ground 8 12 8.5/10
Phase to phase 12 11.4 -12.4 10.5 -11.5 11.3
Phase to structure 9.5 7.5 - 9.7 7.5 - 8.5 7.5
To clarify the necessary clearance at the entrance to UHV substations, [20] flashover tests were
carried out using shielding rings under dry and wet (water spray) conditions. It was confirmed
that flashover voltages decrease under wet conditions, as shown in Fig. 3.1.10, because water
droplets on the shielding ring enhance the local electric field and lead to the occurrence of
streamers. This means that with large shielding rings such as those for bushings, switching
impulse voltages possibly increase under dry conditions, but decrease under wet. Accordingly, in
Japan, the withstand voltage test of a bushing was performed under dry and wet conditions. In the
test, waveforms of 250/2500 µs (standard) and 500/3000 µs were both applied taking into
account that the critical front duration tends to be longer as the gap length tends to be larger.
89
Further information on clearances will be provided by CIGRE WG B3.22.
Dry
Wet
Rod - plane
［Rod-plane］
］
Figure 3.1.10 Critical 50% flashover characteristics (Phase to structure)
90
3.2 Transformers
3.2.1 Insulation level

The insulation level for power transformers should be specified based on insulation coordination
with consideration for margin and the practise of utilities in terms of protection levels of different
types of substations such as “AIS”, “MTS”, “GIS”, etc., and arrangement/connection points in
the substations for applied line arresters. For the 800 kV class system voltage, the insulation level
for transformers should be selected or specified based on Tables 3.2.1 and 3.2.2, standardized in
IEC and IEEE, respectively. The IEC and IEEE standards are lined up by multiple values of the
insulation level for the 800 kV class, with discretion for the selection of insulation level by utility
and taking into consideration the decision of rational specifications for power transformers.
Table 3.2.1 Rated withstand voltages for transformer windings with Um=800 kV
(IEC 60076-3 2000-03, Part 3 : Insulation levels, dielectric tests and external clearances in air)
Um ; Highest Rated switching impulse Rated lightning Rated short-duration
voltage for withstand voltage phase- impulse withstand induced AC withstand
equipment (kVr.m.s) to-earth (kVpeak) voltage (kVpeak) voltage (kVr.m.s)
1 550 630
1 300 1 675 680
800 1 425 1 800 N/A
1 550 1 950 N/A
2 100 N/A
Table 3.2.2 Dielectric insulation levels for Class Ⅱ power transformers

(IEEE Std C57.12.00-2000 : IEEE Standard General Requirements for Liquid-Immersed
Distribution, Power, and Regulating Transformers)
Low frequency test levels
Basic
Nominal lightning Switching Induced-voltage test (phase to ground)
Chopped
system impulse impulse
wave level
voltage insulation level
(kV crest) One hour level Enhancement
(kV) level (kV crest)
(kV crest) (kV rms) level (kV rms)
Column 1 Column 2 Column 3 Column 4 Column 5 Column 6

1 800 1 980 1 500 690 800
765 1 925 2 120 1 600 690 800
2 050 2 255 1 700 690 800
On the 1100 kV class system voltage, the insulation levels are not specified in any international
standards, and are presently decided by each utility according to individual 1100 kV transmission
projects. Table 3.2.3 shows examples of insulation levels of 1100 kV project transformers that are
actually in operation or are part of an ongoing project. The records for Russia are from the 1980s,
and the insulation levels selected for China and Japan’s projects involve the existence of high-
performance zinc oxide arresters (protection level: 1620 kVpeak at 20 kA). However, as different
insulations levels are decided, it is necessary to standardize them with a lineup of insulation
levels due to the application of arresters with similar characteristics for both projects
91
Table 3.2.3 Rated withstand voltage for UHV transformer windings
Um ; Highest Rated switching Rated lightning Rated short-
voltage for impulse withstand impulse duration induced
Country
equipment voltage phase-to-earth withstand voltage AC withstand
kVr.m.s kVpeak kVpeak voltage kVr.m.s
Russia 1 200 2 100 2 550 1 100
China 1 100 1 800 2 250 1 100
Japan 1 100 1 425 1 950 1 100
3.2.2 Equivalent surge capacitance

As there is no official data on equivalent surge capacitance for UHV transformers, relevant
research is underway. In the meantime, it is assumed that the values would be quite high
compared with 500 kV class transformers because of the extremely large capacity of UHV
transformers (the power supply of UHV transformers for China and Japan’s projects is
1000 MVA per phase). There is an example value of approximately 10,000 pF per phase with
single phase, core type and 1000 MVA per phase, but the value is for high voltage, large capacity
and 6 legs in parallel connection per phase, so it would be changeable according to the type of
transformer, core or shell, the arrangement of windings, and the provisions against surges in the
high-voltage windings, etc. On the other hand, there are another available values that the
equivalent surge capacitances are approximately 1,000-2,000pF per phase. Therefore, in the
general calculations for insulation coordination, a wide range of values or large margin shall be
considered for the equivalent surge capacitances ranging from 1,000 to 10,000pF per phase.
Similarly, an example value of approximately 3000 pF per phase for equivalent surge capacitance
for 800 kV class transformers requires the same considerations as for UHV transformers.
92
3.3 Circuit Breakers
3.3.1 TRV for terminal fault test duties
3.3.1.1 General
For rated voltages equal to or higher than 100 kV, the TRV for terminal fault test duties is
described by two parameters (T10 and T30) or four parameters (T60 and T100).
Voltage
Uc
Uc: TRV peak

U' Uc/ t3:Rate of rise of TRV
0 td t' t3 Time
Fig. 3.3.1 Representation of specified TRV by two-parameter reference line and delay line
Voltage
Uc
U1 Uc: TRV peak

U1: First reference voltage
U1/ t1:Rate of rise of TRV
U'
0 td t' t1 t2 Time
Fig. 3.3.2 Representation of specified TRV by four-parameter reference line and delay line
The peak value of TRV (Uc) applied to a circuit breaker during interruption of a terminal fault is
given by,
2
Uc = Ur k pp k af (1)
3
where,
Uc TRV peak
Ur Rated voltage
93
kpp First-pole-to-clear factor
kaf Amplitude factor
For a given rated voltage, both the first-pole-to-clear factor and the amplitude factors must be
defined for determining the peak value of TRV.
The power frequency recovery voltage, given by Equation (2), is a function only of the rated
voltage and the first-pole-to-clear factor.
U r k pp
U= (2)
3
3.3.1.2 Inherent TRV calculated based on Japan’s UHV system

Inherent TRVs for terminal fault duties on one side and those for Long Line Fault (LLF) on the
other side were analysed using EMTP considering electromagnetically and electrostatically
induced voltages under 3LG conditions in TEPCO’s 1100 kV transmission systems without
MOSAs, shown in Fig. 3.3.3. TRVs for transformer fed faults (Transformer Limited Fault: TLF)
were also calculated for single-phase-to-ground fault (1LG) conditions in addition to three-phase-
to-ground fault (3LG) conditions.
The calculated TRVs, summarized in Table 3.3.1, are classified as TLF, LLF, T60 and T100
duties. Figure 3.3.4 compares typical TRV waveforms with a TRV representation having TRV
peak values twice that of existing 550 kV standards. Furthermore, TRV peak values and rates of
rise are plotted as functions of the breaking currents in Fig. 3.3.5. Most of the calculated TRV
values can be covered by a specified TRV with peak values twice that of existing 550 kV
standards. However, rates of rise of TRV for TLF far exceeds the standard values due to higher
TRV frequency around 5 kHz and therefore special measures might be required for circuit
breakers.
231U 231L
D s/s D9 D8 D10
FDBL
Transmission line
(50km)
Transmission line Transmission line
FBEL (40km)
E s/s (210km)
A s/s
FBDL FEBL FEAL FAEL
B s/s B11 B7 B8 B12 E8 E9 A11 A12

B6 E10 E11
FBBUS E7 A10
B9 B10
B1
FBCL
218 204A 204B
224
Transmission line FCBL
(138km)
C8 C9 : Power transformer
C7
C s/s : Fault point
C1
FCBUS
226
Fig. 3.3.3 1100 kV system used for TRV analysis
94
2500 2000
TRV for T30 with twice TRV peak value of existing 550kV standards
2000
1500
Voltage (kV)
Voltage (kV)
1500
1000
1000
500
500
TLF, Breaking currents : 6.1-7.4 kA LLF, Breaking currents : 3.5-23.4kA
0 0
0 0.2 0.4 0.6 0.8 1.0 0 0.5 1.0 1.5 2.0 2.5
Time (ms) Time (ms)
2000 2000
TRV for T100 with twice TRV peak value of existing 550kV
standards
1500 1500
Voltage (kV)
Voltage (kV)
1000 1000
500 500
T60, Breaking currents : 19.1-26.2kA T100, Breaking currents : 30.5-36.0kA

0 0
0 1.0 2.0 3.0 4.0 5.0 0 1.0 2.0 3.0 4.0 5.0
Time (ms) Time (ms)
Fig. 3.3.4 Comparisons of TRV waveforms with twice TRV peak values of existing 550 kV standard
Table 3.3.1 Inherent TRV for TEPCO 1100 kV system

Duties Breaking currents (kA) Rate of rise of TRV (kV/µ
µsec) TRV peak values (kV)
TLF 6.1 (12)-7.5 (15%) 6.2-14.5 1046-1528
LLF 3.5 (7)-14.5(29%) 0.96-4.25 1049-1879
T60 19.1 (38)-27.2 (54%) 1.08-5.47 890-1586
T100 30.5 (61)-36.0 (72%) 0.84-1.38 927-1571
2500 20
T60-T100 TLF LLF
Rate of rise of TRV u1/t1 (kV/µs)
15
TRV peak value Uc (kV)
2000 Twice TRV peak values of existing 550kV standards
1500
10
Rate of rise of TRV of existing 550kV standards
1000
5
500 1100kV systems without MOSA and T60-T100 TLF LLF
GCB without Opening resister
0 0
0 10 20 30 40 50 0 10 20 30 40 50
Breaking current (kA) Breaking current (kA)
Fig. 3.3.5 TRV peak values and rates of rise of TRV are plotted as functions of breaking currents
Table 3.3.2 Inherent TRV peak values of 1100 kV system for different FPCF
Test- Uc (kV) Uc (kV) Uc (kV) 90% value of Maximum TRV
Faults kaf
duty kpp=1.3 kpp=1.2 kpp=1.1 TRV peak (kV) peak (kV)
TLF T10 1.76 2061 1903 1744 1514 1528
LLF T10 1.76 2061 1903 1744 1746 1747
LLF T30 1.54 1798 1660 1522 1685 1879
BTF T60 1.50 1751 1617 1482 1405 1586
BTF T100 1.40 1634 1509 1383 1512 1571
95
Table 3.3.2 shows the 90% values of inherent TRV peaks calculated for TEPCO’s 1100 kV
transmission lines compared with inherent TRV peak values for different first-pole-to-clear
factors. If the amplitude factor in IEC 62271-100 (with Amendment 3, that will be approved as
edition 2.0) can be applied to the 1100 kV system, a first-pole-to-clear factor of 1.2 covers most
of 90% inherent TRV peak values.
In conclusion, inherent TRVs of TEPCO’s 1100 kV system can be covered by TRV peak values
twice that of existing 550 kV standards except in the case of TLF. [21]
3.3.1.3 First pole-to-clear factor (kpp) for terminal fault test duties
IEC 62271-100 specifies a first-pole-to-clear factor of 1.3 for effectively earthed neutral systems
irrespective of the rated voltage. It is calculated from Equation (3) using the zero-sequence
reactance X0 and positive-sequence reactance X1
3
k pp = (3)
X
2+ 1
X0
In a network with long transmission lines, the first-pole-to-clear factor tends to increase, because
the ratio of X1/X0 of lines relatively becomes smaller. On the other hand, in the case of a network
connected to large power transformers (a star connection with an earthed neutral or a delta
connection), the first-pole-to-clear factor becomes smaller and occasionally less than 1.2, because
the ratio of X1/X0 is equal to or larger than 0.5 (2X1 > X0). Especially in cases where most of the
short-circuit currents are fed through large-capacity power transformers, the first-pole-to-clear
factors are smaller, because the zero-sequence impedance is reduced due to the delta connection
of large-capacity power transformers (X1/X0 approaches unity, or X0 approaches X1).
The impedance of OH lines is nearly in reverse proportional to the square of the system operating
voltage, so the ratio of line impedance to total transmission system impedance tends to be smaller
for a system operating higher voltages. This tendency reduces the zero-sequence impedance in
UHV systems due to the increasing influence of large-capacity power transformers that have
smaller zero-sequence impedance X0 compared with transmission lines. Accordingly, the first-
pole-to-clear factors in UHV systems generally have smaller values than those for the systems
with lower voltages as described in the section 3.1.2.
Figure 3.3.6 shows the calculation results of the first-pole-to-clear factors for 245, 420, 550 and
1100 kV transmission lines in Japan [22] and the Netherlands. The first-pole-to-clear factor
values investigated in recent 550 kV transmission systems are generally smaller than 1.2. For
1100 kV transmission systems, the known values of kpp do not exceed 1.2.
96
1.3
550kV-BTF
550kV-LLF 420kV-BTF
1.2
First pole-to-clear factor

1.1
1.0
0.9
1100kV-BTF
245kV-BTF
0.8
0 10 20 30 40 50 60 70 80 90 100
Short-circuit current divided by
the maximum bus fault current (%)
Fig. 3.3.6 First pole-to-clear factors in 245-1100 kV systems

The first-pole-to-clear factors were evaluated at four different stages of TEPCO’s 1100 kV
transmission lines including 550 kV systems shown in Fig. 3.3.7. Overvoltages of faulted and
healthy phases in the case of 1LG and 2LG were investigated using EMTP.
Ⅰ Ⅱ Ⅲ Ⅳ
UN UN UN UN
～～～～～～～
U2 U1 U1 U1
～～～～～～～～～
U2 U2 U5 U2 ～ U5 U4～～～
U3 U3 U3 ～ U3 ～
～～
Fig. 3.3.7 1100kV transmission network used for calculation

Figures 3.3.8 summarize the calculated results of overvoltages in healthy phases during 1LG and
2LG. Overvoltages prevail between 0.95 and 1.1 p.u. at each substation. Therefore, the first-pole-
to-clear factor of TRV duties was defined as 1.1 in Japan. [21]
p.u p.u
1.2 1.2
Voltage on sound phase(pu)
１LG ２LG
× ×
×
△▲ △
▲ 1.1 ▲
1.1 △ × △
▲ ▲
▲ × ▲ ×
× △
▲ ×▲ △
▲ × △ × ▲ ▲
▲
△ △ △ △ △ △ ×
▲ ▲
△
× × △ ×
1.0 △ 1.0 ×
△ ▲
△
U2 UN U3 U2 U5 U4 U1 U2 UN U3 U2 U5 U4 U1
Ⅰ △ Ⅴ For future Pj, Ⅰ △ Ⅴ For future Pj,
Ⅱ × Ⅵ which is not 1100 Ⅱ × Ⅵ which is not 1100
1 pu = kV 1 pu = kV
explained 3 explained
Ⅲ Ⅶ Ⅲ Ⅶ 3
here. here.
▲ Ⅳ ▲ Ⅳ
Resistance at fault point:0ohm Resistance at fault point:0ohm
Figs. 3.3.8 (a) FPCF for Japanese UHV (1LG), (b) FPCF for Japanese UHV (2LG)
In conclusion, the first-pole-to-clear factor for the rated voltage of 1000 kV and above could be
reduced to less than the existing values of IEC standard for rated voltages up to 800 kV, even
97
though the effect of the line length should be considered in detail.
3.3.1.4 Amplitude factor (kaf) for terminal fault test duties

The amplitude factor of inherent TRV in UHV systems tends to increase due to low losses of
power transformers and transmission lines as well as low damping of travelling waves. For
instance, TRV of test-duty T100 in Japan’s future UHV system will show an amplitude factor kaf
of 1.58 instead of the standard value of 1.4 based on kpp of 1.1. Because of the low kpp, TRV of
the last pole shows the most severe peak value under certain conditions.
Table 3.3.3 gives the amplitude factors in IEC 62271-100 (including Amendment 3, at the final
stage of approval as edition 2.0) for terminal fault interruption by circuit breakers with a rated
voltage of 100 kV and above. The amplitude factor is 1.7×0.9 = 1.53 for test-duty T10 based on a
first-pole-to-clear factor of 1.5 with voltage reduction of 0.9 across the transformer.
Table 3.3.3 Amplitude factors for terminal fault test duties
Test duty Amplitude factor
T10 1.76
T30 1.54
T60 1.50
T100 1.40
TRV calculations performed for UHV networks in China and Japan indicate that the amplitude
factors do not show much difference in those stipulated in IEC 62271-100 for networks with
lower rated voltage than 1100 kV. Concerning TRVs for breaking current equal to 10% of the
rated short-circuit current, it must be checked that it also covers cases of three-phase line faults
with low short-circuit current and full short-circuit source power.
CIGRE WG A3.19 is currently studying such cases. Figure 3.3.9 is an example that shows the
TRV calculated in the case of a three-phase line fault in Hydro-Québec’s 765 kV network.
Vn= 735 kV, Rated Isc= 40 kA, kpp= 1.3, L10
Source TRV parameters: Kaf= 1.40, RRRV= 2.0 kV/us
1500
Line & Source TRV
Pole-1 line TRV, TRV slope= 0.65 kV/us, d= 2.54
IEC Line TRV, L10, Zline= 450 ohms
IEC 2-parameter TRV - T10 (1495 kV - 186 us)
1000
TRV (kV)
500
0
0 500 1000 1500 2000 2500 3000 3500 4000 4500
t (us)
Figure 3.3.9 Comparison of TRV and line-side voltage for a 3-phase line fault in Hydro-Québec’s 765 kV
network (Nominal voltage: 735 kV) with TRV envelopes for T10 duty of 800 kV as stipulated in IEC
As shown in Figure 3.3.9, the TRV peak calculated based on the rated voltage of 735 kV in
Hydro Québec is equal to 1279 kV. The corresponding TRV peak relative to 800 kV
specifications is then,
98
800
1279 × = 1392 kV (4)
735
Hydro Québec uses the value of the 800/735 ratios, when considering the specifications of 800
kV substation equipment, thus adding some additional margins.
In Edition 1.2 of IEC 62271-100 (dated 2006-10) [23], the specified value of the TRV peak for
T10 duty is based on a first-pole-to-clear factor of 1.3 and is equal to
2
Uc = 800 × × 1.3 × 0.9 × 1.7 = 1299 kV (5)
3
From (4) and (5), it appears that some long line faults, associated with low short-circuit currents,
may not be covered by test duty T10 in IEC 62271-100. IEC SC17A has already taken this into
account and has prepared an amendment that is in the final phase of approval (FDIS). In this
Amendment 3 to IEC 62271-100, the first-pole-to-clear factor is raised to 1.5 and the amplitude
factor remains at 0.9 × 1.7 so that the peak TRV is raised to Uc = 1499 kV, higher than the value
in (4). From a design point of view, TRV peak should be lower than the SIWV value, except for
LLF and out-of-phase. TRV peak for 1100 kV lines is discussed in the section 3.3.1.2 and the
WG will be studied more in depth in the near future.
3.3.1.5 Rate of rise of recovery voltage (RRRV) for terminal fault test
duties
Table 3.3.4 gives the RRRV values in IEC 62271-100 for terminal fault interruption by circuit
breakers with a rated voltage of 100 kV and above.
Table 3.3.4 RRRV for terminal fault test duties
Test duty RRRV (kV/µ µs)
T10 7
T30 5
T60 3
T100 2
TRV calculations performed for UHV networks in China and Japan indicate that the values in
Table 3.3.4 are applicable for networks with a rated voltage of 1100 kV, [21], [24] except in the
case of TLF, which corresponds to test-duty T10. RRRV of TLF attains 14.5 kV/µs at maximum.
3.3.1.6 First reference voltage (U1) for terminal fault test duties
The first reference voltage (U1) represents the maximum voltage at which the RRRV is required
to be withstood during test duties T60 and T100 (see Figure 3.3.2). In IEC 62271-100, this value
is presently equal to
2
U 1 = 0.75 × U r k pp (6)
3
CIGRE WG A3.19 has been discussing whether the Equation (6) is also applicable to EHV/UHV
systems as it can allow covering cases of a line fault with 60% of the rated short-circuit current
and whether it is also applicable for terminal fault test duty T100, as shown in the past by WG23
of IEC SC17A when Amendment 1 to IEC 62271-100 was prepared.
Table 3.3.5 shows the 90% values of the first reference voltages calculated for TEPCO’s 1100 kV
99
transmission lines compared with U1 values for different first-pole-to-clear factors. A first-pole-
to-clear factor of 1.2 also covers the 90% values of the first reference voltage.
Table 3.3.5 First reference voltages of 1100kV system for different FPCF
Test- U1 (kV) U1 (kV) U1 (kV) 90% value of first Maximum first
Faults
duty kpp=1.3 kpp=1.2 kpp=1.1 reference voltage (kV) reference voltage (kV)
BTF T60 876 808 741 762.4 843.5
BTF T100 876 808 741 440.7 493.2
3.3.1.7 Effect of MOSAs with different characteristics on TRV

The influence of MOSAs on TRVs for terminal fault duties is analysed by EMTP using MOSAs
with different characteristics, as shown in Table 3.3.6. The V-I characteristics of Type A are
superior to that of Type B.
Figure 3.3.10 shows typical results of TRV waveforms with a TRV representation having peak
values twice that of existing 550 kV standards. MOSA can effectively reduce the TRV peak
values for terminal fault duties, so it should be noted that the TRV peak reduction does not show
a significant difference for MOSAs with different characteristics, because the restriction voltages
around 10–100 A determine the effect of TRV peak reduction and those values are similar for
different MOSAs.
The application of MOSAs can reduce the amplitude factors of TRV for terminal faults, TLF
(transformer limited or secondary faults) and out-of-phase. However, it may not lead to a
reduction large enough for LLF (long line faults whose applicability to UHV systems remains to
be established) where the TRV is generated at both the source and line sides of the breaker
terminals. The phenomena are not observed in transmission systems up to 800 kV. Therefore, the
effect should be considered when the TRVs for UHV are extrapolated and compared with the
existing standards.
2500 kA
TLF, Breaking curresnts : 7.4kA
2000 Uc=2061kV
Voltage (kV)
1500
1000
500 Without MOSA

With MOSA (A type characteristic)
With MOSA (B type characteristic)
0
0 0.2 0.4 0.6 0.8 1.0
Time (ms)
2000 kA
BTF, Breaking curresnts : 26.2kA 2000 kA
BTF, Breaking curresnts : 33.8kA
TRV for T60 and T100 with twice TRV peak value of existing 550kV standards
Uc=1751kV
Uc=1635kV Uc=1635kV
1500 1500
Voltage (kV)
Voltage (kV)
1000 1000
500 500
Without MOSA Without MOSA
With MOSA (A type characteristic) With MOSA (A type characteristic)
With MOSA (B type characteristic) With MOSA (B type characteristic)
0 0
0 1 2 3 4 5 0 1 2 3 4 5
Time (ms) Time (ms)
Figure 3.3.10 Effect of MOSA with different characteristics on TRV

100
Table 3.3.6 V-I characteristic of different types of MOSAs
Characteristic 1 mA 1A 10 A 100 A 1 kA 10 kA 20 kA 50 kA
Type A (kV) 1080 1235 1285 1335 1435 1555 1620 1770
Type B (kV) 1160 1250 1305 1370 1510 1710 1800 1900
3.3.1.8 Calculation of TLF fault current

Severe TRV conditions may occur when there is a short circuit immediately after a transformer
without any appreciable capacitance between the transformer and the circuit breaker. In such
cases, the rate-of-rise of transient recovery voltage (RRRV) may exceed the values specified in
the standards, due to the relatively small capacitance to earth of transformers, i.e. 3000 pF for
800 kV applications and up to 10,000 pF in UHV applications (transformers with six windings in
parallel per phase). The corresponding natural frequency of the transformer leads to TRV having
a rate of rise that is two or three times the value for a terminal fault with the same short-circuit
current. RRRV of 14.5 kV/µs has been reported for the UHV system in Japan. [25]
The system TRV can be modified by capacitance or a resistor so that it is within the standard
TRV capability envelope. As an alternative, a definite-purpose circuit breaker can be specified
for fast transient recovery voltage rise times.
For circuit breakers with a rated voltage equal to or higher than 100 kV, TRVs specified for
terminal fault test duties T30 and T10 cover most applications in this current range. In special
cases where the connection between the circuit breaker and the transformer has a very low
capacitance, a special test duty may be specified or capacitance could be added to allow the use
of a standard circuit breaker. TRVs for these special cases will be covered in a revised IEEE
C37.06-200x. [26]
Taking into account the actual characteristics of power transformers for rated voltages of 800 kV
and higher, such as short-circuit reactance and power, the fault current for transformer-limited
faults is usually between 10 and 16% of the rated short-circuit current of the circuit breaker (a
typical value of 10 kA corresponds to 20% of the rated short-circuit current for a 50 kA circuit
breaker and 16% for a 63 kA circuit breaker). However, a higher-value TLF current is possible
such as in the case of AEP’s 800 kV system in which the 765/500 kV transformers that connect
the AEP system to neighbouring systems have impedance as low as 6.35%, leading to a fault
current of up to 17.2 kA. On the opposite side, the maximum TLF current for Hydro-Québec’s
735 kV systems is only 6 kA.
As the TLF current varies in a wide range depending on the characteristics of the transformer, it
is suggested that a specific TLF breaking current be introduced, independently from the terminal
fault breaking current, as already recommended in IEEE C37.016 for high-voltage circuit
switchers.
3.3.2 TRV of compensated/uncompensated lines
3.3.2.1 Series compensated lines and their effect on circuit breakers

recovery voltage and switching surges
As utilities face increasing opposition to adding new transmission lines to networks, application
of series capacitors becomes an interesting alternative for enhancing transient and steady-state
stability and improving loading between parallel lines.
101
Series capacitor banks (SCB) have already been put into use in EHV systems (Hydro-Québec,
Turkey, BC Hydro, Venezuela, Chile, Sweden, etc.)
For UHV, where the power is generally transmitted over very long distances, the use of series-
compensated lines would certainly be beneficial by improving the voltage control and the voltage
and angle stability as well as increasing the maximum load for transmission.
3.3.2.2 Effects on the lines CB TRVs during line fault interruption

There are numerous published reports on the effects of series capacitors on line CB TRV severity
when clearing faults on series-compensated lines. [27]-[31] Depending on the protection schemes
used to set the bypass conditions for series capacitors, the line CBs can be exposed to severe
TRV exceeding IEC standard values. This is a case where the line CBs must clear a line fault for
which the series capacitor is not bypassed, resulting in a trapped charge on the series capacitor.
The following four types of SCB protection schemes are used:
• Slow re-insertion type: The bank is protected by a self-triggered gap and a bypass CB
only. For this type of protection, the gap is usually set so that it will self-trigger for most
faults on the line (internal faults) except for those with a relatively small fault current.
• Instantaneous re-insertion: The bank is protected by MOVs, forced-trigger spark gap and
bypass CB (as seen in Fig. 3.3.11).
• Fast re-insertion type: The bank is protected by two self-triggered gaps, a low setting gap
with a CB in series and a high setting gap with a bypass CB in parallel.
• Thyristor-controlled series capacitors (TCSC): As shown in Fig. 3.3.12, thyristors are
used to bypass or insert modules of capacitance and inductance to rapidly modulate the
line reactance. [32]
Fig. 3.3.11 Typical series-capacitor bank installation protected by MOVs
102
Fig. 3.3.12 Thyristor-controlled series capacitor
3.3.2.3 Turkey’s network

Iliceto et al. studied the effect of series capacitors on TRV across line circuit breakers on
Turkey’s 420 kV grid. [28],[29] Techniques such as surge arresters, opening resistors on CBs,
MOVs in parallel with CB contacts and fast bypassing of series capacitors were analysed. In the
first installations on the Turkish network, series capacitors were protected by self-triggering spark
gaps that would bypass the capacitors whenever a fault resulted in transient voltage across the
capacitors higher than the gap withstand voltage. Therefore, the increased TRV across the CB
would only be present for faults with a relatively small current so that the voltage across the
capacitor is lower than the setting for the spark gap.
In more recent series capacitor installations on the 420 kV network, the capacitors are protected
by MOVs across a circuit breaker terminal and therefore, the trapped charge is equal to the MOV
protection level. This scheme can lead to higher TRVs than the self-triggered spark-gap-protected
capacitor. For configuration where the series capacitors are located at an intermediate point of the
line, the TRVs are even higher due to the trapped charge on each side of the CB.
Iliceto et al. performed TRV analysis for different fault locations on the line (three-phase to
ground and three-phase isolated from ground). They showed that the highest TRV (4.11 p.u.) was
obtained on the remote CB during a three-phase-to-ground fault located beyond the series
capacitors. References [28] and [29] propose a list of mitigation measures for reducing the TRV
across the CB. By using MOVs in parallel with the interrupting chambers of the CBs, it was
possible to limit the TRV level to below 3.0 p.u. for CBs that are required to synchronize two
islanded parts of the network, and 2.5 p.u. for other line CBs used on series-compensated lines.
This value was deemed acceptable for Turkey’s 420 kV network because the CBs composed of
two breaking units could withstand the high TRV values exceeding IEC standard values.
3.3.2.4 Hydro-Québec experience

In the early 1990s, Hydro-Québec launched a vast program to fortify its 765 kV network. This
program involved installing 37 series capacitor banks for a total of 15,000 Mvars. The protection
scheme consisted of MOVs and a forced-triggered spark gap as described in Fig.3.3.11. The
MCOV protection level is 2.5 p.u. in accordance with the rated voltage of the capacitor. Prior to
this vast program, extensive studies were performed to evaluate the TRV level and switching
surges, taking into account the series capacitors. The results are reported in Ref. [31].
103
The study took into consideration numerous parameters that could impact the TRV level, such as
type, location and duration of fault as well as presence of SA, shunt reactor and opening resistors.
The highest TRVs (around 4.0 per unit) were obtained during interruption (by remote CB) of
ungrounded bi-phased faults near the capacitor bank. As reported by Iliceto et al., the fault
currents are relatively low for cases of high TRV (< 25% of CB rating). These high TRV values
were obtained when no shunt reactors or line surge arresters were connected to the network.
Simulation studies showed that without series capacitors, the worst TRVs were around 3.0 p.u.
during interruption of ungrounded faults.
Therefore, the study demonstrated that TRV values exceeded IEC standard values and some
mitigation measures were necessary to limit the TRV values to the rating of the existing circuit
breakers (2.8 p.u. based on 765 kV). Considering the use of 587 kV MOSAs at both ends of the
lines, the probability of exceeding the CB TRV rating (2.8 p.u.) was calculated by means of
statistical approach considering the fault type occurrence (three-phase grounded and ungrounded,
bi-phased ungrounded, etc.). With the presence of MOSAs, the risk of exceeding the TRV
withstand capabilities of the existing CB was then considered acceptable.
3.3.2.5 BC Hydro experience

BC Hydro has used series-compensated lines in its 500 kV grid since the early 1970s. The first
installations were protected by self-triggered bypass spark gaps. At the beginning of the 2000s,
new lines were constructed and the old series-compensation installations were refurbished by
adopting MOV bypass protection rated at 2.2 p.u. The capacitor banks for the new line were
installed in the middle of the line (330 km).
Extensive EMTP studies were performed to evaluate the level of TRV obtained under the worst
conditions (clearing three-phase ungrounded and grounded faults) with and without line surge
arresters and considering a protection level of 2.5 p.u. for the series capacitors. The detailed
results are reported in Ref. [30]. From these results, the maximum TRV obtained was 1.6 p.u. on
the source side. When faults were statistically applied exceeding the series capacitor banks, it
resulted in TRVs up to 4.8 p.u. and 3.8 p.u., respectively, for three-phase ungrounded and three-
phase grounded faults without line surge arresters. Considering the presence of surge arresters at
both ends of the line, the maximum TRV reached about 3.2 p.u., which is similar to the values
found by Hydro-Québec and Iliceto. Finally, line surge arresters with a 1.57 p.u. protection level
were used to guarantee a maximum TRV of 3.2 p.u.
As for other TRV studies for series-compensated installations, the short-circuit current associated
with the highest TRVs was relatively low (comparable to T10 and T30 test duties) and the time to
peak was quite long.
3.3.2.6 Solutions to reduce TRVs

All TRV studies on series-compensated lines clearly show TRV values far exceeding IEC
standard values when no mitigation measures are taken. Different solutions can be used to reduce
TRV stress on series-compensated line CBs:
• Line surge arresters at both ends: This can reduce peak TRV to about 3.2 p.u. considering
a 2.5 p.u. MCOV protection level for series capacitors and 1.7 p.u. for line surge arresters;
• MOVs in parallel with the interrupting units of CBs: This solution has been successfully
applied on Turkey’s 420 kV series-compensated network. The full simulation results are
104
presented in Ref. [29]. MOVs in parallel with circuit breaker chambers were applied for
both protection schemes of series capacitors, i.e. spark gaps and MOVs.
• Opening resistors: This solution is efficient only if used on CBs at both ends of the line.
With a resistor value of 800 ohm and insertion time of 16 ms, the maximum TRV
obtained from simulations on Hydro-Québec’s 765 kV network was reduced from 3.36 to
2.22 p.u. for three-phase ungrounded faults.[31]
• Fast protective device (FPD): Recent development of FPD for HV series capacitors has
become very attractive for reducing TRV during line fault clearing. This new device is
well documented in Refs. [33] and [34]. It was used for the first time on a pilot
installation at Kamouraska series capacitors in Hydro-Québec’s 315 kV network.
The FPD consists of a very fast switch and a fast mechanical switch that makes it possible to
bypass the capacitor bank within 1 ms (Figs. 3.3.13 and 3.3.14). Due to the high closing speed of
the FPD and its ability to be triggered for low series-capacitor voltages, it is possible to bypass
the series capacitor well in advance of the line CBs opening. The TRVs of series-compensated
line CBs are then comparable with the standard TRVs for uncompensated lines. This solution has
been operated successfully on a 315 kV line at Hydro-Québec since October 2003. The FPD
comprises a plasma switch in parallel with a very fast mechanical switch, as shown in Fig. 3.3.13.
A second installation is on the way for a 230 kV compensated line. It will also be considered for
future installation on 765 kV lines following the positive field experience of the pilot project at
315 kV and after proving its performance and reliability via extensive laboratory tests for its
application on the 765 kV network. [34]
Fig. 3.3.13 Fast protective device: plasma switch and fast mechanical switch
Fig. 3.3.14 Fast protective device installation
105
3.3.2.7 Effects on switching surges
The effect of series capacitors on switching surges is reported in Ref. [27] for a 345 kV series-
and shunt-compensated network. During a reclosing sequence (after fault clearing), very high
switching surges were observed due to the trapped charge on the capacitors (if the series
capacitors are not bypassed during the reclosing sequence). The total switching overvoltage could
reach 4.0 to 4.5 p.u., which is very severe even on a 345 kV system. At higher voltage levels, the
relative switching surge withstand of the equipment is usually less compared to lower voltages.
If shunt reactors are connected on the line, there will be a reduction of the trapped voltage on the
series capacitor due to oscillation between the shunt reactor and the series capacitor, although this
reduction will not be so significant due to the usually high quality of the shunt reactor (low loss).
However, since surge arresters normally protect the shunt reactors, the switching overvoltage
would be limited to the protection level of the arresters. If shunt reactors are not used, it is highly
recommended that surge arresters be installed on the line.
3.3.2.8 Shunt compensated lines and their effect on circuit breakers

recovery voltage and switching surges
It is well known that long lines at EHV are usually shunt-compensated to compensate for the
Mvars generated by the line under weak load conditions. Different approaches are used by
utilities for the way in which they operate their shunt reactors. Shunt reactors are used in
transmission OH lines for 2 main purposes:
• Switchable shunt reactors at busbars or tertiary windings of power transformers for
steady-state voltage control issues;
• Fixed shunt reactors at the ends of OH lines to deal with switching transient overvoltage
(TOV) requirements, related to line closing and/or opening operation (including fast
auto-reclosing issues)
During a reclosing sequence on the line, shunt reactors have a positive effect in reducing the
switching surges because the line charging will discharge through the shunt reactors before
reclosing the circuit breaker. The trapped charge during reclosing of the line will depend on the
fault location, the reclosing delay and the damping provided by the oscillating circuit (given by
the shunt reactor Q factor).
Shunt reactors also help to reduce TRV across the line circuit breakers especially during
interruption of non-earthed faults and during line charging current interruption because the line
will again discharge through the shunt reactors, leaving a reduced trapped charge on the line side
of the CB. Results from EMTP studies on Hydro-Québec’s 735 kV system (series- and shunt-
compensated lines) show an overvoltage reduction of about 10% during fault interruption when
shunt reactors are connected to the line. [31]
3.3.3 Characteristics of short-line faults
3.3.3.1 General
In IEC 62271-100, L90 and L75 are the type test duties for short-line fault (SLF) conditions such
as line faults that occur at a distance ranging from 100 m to several kilometres down the line. The
106
fault current is equal to 90 and 75%, respectively, of the rated short-circuit breaking current.
The severity of SLF test duties depends mainly on the rate of rise of recovery voltage, which is
determined by the line surge impedance and slope of the fault current.
The equivalent line surge impedance (Z) for each pole-to-clear is given by Eqs. (7), (8) and (9),
where Z0 is the zero-sequence surge impedance and Z1 is the positive-sequence surge impedance.
3 Z1 Z 0
Z first pole = (7)
Z1 + 2 Z 0
Z 1 (Z + 2 Z 0 )
Z sec ond pole = (8)
2 Z1 + Z 0
Z third pole =
(2 Z1 + Z 0 ) (9)
3
Various line surge impedance values for UHV systems are given in Table 3.3.7 as well as values
for EHV systems. It should be noted that in the case of three-phase line faults, the surge
impedance is highest for the third pole to clear.
3.3.3.2 Single-phase faults as the basis for rating

In IEC 62271-100, a single-phase test at phase-to-earth voltage is considered to cover all types of
short-line faults. This is supported by the following considerations:
• Line surge impedance relative to the last pole to clear is highest in the case of a three-
phase fault, and it is also the surge impedance for a single-phase fault. Therefore, the
RRRV obtained during a single-phase fault interruption covers all SLF conditions;
• A single-phase short-line fault test demonstrates an arcing window of (180°−dα) that
covers the requirements for all multi-phase fault cases in effectively-earthed and non-
effectively-earthed systems;
These considerations could also apply in the case of UHV. However, CIGRE WG A3.19 has
been discussing whether it is recommended that SLF test duties be performed as single-phase
with line characteristics as defined hereafter for single-phase faults.
3.3.3.3 Line surge impedance and bundle contraction

IEC 62271-100 and IEEE C37.04 stipulate standard line surge impedance of 450 Ω, considering
the completed bundle contraction of multi-bundle conductors due to magnetic attraction caused
by a large fault current through the conductor, although the equivalent line surge impedance
calculated without bundle contraction is less than 360 Ω (see IEEE C37.011-2005 and the values
in Table 3 for system operating voltage of 500 and 765 kV).
Table 3.3.7 summarizes analytical line surge impedance under both normal conditions and
completed bundle contraction for several transmission lines. The equivalent line surge impedance
of multi-bundle conductors is calculated as 250–311 Ω for 4-8 bundle conductors under normal
conditions without taking into account bundle contraction. These values are assessed based on the
bundle and tower configurations illustrated in Fig. 3.3.15 with a constant parameter line model
supported by EMTP. The line surge impedance tends to be larger if a transmission line is high
from ground.
107
Table 3.3.7 Surge impedance of transmission lines
Equivalent surge
Highest Conductor Number Conditions Z0 Z1 impedance (ohm)
voltage size of
(kV) (mm2) conductor
(TRV frequency) (ohm) (ohm) 1st 2nd 3rd
pole pole pole
Normal conduction
550 (60 kHz) 444 226 270 281 299
410 6
(Japan) Bundle contraction
(60 kHz) 580 355 408 417 430
800 Normal conduction
(27.5 kHz) 403 254 290 296 304
(South 428 6
Bundle contraction
Africa) 509 359 398 403 409
(27.5 kHz)
Normal conduction
1,050 406 210 250 260 275
520 8 (26.2 kHz)
(Italy) Bundle contraction
(26.2 kHz) 532 343 389 396 406
Normal conduction
1,100 (25 kHz) 476 228 276 289 311
810 8
(Japan) Bundle contraction
(25 kHz) 595 339 396 407 424
TRV frequency can be obtained in case of 800 kV South Africa project as follows.
Impedance at the source side: 800kV/√3/40kA/(2×3.14×50Hz)=36.78mH, Impedance at the line side:
36.78mH×1/9=4.09mH for L90, TRV frequency: 450 ohm/(4×4.09mH)=27.5kHz for L90
19m
15.5m
South Africa Italy

800kV 1050kV 16m
52m
13.75m 13.75m 16.5m
15.8m 15.8m 0.40m

810 mm sq.
8 conductors
100.75m
0.32m 0.45m
72.5m
90m
642 mm sq. 510 mm sq.

50m
39m
6 conductors 8 conductors
120m
45m
32m
Japan
1100kV
Fig.3.3.15 Collision time of multi-bundle conductors
Table 3.3.7 shows that line surge impedance for the third pole to clear increases several
percentage points compared with the values for the first pole to clear. If bundle contraction is
assumed, line surge impedance for the third pole to clear increases by 36-56% and approaches the
standard value of 450 Ω. The change in line surge impedance depends on the multi-bundle
conductor design (materials, cross section, spacer distance and spacer size) and the line tension in
addition to the fault current. Several studies have reported the time required for bundle collision.
Table.3.3.8 Collision time of multi-bundle conductors
Country Size Number of Span Sub-conductor Initial Breaking Time to bundle Time to bundle
(mm2) Conductor (m) distance (mm) tension (kN) current (kA) collision, Cal. (sec) collision, Exp. (sec)
Italy 520 8 --- 450 --- 50.0 0.166 ---
410 6 45 400 34 40.8 0.140 0.110
Japan
410 6 45 400 34 53.2 0.106 0.080
Japan 810 4 45 550 49 40.8 0.148 0.124
810 4 45 550 49 53.2 0.114 0.090
108
810 8 50 400 53 50.0 0.202 ---
810 8 45 400 60 50.0 0.149 ---
For example, CIGRE WG 13.01 [35] indicates that twin conductors, in a 686 mm2 cross section
carrying 40 kA, collided at 50 ms after a fault occurrence. Table 3.3.8 summarizes the collision
times of various multi-bundle conductors surveyed in Japan, which collided at 100-200 ms when
carrying 30-50 kA through the line.
The analytical times to bundle collision are in relatively good agreement with the experimental
data. The analysis also provides the time to bundle collision in the case of 8 conductors, 810 mm2,
calculated at a breaking current of 50 kA with a DC time constant of 150 ms.
Figure 3.3.16 shows a typical result of a change in sub-conductor distance as well as line surge
impedance of the third pole to clear. The results are calculated for 8 conductors, 810 mm2, at a
breaking current of 50 kA with a DC time constant of 150 ms. The electromagnetic force
generated between sub-conductors increases with an increase in current and decrease in sub-
conductor distance. The sub-conductors collided at 149 ms after the fault occurred. Line surge
impedance of the third pole to clear increases and gradually approaches the standard value at the
completed bundle contraction. However, impedance is estimated to be less than 350 Ω for about
5 cycles (100 ms) after the fault initiation.
20
Current per conductor [kA] τ=150ms
10
0
810mm sq.
800 8 conductors
Electro-magnetic force [N/m]
600
400
400
200
0
Distance between sub-conductors [mm]

400
300
Bundle collision (0.15sec)
200 Fault occurrence
100
0
Line surge impedance [ohm]
450 Standard line surge impedance : 450 ohm
400
350
300
Line surge impedance of 3rd pole to clear is
250 less than 350 ohm for 5 cycles after the fault.
200
0.0 0.02 0.04 0.06 0.08 0.10 0.12 0.14 time [sec]
Fig.3.3.16 Collision time of multi-bundle
UHV transmission lines normally employ multi-bundle conductors with a larger cross section for
reduced radio interference and corona noise, which results in a longer time required to obtain
bundle collision. Since bundle contraction is not supposed to be completed at the time the fault
current is interrupted (50–80 ms after fault initiation), line surge impedance for L90 and L75
duties could be reduced for rated voltages exceeding 800 kV.
109
3.3.3.4 Peak factor “k”
The line-side contribution to TRV is equal to the instantaneous value of the line-side voltage at
current zero multiplied by the peak factor “k” (“d” in ANSI/IEEE standards).
The peak factor is defined in Table 4 of IEC 62271-100. Its value is independent of the rated
voltage and is equal to 1.6. This value is also considered suitable for UHV applications, taking
into account that the initial TRV peak is of secondary importance for SLF interruption, compared
to the two more influential parameters: RRRV and di/dt. A conservative estimate of the peak
factor is given by Equation (A-29) in IEEE C37.011-2005, shown below as Equation (10):
Z0
peak factor = 0.4 × (2 + ) (10)
Z1
Using the values of Z0 and Z1 given in Table 3.3.9 for 1000 kV networks in Japan and Italy, and
assuming that there is no bundle contraction, k is equal to 1.64 and 1.57, respectively. A peak
factor of 1.6 can be recommended for UHV applications; however, WG A3.22 will investigate
further whether the formula and the damping factor are applicable to UHV.
3.3.4 DC time constant
3.3.4.1 General
Large power generators and large-capacity power transformers lead to a higher X/R ratio, which
contributes to the increase in DC time constant in fault currents. EHV and UHV transmission
lines employ multi-bundle large-diameter conductors in order to reduce corona noise as well as
increase transmission capacity. For example, UHV transmission lines use 8 conductors, 400–
810 mm2, depending on the allowable level of corona noise. The application of multi-bundle
conductors also increases the DC time constant (τ = L/R) because the reduction in resistance (R)
exceeds the reduction in inductance (L).
Table 3.3.9 summarizes analytical results of DC time constants calculated using various tower
designs with different multi-bundle conductors used in different projects. The constants were
obtained using the ratio of positive-sequence inductance to positive-sequence resistance of the
lines. In IEC 62271-100, a special-case time constant was added to the standard value of 45 ms,
based on a survey conducted by CIGRE WG 13.04. [35] The special-case time constant is 75 ms
for a rated voltage of 550 kV and above, which corresponds to the medium value of constants
surveyed for 800 kV lines.
Table 3.3.9 DC time constants of short-circuit currents in EHV / UHV transmission line
Conductors
Maximum voltage 2 DC time constant (ms)
Size (mm ) Bundle
765 kV (Canada) 686 4 75
800 kV (USA) 572 6 89
800 kV (South Africa) 428 6 67
800 kV (Brazil) 603 4 88
800 kV (China) 400 6 75
1,200 kV (Russian) 400 8 91
1,050 kV (Italy) 520 8 100
1,100 kV (Japan) 810 8 150
1,100 kV (China) 500 8 120
110
It is expected that the DC time constants for UHV systems will be higher than the standard values
due to the use of multi-bundle conductors with larger diameter in addition to the existence of
large power generators and large-capacity power transformers. The influence of the high DC
component on test-duty T100a was evaluated by the energy of the last major loop before the
interruption, slope of current at the time of interruption and TRV characteristics. The results do
not show any significant difference when the constant exceeds about 120 ms. [36] Therefore, it is
advisable to use a higher special-case time constant of about 120 ms for rated voltages exceeding
800 kV.
3.3.4.2 DC time constant in UHV system in Japan

Network losses are smaller in UHV systems compared to 800/550 kV or lower voltage systems,
since multi-bundle conductors are frequently used for transmission lines. In addition, large power
sources tend to be located close to UHV networks, and large-capacity transformer banks are
connected to the networks. These characteristics lead to a higher X/R ratio and subsequently, a
longer DC time constant in fault currents.
On the other hand, the influence on interruption performance saturates as the time constant
becomes longer than about 100-120 ms. Figure 3.3.17 shows an example rate of decrease of
interrupting performance [36] assuming that interrupting capability is determined by the rms
value of the breaking current.
100
Interrupting capability (%)
90
Guideline of interrupting capability
80
Rate of rms value of asymmetrical current to the critical interrupting current
70
CB Type A B C
Interruption success
Interruption failure
0 45 60 80 100 120 150

Time constant of DC component in fault current (ms)
Fig.3.3.17 Influence of time constants of DC component on interrupting capability
In Japan, estimation was conducted using actual electric parameters of each network component.
Table 3.3.10 shows typical values of electric parameters of each system component. DC time
constant is estimated by the X/R separation method.
Table 3.3.10 Typical values of each component for estimation of DC time constant
Equipment Resistance Reactance Time constant
R (%) X (%) X/R (ms)
Generator 0.2-0.3 23 240 ~ 370
Ta=L2/R=0.2-0.3s
X2=Xd”=25%,Xd’=30% (at 1.3GVA)
500kV Transformer 0.1-0l.3 8-15 150 ~ 240 Xt=12-15%,R=0.2-0.3% (at 1.2-1.5GVA)
UHV Transformer 0.067 6 260 Xt=18%, R=0.2% (at 3.0GVA)
UHV Transmission Line
0.0025 0.125 150
(810mm2, *8 bundle)
UHV Transmission Line
0.0035 0.125 110
(610mm2 *8 bundle)
500kV Transmission Line 0.0025 0.05 70 810mm2 *4 bundle
Note: Values are 1000MVA base
111
The 500 kV transmission lines that supply fault currents to the UHV system are shorter in length,
and the DC time constant tends to be longer because they have the shortest time constant among
those in the components considered in Table 3.3.10. Based on the estimation of a simple model
network in which the DC time constant ranges from 100 to around 200 ms (as shown in Table
3.3.11), and the detailed analysis with actual UHV networks in the first and second stage in
which the DC time constant ranges from 100 to around 150 ms (as shown in Table 3.3.12),
therefore a specification of 150 ms was decided.
Table 3.3.11 Estimation of simple model network (Short circuit current less than 40kA)
550kV Line UHV Line
(1) (2) (3)

Model A B
The influence of 500kV Transmission The influence of 500kV Transmission
lines is comparatively small. lines is comparatively large.
Max. Short Circuit current (kA) (1): 40, (2): 35, (3): 30 (1): 40, (2): 35, (3): 30
Conditions Power source: 50 GVA Power source: 100 GVA
500kV Line: 20 km*5 route 500kV Line: 70 km*5 route
UHV Transformer: 60 GVA UHV Transformer: 60 GVA
UHV Line: 40 km (810sq* 8bundle) UHV Line: 40 km (610sq* 8 bundle)
Reactance’s (X) (1): 1.41, (2): 1.66, (3): 1.91 (1): 1.42, (2): 1.67, (3): 1.92
Resistance’s (R) (1): 0.023, (2): 0.028, (3): 0.033 (1): 0.047, (2): 0.054, (3): 0.061
DC Time Constant=(X/R/(2*pai*f) (1): 195 ms, (2): 189 ms, (3): 184 ms (1): 96 ms, (2): 98 ms, (3): 100 ms
Table 3.3.12 Estimation of time constant of DC component in actual UHV networks

1100 kV Transmission Line
UHV Network in Japan**

1100 kV Transmission Line 1100 kV Transmission Line
D C A D C B A
550 kV Grid 550 kV Grid
E E
Stage First stage Second stage

Time constants of A 156 149-165
DC component in B - 117-144
fault currents at C 149 111-142
different D 153 111-145
Substations (ms) E 154 103-135
** 500kV and lower network is also modeled.
3.3.4.3 Time constant for T100a and UHV applications

Different time constants of DC decrement have been presented for UHV networks. For example,
it is 124 ms in China and 150 ms in Japan. For the purpose of standardisation, a compromised
solution should be recommended as the preferred or rated value. For this purpose, consideration
112
must be given to the three parameters that influence the interruption of asymmetrical currents
required in T100a test duty:
• Energy of the last major loop before interruption,
• Slope of current (di/dt) at the time of interruption,
• TRV characteristics (peak and rate-of-rise).
According to Edition 2.0 of IEC 62271-100 (see 6.102.10.2.1.2 in document 17A/768/CDV), it is
possible to compare T100a breaking capabilities using different time constants if the duration and
peak value of the major loop to be interrupted do not differ by more than 10%. In that case, the
energy in the last major loop can be assumed to be proportional to the product of the peak current
multiplied by the duration of the major loop considering a constant arc voltage, and equivalence
between asymmetrical currents can be achieved as defined by IEC. [37]
In practical cases, it can be assumed that interruption occurs after the third major loop of a
current established with 100% asymmetry. It can be seen in Fig. 3.3.18 and Tables 3.3.14 and 15
that when time constants of 120 and 150 ms are considered, both the loop duration and amplitude
of the last loop of current before interruption differ by less than 10%. Therefore, the IEC law of
equivalence can be applied and a test based on a time constant of 120 ms can be used to
demonstrate breaking capability with a time constant of 150 ms.
Figure 3.3.18 Evolution of asymmetrical current in a network of time constants

45ms, 100 ms, 120 ms and 150 ms (f = 50 Hz)
Table 3.3.14 Duration of the third major loop as function of the time constant (f = 50 Hz)
Time constant From (ms) To (ms) Duration Relative value of loop duration
45 ms 43.65 55.8 12.15 85.9%
100 ms 42.65 56.8 14.15 100%
120 ms 42.35 57.15 14.8 104.6%
150 ms 42.25 57.35 15.1 106.7%
Table 3.3.15 Amplitude of the third major loop and product “Peak x loop duration”
as function of the time constant of the network (f = 50 Hz)
Time constant Third peak (p.u.) Third peak (%) Relative value of Peak x loop duration
45 ms 1.33 82.6% 70.9%
100 ms 1.61 100% 100%
120 ms 1.66 103.3% 108.1%
150 ms 1.72 106.9% 114.1%
* 1 p.u. = peak value of symmetrical current
Taking into account the criteria introduced in Edition 2.0 of IEC 62271-100, a test performed
with (and not less than during actual tests) the parameters of the major loop specified for L/R
113
=120 ms would be acceptable for demonstrating T100a with L/R = 150 ms. Alternatively, the
r.m.s. value of the test current could be increased to meet the relative value of the peak current ×
loop duration specified for 150 ms.
It should also be noted that a test based on a time constant of 120 ms leads to di/dt at current
interruption and TRV characteristics that are both more severe than those required for a time
constant of 150 ms. Moreover, it must be taken into account that during testing in high-power
laboratories, it may be difficult to meet the long duration of a major loop of current.
In view of these considerations, and taking into account that the three parameters for T100a
interruption based on a time constant of 150 ms can be covered by the previously described test
based on a time constant of 120 ms, it is recommended that 120 ms be selected as the time
constant for networks with a rated voltage higher than 800 kV.
3.3.5 TRV with/without opening resistors

3.3.5.1 General
An opening resistor scheme can reduce the switching overvoltages and also have a considerable
effect to mitigate the switching duties of a circuit breaker. In the UHV system in Japan, the
opening resistor method with a 700 ohms resistance is used to realise a reduced insulation level
and compact facilities. The effects of opening resistor on TRV include the following aspects.
a) Damping of transient oscillation of TRV amplitude
The opening resistor reduces the amplitude of TRV oscillations. It can considerably mitigate the
TRV peak as well as the initial rate of rise of TRV for testing duties of a terminal fault, a short
line fault, and capacitive current switching.
b) Phase angle shift of breaking current
The opening resistor is inserted into the original circuit, which causes the phase angle shift of the
current though the resistor contacts. This phase shift can significantly reduce the severity of the
switching duties for capacitive current switching and out-of-phase conditions. For example, the
TRV for capacitive current switching approaches a sine waveform instead of (1-cosine)
waveform. The TRV behaviour in out-of-phase condition is discussed in the clause 3.3.6 in detail.
3.3.5.2 Effect of opening resistors on TRV

The Effect of opening resistors on TRV in the Japan’s UHV system was calculated by EMTP. I-4
model and II-1 model (Simplified model) in Figure 3.3.19 (a) and (b) are used for analysis.
D
50 km
B E A E A
40 km 210 km 210 km
138 km
B
Figure 3.3.19 (a) Model I-4 Figure 3.3.19 (b) Model II-1
The resistance value of 700 ohms for the closing and opening resistors was selected to suppress
the switching surges as described in the section 2.9.1.2. Here, the resistance of the opening
114
resistor value was changed from 500 to 750 ohms in order to study the effect on TRV reduction.
A summary of this study is shown in Figure 3.3.20, and indicates that the effect of varying
opening resistors in this range is quite small. The characteristics of the surge arresters are given in
Table 3.3.6.
2000 TLF 2000 T30 2000 T60 2000 T100
TRV peak (kV)
Without resistor Without resistor

1500 1500 Without resistor 1500 1500
With opening resistor Without resistor With opening resistor

1000 With opening resistor 1000 1000 1000
With opening resistor
500 500 500 500
Rate of rise of TRV peak (kV/µs)
0 0 0 0
Without resistor
10 10 10 10 Without resistor
Rate of rise of TRV of T10 for IEC standard
Rate of rise of TRV of T30 for IEC standard Without resistor With opening resistor
5 5 5 5
With opening resistor Without resistor Rate of rise of TRV of T60 for IEC standard
Rate of rise of TRV of T100 for IEC standard
With opening resistor With opening resistor
0 0 0 0
750 650 500 (ohm) 750 650 500 (ohm) 750 650 500 (ohm) 750 650 500 (ohm)
Resistance of opening resistor Resistance of opening resistor Resistance of opening resistor Resistance of opening resistor
Fig. 3.3.20 Effect of opening resistor value on TRV parameters
3.3.5.3 Category of fault current and modes

Fault current values are categorized based on fault mode and reveal that the range of fault current
depends on the fault mode. The results are shown in Fig. 3.3.21.
Mode1 Mode2 Mode3 Mode4
Mode1:
Transformer limited fault corresponding to T10
Mode2: Long line fault corresponding to T30
TLF LLF BTF BTF Mode3: Bus terminal fault corresponding to T60
Mode4: Bus terminal fault corresponding to T100
T10 T30 T60 T100

-8 kA 9-16 kA 20-30 kA 30-40 kA
0-16% 18-32% 40-60% 60-80%
0 10 kA 20 kA 30 kA 40 kA 50 kA Breaking current
Note: Japan's UHV system is designed for fault current of 50kA (100%) for the final stage.
Figure 3.3.21 Fault current and modes
3.3.5.4 TRV for main interrupters with/without opening resistors

1) Terminal fault
In Japan’s UHV system, the opening resistors as well as the surge arresters are considered in the
early stage of the system study. The TRV specifications were determined based on the
application of the opening resistors combined with the surge arresters. The TRV without opening
resistors and with surge arresters in addition to the case without the opening resistors nor surge
115
arresters are not officially considered. In these two cases, only the maximum values are indicated
in parentheses for reference.
(A) U1/t1 (dv/dt); Various calculation results for fault clearing are shown in Fig. 3.3.22. The
results include the following three cases: with opening resistors (R) and surge arresters (SA);
without opening resistors and with surge arresters; and without opening resistors or surge
arresters. The values are grouped in pairs. The first is with opening resistors and the other is
without opening resistors. By adding opening resistors, the U1/t1 values are reduced, especially
for regions with a smaller current.
The rate of rise is greatly reduced by opening resistors for all duties. Opening resistors are
especially effective for smaller current values due to the suppression of transient phenomena.
Table 3.3.16 shows the maximum calculated values (in parentheses) and the specified values for
the case with opening resistors and surge arresters with U1/t1 values for Japan’s UHV system.
The results show that the values with opening resistors are lower than the extrapolated standard
values for the reduction of the fault current by using resistors for the main interrupters.
20
Opening
MOSA T60-T100 TLF LLF
Rate of rise of TRV u1/t1 (kV/µs)
resistor
15 Without Without
With Without
With With
10
Rate of raise of TRV of existing 550kV standards
Effect of opening resistor on RRRV
reduction especially for TLF
5
0
0 10 20 30 40 50
Breaking current (kA)
Figure 3.3.22 Calculation results for U1/t1 of Japanese UHV system
Table 3.3.16 Specified and calculated values of U1/t1 of Japan’s UHV system
System condition T10 (16%; 8 kA) T30 (30%;15 kA) T60 (60%; 30 kA) T100 (80%; 40kA)
Without R and without SA (14.5 kV/µs) (4.3 kV/µs) (5.5 kV/µs) (1.4 kV/µs)
Without R and with SA (15.5 kV/µs) (4.3 kV/µs) (5.5 kV/µs) (1.4 kV/µs)
With R and With SA* 3 kV/µs 3 kV/µs 3 kV/µs 2 kV/µs
(2.3 kV/µs) (2.5 kV/µs) (4.6 kV/us) (1.2 kV/µs)
Existing 550 kV standards 7 kV/µs 5 kV/µs 3 kV/µs 2 kV/µs
Note; * Maximum calculated values are indicated in ( ).
(B) U1 (first peak voltage); Various calculation results for fault clearing are shown in Fig.
3.3.23. The majority of U1 values are covered by the standard values extrapolated from IEC
62271-100 (2002) (1.3 × 1 p.u. = (√2 × 1100 kV / √3) = 1167 kV), but they are not covered by
the latest standards in 2006 (1.3 × 1 p.u. × 0.75 = 876 kV). The values with opening resistors are
lower than those without opening resistors, because t1 (corresponding to U1) is defined by the
timing of the arrival of the reflection wave from the line end, which does not depend on whether
there are opening resistors or not, but the U1/t1 value is reduced by the use of opening resistors.
116
U1, which are defined based on a first-pole-to-clear factor of 1.1 with the use of surge arresters.
2500
Opening
First reference voltage U1 (kV)

MOSA
resistor
2000 Without Without
With Without
With With
1500
1000
500
0
20 0
30 10 40 50
Figure 3.3.23 Calculation results of U1 for Japanese UHV system
Table 3.3.17 Specified and calculated values of U1 of Japan’s UHV system

System condition T30 (15kA) T60 (30kA) T100 (40kA)
Without R and without SA (1278 kV) (1110 kV) (1047 kV)
Without R and with SA (1278 kV ) (922 kV) (830kV)
With R and with SA 988 kV (FPCF=1.1) 988 kV (FPCF=1.1) 988kV (FPCF=1.1)
(833kV) (857 kV) (1012 kV)
IEC value (extrapolated) --- 876 kV(FPCF=1.3)
Note: Extrapolated IEC value is estimated as 0.75x1.3x1p.u. (898kV)
Maximum calculated values are shown in parentheses
(C) Uc (TRV peak voltage); Various calculation results for fault clearing are shown in Fig.
3.3.24. The Uc is reduced by using opening resistors, especially in lower-current regions due to
the high impedance ratio of transformers and suppression of transient phenomena by the resistors.
The data with opening resistors is covered by the extrapolated IEC standard values even when the
first-pole-to-clear factor is 1.1. In the case of LLF, Uc without opening resistors increases even
with the use of surge arresters, because although surge arresters reduce the overvoltage to ground,
the TRV between CBs results in a difference in voltage at both sides of the CB.
Uc. In the case without opening resistors (1st and 2nd rows), the Uc values are suppressed by surge
arresters except for T30 duty, which corresponds to LLF conditions. In the case with opening
resistors, Uc values are greatly reduced and specified values are expressed as 1385 kV (1.1×1.4)
for all ranges from T10 to T100.
117
2500
TRV peak value Uc (kV)

2000 Twice TRV peak values of existing 550kV standards
FPCF=1.3
FPCF=1.2
1500 FPCF=1.1
1000
MOSA Opening resistor T60-T100 TLF LLF
Effect of opening resistor Without Without
500
on TRV reduction With Without
With With
0
20 0 30 1040 50
Figure 3.3.24 Calculation results of Uc for Japan’s UHV system
Table 3.3.18 Specified and calculated values of Uc of Japan’s UHV system

System condition T10(8kA) T30(15kA) T60(30kA) T100(40kA)
Without R and without SA (1528 kV) (1879 kV) (1586 kV ) (1571 kV)
Without R and with SA (1478 kV) (1879 kV) (1303 kV) (1337 kV)
With R and with SA 1385 kV: 1.1x1.4 1385 kV: 1.1x1.4 1385 kV: 1.1x1.4 1385 kV:1.1x1.4
(890 kV) (1217 kV) (1157 kV) (1309 kV)
IEC value (extrapolated) 1786 kV: 1.3x1.7x0.9 1752 kV (1.3x1.5) 1634 kV: 1.3x1.4
Note: The maximum calculated values are shown as ( ) for reference.
2) Short line fault (SLF) interruption

SLF interruption is a phenomenon where a high rate of rise of recovery voltage occurs at the
initial stage of interruption. The initial rise of TRV is reduced by about 60% by opening resistor.
3) Capacitive current switching
Capacitive current switching duties become most severe with no load switching, but with opening
resistors, the recovery voltages between the interrupters are divided by the ratio of resistance.
This significantly mitigates the capacitive current switching duties.
Fig.3.3.25 TRV for small capacitive switching with/ without opening resistor
118
Fig.3.3.26 TRV for small capacitive switching with/ without opening resistor
3.3.5.5 TRV for resistor interrupter

1) Terminal fault
1. U1/t1(dv/dt): Figure 3.3.27 shows the TRV for the resistor contact of the interrupter, and the
U1/t1 is determined by T10 duty. Table 3.3.19 summarizes the TRV duty for the resistor contact.
Table 3.3.19 Summary of TRV condition for resistor contact
Basic specification Interrupting current U1 U1/t1 t1 Uc t2
With opening resistor 750 ohms 1000A 250kV 2.3 kV/µs 109 µs 1200 kV 2200 µs
Resit er Cont act (t 10) Resist erCont act (T30- T100)
T10 T30- T100

2.3kV/ us- 250kV 2.3kV/ us- 250kV，1200kV- 2200us
2kV/ us 1200kV- 2200us
500 1400
1200
400
Voltage (kV)
1000
Voltage (kV)
300
800
200 600
100 400
200
0
0
0 50 100 150 200 0.00 1.00 2.00 3.00 4.00 5.00
time (micro sec) Time (ms)
Fig. 3.3.27 TRV shape of resistor contact with opening resistor
2) Capacitive current switching

Because resistance is inserted in the circuit after the main circuit is interrupted, the TRV wave
shape varies depending on the line length. In the case of a long line (210km), the TRV wave
shape becomes [(cos θ -cos( ω t + θ ), where θ =tan-1 ω CR).
119
Figure 3.3.28 Four TRV parameters for small capacitive switching duty for resistor interrupter
The peak value decreases slightly compared to that of (1- cos ω t), however, the initial part of the
rate of rise of TRV increases steeply (See Figure 3.3.29).
50 Hz
2
1 cos
Basic condition
cos cos
Long line(210km) condition
10_
Figure 3.3.29 TRV wave shape of line-charging switching for resistor interrupter
3.3.5.6 Evaluation of switching duties with opening resistors

1) Switching duties of main interrupter with opening resistors
Theoretically, annex F of the IEC standards proposed the following approach to evaluate the
switching duties for SLF and BTF. Let the inherent TRV be V(t).
V(t)＝（u1/t1）×t （0<t<t1） (11)

(Uc-u1)/(t2-t1)×(t-t1)＋u1 （t1<t<t2） (12)
Uc （t2<t） (13)
Next, the current will be expressed as;
I(t)＝√２×I rated×sinωt (13)
The network impedance without the opening resistors is given by;
Z(t)＝V(t) / I(t) (14)
The equivalent network impedance with the opening resistor will be theoretically given by;
Zr(t)＝(R×Z(t))/ (R+Z(t)) (15)
Then the switching duty with the opening resistor will be theoretically as follows;
120
Vr＝I(t)×Zr(t) (16)
2) Switching duties of resistor interrupter with opening resistors

The interrupting current for resistor interruption is different from that for the inherent one.
Let the inherent interrupting current be I(t), for example, then usually I(t)=E*sin(wt), where E(t)
is the system operating voltage.
Assuming the network impedance at the circuit breaker before interruption is Zo(t)
Zo(t)＝E(t) / I(t) (17)
When the opening resistor is inserted into the network impedance, the new impedance Zrr will be,
Zrr(t)＝R+Z(t) (18)
Then the interrupting current through the resistor contacts will be,
Irr(t)＝E(t) / Z(t) (19)
Finally, the switching duty for the resistor unit will become,
Vrr(t)=Z(t) ×Irr(t) (20)
3.3.6 Phenomena related to out-of-phase

3.3.6.1 General
Interconnections at the EHV level are increasingly used for multiple powerflows between regions
and between countries. Such large powerflows require substantial phase shift (voltage angle) to
transmit energy in an AC system. Even in a meshed network such as the interconnected European
420 kV system, substantial phase angles can occur between the northern and southern parts of the
grid. Angles as large as 60º are no exception and the UCTE has taken the decision to install an
online phasor measurement system to monitor stability limits within the European grid. In Brazil,
the phasor angle difference along the 800 kV transmission system from Itaipu to Sao Paolo (triple
circuit, ~1000 km, 60 Hz, series compensated) can reach values of 100º or more under full-power
operation. When the systems with such large phasor angles are split up into two sections under
stable conditions, the dynamic out-of-phase angles will inherently be large as well. Out-of-phase
angles as large as 180º are therefore considered realistic. Generally utilities that exploit long-
distance transmission systems specify out-of-phase angles as large as 120, 150 and 180º.
Similarly, it can be expected that UHV grids where regions with a large surplus of generated
power are interconnected with regions with a large load demand, will show substantial angles
between the voltages at both sides. Whether these angles will show up under circumstances of
splitting up the system depends to a large extent on the number of parallel branches and the
meshed structure of the UHV network. Although large out-of-phase angles may occur in the
heavily meshed European network, the probability of out-of-phase conditions with a large angle
is rather low. UHV networks, however, tend to be simple and far from meshed. Moreover, the
length of OH lines is usually long, so that the probability of occurrence of large out-of-phase
angles in general cannot be ignored.
121
Breaking the current under out-of-phase conditions with large out-of-phase angles (roughly more
than 90º) deserves some attention, as the dielectric TRV stress may be beyond the capabilities of
the involved circuit breakers. In the IEEE and IEC Standards that cover voltages up to and
including 800 kV, specified recovery voltages under out-of-phase conditions are applicable for
out-of-phase angles as large as 105º (see Clause 6.110.3 of IEC doc 17A/768/CDV). Looking at
the peak values of the recovery voltage, it may be that only angles less than 105º are covered.
Apart from circuit breakers, other equipment may be stressed by out-of-phase current that may be
larger than the short-circuit current contribution of that specific branch in the network.
In order to prevent the occurrence of large out-of-phase angles and/or overstressing of the circuit
breakers, utilities may apply special protection schemes. This is quite normal for large generating
plants, but less common for interconnections at the transmission level. The most familiar
protection systems are the power swing relay and the out-of-phase relay, which may be separate
devices or may be functions in an advanced (digital) distance/zone protection relay. While the
power swing relay detects certain slow changes in complex impedance as measured by the relay,
the out-of-phase relay continuously monitors the voltage angle between the line’s ends.
The power swing relay acts on large angle variations, but small angles, by giving a trip command
(or an alarm, but then it acts as a monitor rather than a protection). The relay offers an alternative
to block tripping for operating areas where there is some question as to whether tripping is
necessary or where tripping must be prevented. The setting of the angle variation threshold for
blocking is based on extensive computer simulations of the network under several dynamic
loading conditions.
The out-of-phase relay blocks the tripping command of the conventional protection relay
functions (distance/zone protection) when a certain threshold is passed by the out-of-phase angle.
The block will last until the out-of-phase angle decreases to a value where the circuit breaker is
able to clear the current.
Some utilities are in favor of such protection functions, but they are not unanimous about the
settings of the devices. Other utilities feel that such functions interfere with the reliability of the
main protection systems. Of course, network topology and the utility’s experience with out-of-
phase conditions play an important role in the implemented policy. For instance, Brazil’s 800 kV
system employs out-of-phase blocking relays, but Canada’s system has no protection device to
block the circuit breakers in the case of large out-of-phase angles. In India’s extended 420 kV
network, out-of-phase protection has not been applied nor is it foreseen for the emerging 800 kV
network.
In Japan, the out-of-phase protection is performed by two stages. The first stage is to prevent the
occurrence of out-of-phase conditions, which may result in large area blackouts. It caries out
online forecast calculations of the phase angle difference between a generator internal voltage
and the main system internal voltage; examples of two cases leading to stable conditions and
three cases leading to unstable conditions after a fault occurrence are given in figure 3.3.30.
When the predicted phase angle difference exceeds a pre-determined threshold value, due to an
electrical disturbance such as a short-circuit, some generators are tripped so that the stability of
the entire network is secured. About 200 to 300 ms before the phase angle reaches the threshold
value, trip signals are sent out from a central system protection unit. By offline studies, the
threshold values have been determined for each network condition (generally above 100°). The
second stage is to split up the network in case the fore-mentioned protection system does not
operate properly. Protection relays detect the out-of-phase condition by measuring the phase
122
angle difference between the busbar voltages at both sides of a transmission line or a transformer.
When the electrical centre of the out-of-phase condition is within this OH-line or this
transformer, and the phase angle approaches and exceeds 180°, the relay will give a tripping
command. (Much smaller phase angle differences than 180° could lead to unnecessary trips as
the power swing may settle down to stable conditions; see figure 3.3.30.) The out-of-phase
current is interrupted approximately 45 to 70 ms after the detection of phase opposition (180°).
As the slip interval between the generators at both sides is about several hundreds to 2000 ms per
turn, there is a certain possibility of interruption under a condition close to 180°.
250
δο 0) =32.5 degrees
Case 1 (δο
200
δ 0) =45 degrees
Case 2 (δ
δ (electrical degrees)
δ 0) =55 degrees
Case 3 (δ
150
100
50 δ 0) =32.5 degrees, Fault cleared at 4.5 cycle

Case 4 (δ
δ 0) =32.5 degrees, Fault cleared at 3 cycle

Case 5 (δ
0 0.1 0.2 0.3 0.4 0.5

Time elapsed from a fault occurrence (sec)
Note: Curves of cases 1, 4 and 5 quoted from "Element of power system analysis, second edition", written by
W. D. Stevenson. Jr., McGraw-Hill Book Company Inc.
Figure 3.3.30 Example of power swing curves from [38] (cases 2 and 3 are added)
Specifications for TRV at out-of-phase switching at the UHV level are compared in Ref. [39]. In
Japan, Russia and China, a voltage factor of 2.0 is used, resulting in a recovery voltage of 2 p.u.
This voltage factor is based on an out-of-phase angle of 180º, under conditions of a first-pole-to-
clear factor close to one. In Italy, based on an out-of-phase angle of only 90º, the voltage factor is
1.4, probably with a low first-pole-to-clear factor as well. The applied voltage for making the out-
of-phase current is 1.41 p.u. in Japan, Russia and China, suggesting an angle of 90º, while in Italy
a voltage factor of 1.0 p.u. is used (60º). Peak TRV value is 2.5 p.u. in Japan and China, with an
amplitude factor of 1.25, common to IEC. In Russia, the peak value is 3.0 p.u., suggesting an
amplitude factor of 1.5. In Italy, the peak value is 2.21 p.u., which leads to an amplitude factor of
1.58. RRRV is specified as 1.54 kV/µs in China, as in the IEC Standard. In Italy, it is 3.0 kV/µs
and in Japan, 6.0 kV/µs. [40]
The application of MOSAs can suppress the TRV peak values for terminal faults, resulting in
reduction in the amplitude factor of TRV for terminal faults, TLF (transformer limited or
secondary faults) and out-of-phase. However, it may not lead to sufficient reduction for LLF
(long-line faults) where TRV is generated at both the source and line sides of the breaker
terminals. In addition to MOSAs, the application of opening resistors will lead to considerably
lower TRV peak values and RRRV values, as shown in Fig. 3.3.31.
123
3500
TRV peak without MOSA

3000
Peak voltage of TRV (kV)

TRV peak with MOSA
2500
2000
TRV peak with breaking resistor (700 ohm)
Twice TRV peak value of existing 550kV standards

1500
1000
60 90 120 150 180 210
Out-of-phase angle (deg)
Fig. 3.3.31 TRV out-of-phase switching vs out-of-phase angle
The triangular waveforms of TRV in China's UHV network [41] as well as in the report [42]
were observed in the case of out-of-phase conditions on a long OH line. The TRV at line-side can
be calculated by the travellling time multiplied by the rate of rise of the recovery voltage, where
RRRV = Zeq * di/dt and i is the peak value of IOP, the out-of-phase-current. Equivalent surge
impedance Zeq is given by that for the first pole to clear (similarly to determine ITRV) but not
for the last pole to clear (That is used to determine the RRRV of SLF interruption). The value
ranges from 300 to 350 ohm. For 50 Hz, this leads to RRRV = 0.13 to 0.16 * IOP. [43]
When the busbar side of the circuit breaker is to be represented by another OH line, a triangular
waveform will also appear at the source side, thus increasing the RRRV to, for instance, a double
value, but with different reflection times, as shown in Ref. [41].
The amplitude of IOP depends on the short-circuit power at the busbars at both line ends and the
line impedance. The longer the line, the lower the amplitude on one hand, but the longer the
reflection time on the other hand. The combination tends to produce slightly higher TRV peak
values for longer lines. As long as the short-circuit power at the busbars is rather low or the out-
of-phase angle rather small, the TRV peak value will be covered by 2.5 p.u. (product of an out-
of-phase factor of 2.0 and an amplitude factor of 1.25), as specified in the present standards (for
rated voltages up to and including 800 kV). However, at higher values of the short-circuit power
or out-of-phase angle, peak values exceeding 2.5 p.u. will appear.
Figure 3.3.32 shows the TRV calculated in Ref. [41] under out-of-phase conditions in Jin-Nan
and Nan-Jing lines in an 1100 kV network in China.
124
Figure 3.3.32 TRV waveform of Nanyang CB during out-of-phase conditions
while the oscillation centre is on the Jing-Nan line
The out-of-phase angle and the short-circuit power at the busbar are not given in Ref. [41]. In this
case with a rather low out-of-phase current, the peak value of the TRV and RRRV are covered by
that as specified for the out-of-phase test duty. The peak of 2035.9 kV is also covered by the
value calculated for terminal fault test duty T10 with a first-pole-to-clear factor of 1.3 (2055 kV);
it would not be covered if a first-pole-to-clear factor of 1.2 is taken for T10. The 0.716 kV/µs rate
of rise is sufficiently covered by the value for T10. It is required to investigate more detailed
whether T10 (or T30) test duty generally can cover the requirements for out-of-phase duty for
UHV systems.
For circuit breakers equipped with pre-insertion resistors, Ref. [41] shows that a much higher
TRV peak value is applied to the auxiliary interrupter, compared with the main interrupter, and
values exceeding 2.5 p.u. are already obtained at smaller out-of-phase angles. Also, simulation
results from Japan show that the peak value of TRV across the auxiliary interrupter is higher than
the TRV across the main interrupter (2.41 versus 2.28 p.u.). [21] The influence of PIR on the
TRV waveform is explained in Section 3.3.6.2.
The effect of MOSAs on the TRV waveform can be understood from Refs. [42] and [25]. The use
of special MOSAs with a very low SIPL can limit the TRV peak value to within 3.2 p.u., as
shown in Fig. 3.3.30. MOV in parallel to the arcing chambers can limit the TRV peak value to
3.0 p.u., when the involved circuit breakers are required to synchronize the systems.
3.3.6.2 Effect of opening resistors under out-of-phase conditions

There is a report from China [41] indicating that TRV peak under out-of-phase condition with
opening resistor scheme is higher than that without opening resistor, while similar study in Japan
showed that TRV peak is reduced by applying the opening resistor method [21]. This section
provides technical explanations for these conflicting results.
In general, the opening resistors can mitigate the TRV oscillation for terminal fault duties and
reduce the RRRV and peak of TRV. In the case of out-of-phase conditions with or without
opening resistors, TRV peak is affected following several parameters.
1) Frequency of line side oscillation
2) Whether the line side oscillation is at one side of CBs or both sides
3) Phase angle between two network under Out-of-phase condition
125
When the opening resistor is applied, two additional parameters affect the TRV.
4) Phase shift
5) Amplitude of line side oscillation (Before interruption)
3.3.6.2.1 Phase shift

The phase angle of the interrupting current is decided by the total impedance of the network and
is normally delayed 90 degrees from the voltage if there is no opening resistor. In the case of
opening resistor applications, the phase angle of the interrupting current is determined by not
only the back (network) impedance but also the opening resistor. Considering the several hundred
ohm of opening resistor and the back impedance corresponding to 50 to 63 kA short-circuit
current in the network, the opening resistor will be dominant and the phase shift of interrupting
current will be up to around 10 degrees. In the case of a network with small short-circuit current,
the back impedance becomes large and comparable to the resistance value of the opening resistor.
In such cases the phase shift will go to up to around 40 degrees. This phase shift influences the
TRV peak value when the peak of the transient oscillation of the line side comes close to the peak
of the power source voltage.
3.3.6.2.2 Amplitude of line side oscillation

Before the out-of-phase condition is interrupted, the voltage between two networks is distributed
along the back (network) impedances and opening resistor. Voltage distribution on the back
impedance will increase the amplitude of the transient oscillation at the line side. For networks
with large short-circuit currents, voltages across the back impedance is small and around several
to 10%. In the case of the networks with small short-circuit currents, the voltage on the back
impedance becomes large and is up to 20% on one side.
3.3.6.2.3 Other parameters

1) Frequency of line side oscillation
Frequency of line side oscillation is roughly in proportion to the line length, but varies depending
on the degree of shunt reactor compensation. If the degree of compensation increases, the
frequency of line side oscillation goes lower, and when this frequency is close to power
frequency, the peak of line side oscillation is likely to overlap the source voltage peak and cause
a severe TRV peak.
2) Whether the line side oscillation is at one side of CBs or both sides
In the IEC standard, line side oscillation is assumed to be on one side of the CB. In the case of a
switching station it is possible to have oscillations on both sides of the CB. If the line length of
both sides is almost the same, the oscillation frequency is also same in opposite polarity and then
severe TRV will appear.
3) Phase angle between two networks of Out-of-phase condition
Phase angle between two networks of Out-of-phase condition is also important as discussed in
other section.
3.3.6.2.4 Simplified calculation
126
Simplified calculation is carried out assuming the UHV pilot plant in China as Network 1 and
Japanese planned UHV network as Network 2 respectively. Parametric conditions for this study
are shown in Table 3.3.20. NOTE: parameters of NETWORK1 and 2 are chosen only for this
study and are NOT same with actual network parameters of neither Japan nor China. Generally
there are surge arresters in actual UHV network in China and Japan, and the transient voltage to
the ground is cut off at about 1300 kV by these arresters. However, no arresters are modelled in
this study to grasp the effect by deference of network condition to TRV without non-linear
elements.
Examples of voltage waveform of the line side voltage are shown in Fig. 3.3.33 and comparison
of TRV voltage is shown in Table 3.3.21. Major points for affecting parameters to TRV peak are
listed in Table 3.3.22. From Table 3.3.22 it is understood that, condition of Network 1 was of
severe condition for out-of-phase TRVs with opening resistor and resulted in higher TRV peak
than that of without opening resistor. More than 20% of voltage is distributed on back impedance
and then, such level of amplitude of line side oscillation occurred. Frequency of line side
oscillation is about 120 Hz. Phase shift of interrupting current is around 40 degrees. These
parameters led the superimposing of side line oscillations peak to near peak of power frequency
voltage with opening resistor. Adding to this, Network 1 is the case of switching station in the
middle of 600 km length OH-line. This implies line side oscillations are applied to each side of
terminals of Circuit breaker at reverse polarity.
3.3.6.2.5 Summary
TRV peak for out-of phase condition with opening resistor method is affected by the system
configuration and network parameters, and vary. Attention shall be paid to these phenomena
when applying the opening resistors.
Table 3.3.20 Simplified network for study
NETWORK 1 NETWORK 2
System voltage 1100kV / root 3 (=635.1 kV rms) 1100kV / root 3 (=635.1 kV rms)
Back impedance 741 mH 162 mH
(SC current = 2.73 kArms) (SC current = 12.5 kArms)
OH line capacitance 3 micro F (both sides) 2.1 micro F (One side)
(0.01 micro F *300 km) (0.01 micro F *210 km)
Sh. reactor 1688.5 mH 4825 mH
(100% compensation) (100% compensation)
Opening resistor 600 ohm 600 ohm
Phase difference 180 degree 180 degree
Frequency 50 Hz 50 Hz
Configuration Switching station Substation
NOTE: 1) Parameters of NETWORK1 and 2 are chosen only for this study and is NOT same with actual
network parameters of Japan nor China, 2) No arresters are modelled in this study to grasp the effect by
deference of network condition to TRV without non-linear elements.
Table 3.3.21 Comparison of TRV peak voltage

(TRV peak / peak voltage of line side to ground)
NETWORK 1 NETWORK 2
Without resistor 100 % / 100 % 86 % / 130 %
With resistor 117 % / 116 % 80 % / 72 %
127
Without resistor With resistor (600ohm)
2.0 Voltage at CB line side 2.0 Voltage at CB line side

[MV]
1.5
[MV]
1.5
Source voltage
Source voltage
1.0 1.0
NETWORK 1
0.5 0.5
0.0 0.0
- 0.5 - 0.5
- 1.0 - 1.0
- 1.5
Voltage of back impedance - 1.5
Voltage of back impedance
- 2.0 - 2.0
10 14 18 22 26 [ms] 30 50 54 58 62 66 [ms] 70
(file CUD007.pl4; x- var t) v:SOU1 v:CB11 v:SOU1 - CB11 (file CUD008.pl4; x- var t) v:SOU1 v:CB11 v:SOU1 - CB11
factors: 1 1 1 -1 factors: 1 1 1 -1
offsets: 0 0 Current interruption
0 0 offsets: 0 Current interruption
0 0 0
Voltage at CB line side Voltage at CB line side

2.0 2.0
[MV]
1.5
Source voltage [MV]
1.5
Source voltage
1.0 1.0
NETWORK 2
0.5 0.5
0.0 0.0
- 0.5 - 0.5
- 1.0 - 1.0
- 1.5 - 1.5
Voltage of back impedance Voltage of back impedance
- 2.0 - 2.0
10 14 18 22 26 [ms] 30 50 54 58 62 66 [ms] 70
(file CUD037.pl4; x- var t) v:SOU1 v:CB11 v:SOU1 - CB11 (file CUD038.pl4; x- var t) v:SOU1 v:CB11 v:SOU1 - CB11
factors: 1 1 1 -1 factors: 1 1 1 -1
offsets: 0 0 Current
0 0 interruption offsets: Current
0 0 interruption
0 0
Fig. 3.3.33 Typical waveforms of line side TRV (Red: Source voltage, Brown:Line
side oscillation, Blue:Voltage at CB line side terminal )
Table 3.3.22 Comparison of two network characteristics with opening resistor method
NETWORK 1 NETWORK 2
Phase shift (Angle from voltage Large: Due to large back Small
zero to current zero point) impedance
TRV amplitude at line side Large: Due to large back Small
impedance
TRV frequency at line side Middle: Due to long line and High: Due to relatively short
compensation line
System configuration Both sides are line Normal: One side is line and
(Switching station) another side is transformer
3.3.7 High frequency components imposed on fault currents

Fault currents are distorted by the superposition of high-frequency surge currents caused by
sudden voltage change. Since an 1100 kV system has twice the operating voltage of a 550 kV
system, and high-frequency surge currents attenuate slowly due to small transmission line
resistance, the fault current distortion of the 1100 kV system becomes larger than that of the 550
kV systems. Interruption duty becomes 10% more severe due to current distortion. Figure 3.3.34
shows an example of the analysis waveform of a fault current.
128
Fig.3.3.34Analysis waveform of fault current (example)
Figure 3.3.35 shows the fault current waveform of an 1100 kV transmission line having a present
operational voltage of 550 kV. The fault current is about 2.1 kA and was caused by a two-phase
short circuit. The fault occurred outside the protection range, and the fault current flows through
this line.
Consequently, the current does not drop to zero after the fault is eliminated. High-frequency
surge currents attenuate within about 1.5 cycles. To understand this phenomenon in more detail,
it is necessary to accumulate additional data on fault currents.
Fig. 3.3.35 Double phase fault in 1100 kV line operated at 550 kV (Japan)
129
Fig. 3.3.36 Double phase fault in 735 kV line (Hydro Québec, Canada)
The majority of high-frequency-component fault currents in Canada’s 735 kV transmission lines

(686 mm2 × 4 conductors) were attenuated within one cycle and no prominent oscillations were
observed at the clearing instant, as can be seen from the actual case in Fig. 3.3.36.
3.3.8 Capacitive current

3.3.8.1 General
As capacitive current switching tests must be performed as single-phase for UHV applications, it
is necessary to define the voltage factor (k) to be applied to the phase-to-earth voltage in order to
reproduce the voltage stress of a three-phase breaking operation.
Figure 3.3.37 gives the recovery voltage on the first pole-to-clear as a function of C1 and C0.
Fig. 3.3.37 Recovery voltage as a function of C1 and C0
130
Typical values for the ratio of C1 / C0 are between 1.5 and 2.0. For example, the calculated values
for C1 and C0 for Eskom’s 800 kV lines with single circuits and 6 × Tern ACSR conductors per
bundle are as follows:
Horizontal configuration: C1 = 13.2 nF/km : C0 = 8.9 nF/km C1 / C0 = 1.48
Delta conductor configuration: C1 = 14.8 nF/km : C0 = 7.6 nF/km C1 / C0 = 1.95
3.3.8.2 Voltage factor for UHV
3.3.8.2.1 Switching overhead-lines under normal conditions

Depending on the ratio of C1 and C0, a voltage factor must be applied during the single-phase test
in order to obtain the same TRV peak as during three-phase operation. Using the data from the
previous example, the peak recovery voltage is 2.1 and 2.3, respectively, and the voltage factor
for the single-phase test would then be 1.05 and 1.15, respectively.
However, when determining the voltage factor, it must be taken into account that restrike will not
necessarily occur at the peak value of recovery voltage, but would more likely occur during the
period between one-quarter and one-half cycle of the power frequency after current interruption.
For example, Fig. 3.3.38 shows that in the case of capacitor banks with ungrounded neutral, the
peak recovery voltage on the first pole to clear can be obtained during a single-phase test using a
voltage factor of 1.25, but in order to cover the recovery voltage during the first half-cycle, the
voltage factor must be 1.4, as defined in IEC 62271-100 and ANSI/IEEE C37.04.
Based on the previous considerations, in the case of unloaded line switching for which the ratio
C1 / C0 is between 1.5 and 2.0, a voltage factor of 1.2 is recommended for operations with no
other phase faults.
TRV 3-PHASE CAPACITIVE CURRENT SWITCHING (50Hz)
Capacitor bank with isolated neutral
3
Single-phase test with kc= 1,4

2,5
2
TRV (p.u.)
Three-phase test
1,5
1
Single-phase test with kc= 1,25
0,5
0
0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 5 5,5 6 6,5 7 7,5 8 8,5 9 9,5 10
T (ms)
Figure 3.3.38 Comparison of recovery voltages of three-phase and single-phase tests
3.3.8.2.2 Switching overhead-lines under fault conditions

The voltages and currents in the case of a fault interruption in a transmission line, depend on the
circuit parameters and the interrupting sequence. The highest voltage appears on the healthy
131
phase, which is interrupted prior to the faulted phase. The highest current occurs on the last
interrupted phase when the faulted phase is the first pole to be interrupted. [44]
The recovery voltage is typically 2.74 p.u., when switching off a healthy phase in a 550 kV
network in the case of a phase-to-earthed fault. For phase-to-phase fault conditions, the recovery
voltage and capacitive current are less severe than for phase-to-earthed fault conditions.
Based on the previous considerations, a voltage factor of 1.4 would be required where capacitive
current switching must be verified in the presence of single- or two-phase earthed faults.
3.3.8.3 Shunt compensated lines

Shunt reactors, which are used to limit switching and temporary overvoltages at planned
commutation and fault clearing, are placed on at least one end of the line, but usually on both
ends of a UHV line that exceeds 100 km. In this case, switching-off of the line is accompanied on
the line side by slowly attenuating low frequency, close to 50/60 Hz, so the resulting voltage
across the circuit breaker is lower than in the case of uncompensated lines. This, however,
requires the connection of shunt reactors to the OH line before de-energization of the line, which
is common practise. But, in the case of transfer tripping in single- or three-phase mode (such as
with three-phase high-speed auto-reclosing applications), it is possible for the OH line to be de-
energized without shunt reactors connected to the heavily loaded line.
3.3.8.4 Current
An increase in the rated voltage of UHV lines having bundled phases results in a rise in specific
line capacitance and a steep increase in the charging current per km. As the distance between
UHV substations tends to increase with voltage, the charging current of UHV lines could be
extremely large.
The capacitive current that must be interrupted varies widely depending on the length and type of
line considered. For UHV systems, the value ranges from 1000 A (210 km line in Japan) to
2030 A (700 km line in Russia). Similar values are already required for capacitor banks in high-
voltage systems, such as 1200 A as stipulated in the draft revision of IEEE C37.06/8.2-2007.
Compensated lines with shunt reactors have lower charging currents than uncompensated lines.
The charging current of a compensated line is equal to the value for the uncompensated line
multiplied by one, minus the ratio of reactor reactance and line capacitive reactance.
Interruption of such high capacitive currents is not necessarily difficult for high-voltage circuit
breakers as the highest probability of restrike is usually met with short arcing times, i.e. with
lower currents. Accordingly, test duty LC1 in IEC 62271-100 is required to be performed with a
low percentage (10 to 40%) of the rated capacitive breaking current.
Adopting the lowest percentage (10%) for test duty 1 or even introducing a lower value such as
5%, while still performing test duty LC2 with 100% of the rated capacitive breaking current,
would make it possible to test the circuit breakers in the most severe cases.
3.3.9 UHV reactor current switching

3.3.9.1 General considerations
Shunt reactor switching is addressed as an IEC standard. [45] Here, general considerations are
reviewed briefly, followed by a more detailed description of UHV applications.
132
Shunt reactor switching is a unique and onerous duty for circuit breakers. At high voltages, the
shunt reactor current is low (in the order of a few hundred amperes at most) and current
interruption is readily achieved. However, due to the nature of the interaction between the
breaker and the circuit, the current is forced to zero prior to the natural power-frequency zero-
crossing, a phenomenon known as current chopping. Current chopping results in stored energy in
the reactor and when that energy is converted to capacitively stored energy, chopping overvoltage
occurs. This overvoltage is referred to as suppression peak overvoltage (Fig.3.3.39). The transient
recovery voltage (TRV) across the breaker is then the difference between the source-side power
frequency voltage and the suppression peak overvoltage.
Fig. 3.3.39 Suppression peak overvoltage

The general case for shunt reactor switching is the four-legged scheme as shown in figure 3.3.40.
The suppression peak overvoltage ka in p.u. is given by:
2
1  ic   L 
k a = (1 + K ) 1 +     − K …(21)
(1 + K )  u o   CL 
where K is the neutral shift in pu, ic the chopped current level, uo the peak voltage across the
reactor at current chopping, L the inductance of the reactor and CL the total load side capacitance.
Fig. 3.3.40 General four-legged shunt reactor switching case

The peak value of the TRV across the breaker krv in pu is given by:
k rv = 1 + 2K + k a …(22)
133
For four-legged reactor schemes at EHV, the neutral reactor is bypassed (neutral earthed) prior to
switching the main reactor. The same procedure would apply at UHV and K can be set to zero in
Equations (14) and (15) giving:
2
i  L
k a = 1 +  c 
 uo  CL …(23)
and
k rv = 1 + k a …(24)
The frequency of the suppression peak overvoltage is given by:

1
f=
2π LCL
…(25)
Using the chopping number approach, Eqn (16) can be written as [44]:
1.5Nλ2  CP 
k a = 1+  + 1
ωQ  CL  …(26)
where N is the number of interrupters in series in the breaker, λ is the chopping number in AF-0.5,
Cp is the grading capacitance across one interrupter, ω is the angular frequency and Q is the
three-phase reactor rating in VA. Equation (26) assumes that CS >> CL.
As for EHV shunt reactor switching, UHV reactor switching will be treated on a single-phase
basis including for type testing. Type testing is done on a unit basis and the test results are
extended by calculation to the multiple interrupter cases. As the behaviour of multiple interrupter
breakers for reactor switching is not fully understood, some caution should be exercised and
appropriate TRV limitation measures should be provided.
Reignition overvoltages are generated by reignitions following an initial interruption. Reignitions
are provoked when the voltage between the contacts exceeds the dielectric withstand of the
contact gap. The rate of rise of a reignition overvoltage is between those observed during fast-
front transients and those of lightning overvoltages, while that of a chopping overvoltage are
similar to that of slow-front transients. [46], [47]
The amplitude of a reignition overvoltage sometimes becomes higher than that of a chopping
overvoltage, when the reignition occurs near the crest of the load side oscillation voltage,
therefore the destructive capability of a reignition overvoltage on the insulation of a reactor as
well as a circuit breaker must also be considered. Reignition may occasionally be followed by
high-frequency arc extinction that could trigger voltage escalation. In South Africa’s 800 kV
systems, the highest switching and reignition overvoltage in a full GIS system were measured as
1.6 p.u. and 1.8 p.u. respectively. In Japan, high frequency arc extinctions were occurred in actual
300kVGIS connected to a cable in the field. [48], [49]
3.3.9.2 TRV limitation

Due to the well-known effects of reignition on circuit breakers, [50] the intent of TRV limitation
is to either significantly mitigate reignition, i.e. limit the voltage at which it occurs, or to avoid
134
reignition altogether. Firstly, however, is the selection of breakers: from Equation (26), the
breaker should have the lowest possible values for λ and Cp and the fewest number of interrupters
N in series. Auxiliary measures for consideration are described in Table 3.3.23 below.
Table 3.3.23 TRV limitation measures
Measure Advantage Disadvantage
Opening resistor Functions by phase shifting the current to Increases the mechanical complexity of
an angle << 90° thereby reducing the value the circuit breaker by the addition of the
of ka and krv resistor and associated switch
Metal oxide varistor Limits the voltage across the circuit Adds to the complexity of the circuit
breaker to its protection level and thereby breaker particularly if GIS; the varistor is
limiting krv exposed to repeated mechanical
vibration each time breaker is operated
Controlled Achieves a contact parting time such that Assumes a suitable minimum arcing
switching current interruption is achieved at the first time in order to achieve the purpose;
subsequent current zero, i.e. no reignitions minimum sizing of reactors for UHV
must be established to determine
viability of this option
Note: In Japan, installation of MOSAs to ground, between CB and reactor is more common practise to
protect the chopping and reignition overvoltages in addition to lightning overvoltage.
3.3.9.3 UHV shunt reactor switching circuit breaker requirements

In addition to all else, Equation (26) shows that shunt reactor switching circuit breakers can be
rated based on minimum reactor switching. Such a minimum rating at UHV has yet to be
established, so for demonstration purposes, this value is taken as 600 MVar based on Ref. [51].
Furthermore, it is assumed that the circuit breaker will have at least two interrupters in series
(best-case assumption) and the chopping number of a single interrupter is estimated at 30×104
AF-0.5. The value of ka and krv can now be calculated:
1+
(
1.5(2) 30 × 10 4 )
2
ka = (
314 600 × 10 6 ) …(27)
= 1.56 p.u.
krv = 2.56 p.u. …(28)
This is a minimum value for the reason stated above and because CP/CL was taken as being
negligible. For a CL value of 3000 pF, the frequency of TRV will be 958 Hz.
The TRV for 1100 and 1200 kV systems is shown in Table 3.3.24. Prospective biased switching
surge test values for live circuit breakers are also shown for comparison.
Table 3.3.24 TRVs for 600 MVar shunt reactor switching
System voltage (kV) TRV (kV peak) Prospective biased switching surge total test voltages (kV peak)
1100 2304 2556 - 2729
1200 2508 2636 - 2809
The aforementioned chopping number is applicable for circuit-breakers of the puffer technology.
Circuit-breakers based on auto-blast technology will show a lower chopping number, for instance
10 × 104 AF-0.5.
135
3.3.9.4 Discussion and conclusion
When a safety margin is applied to the TRV values in Table 3.3.24, the resulting withstand
voltage will approach, or perhaps even exceed, open breaker withstand requirements. Reaction to
this leads to the following considerations:
1. Voltage limitation by means of varistors across the breaker will most likely be required,
particularly if the chopping number is as high as estimated and the actual reactor ratings
are even lower than 600 MVar. In Japan, installation of MOSAs to ground, between CB
and reactor is more common practise to protect the chopping and reignition overvoltages
in addition to lightning overvoltage. WG A3.22 will study and compare these
countermeasures more in depth based on the actual experience.
2. Controlled switching is obviously mandatory to avoid reignition. As noted earlier, such
switching requires a suitable minimum arcing time. A single 500 kV interrupter breaker is
reported to have a minimum arcing time of 9 ms at 50 Hz. [52] Assuming that the
frequency of the load side is unchanged and that the amplitude of load-side oscillation at
1100 kV is twice that at 550 kV, the minimum arcing time for the single 550 kV
interrupter and the two 1100 kV interrupter breakers should be almost the same.
Controlled switching can thus be equally applied for the latter case.
3. Controlled switching on shunt reactors ideally requires a maximum arcing time without
reignition smaller than half a cycle (9 ms at 50 Hz). However, even if the CB does not
meet this requirement, it could still be acceptable by controlled switching techniques to
limit the reignitions at small arcing times, which will not cause excessive reignitions
overvoltages.
A first review of this subject concludes that, while much development work remains to be done,
reactor-switching breakers may incorporate both voltage limitation and controlled switching.
3.3.10 Controlled switching
3.3.10.1 General
Controlled switching systems (CSS) have become an economical solution and are commonly
used to reduce switching surges for various switching applications. Recent developments in
transformer switching, taking into account residual flux, can realise effective mitigation of severe
inrush current and temporary overvoltage that may lead to inadvertent operation of protective
relays and degradation of power quality. CSSs combined with metal oxide surge arresters can
reduce undesirable overvoltage caused by energization of long transmission lines to achieve
insulation coordination. The limited number of applications for line switching may stem from
initial difficulties caused by insufficient technical considerations including idle time
compensation. The upcoming IEC62271-302 Technical Report ‘High voltage alternating current
circuit breakers with intentionally non-simultaneous pole operation’ will standardize the testing
procedures required for CSSs based on the recommended evaluation tests by CIGRE WG A3.07
[46][52]. The CIGRE guide emphasizes the importance of compensating for variations in
operating time because CSSs require accurate operation consistency throughout the lifetime of
the circuit breaker. Variations in operating time due to external variables such as ambient
temperature, control voltage and mechanical energy of the drives can be compensated by the
136
controller using the dependence of Installation (Units)
variations on the variables
Line
1%
Transformer
4000
evaluated according to the testing 14%
requirements. Recent research

Reactor CIGRE
Line 18% Survey
3000 1%
indicates that the closing time for Transformer Capacitor
Capacitor
67%
certain drives is delayed after a few 2000

44%
CIGRE
32%
Survey
hours of idle time. Accordingly, Reactor
compensation is essential even for 1000
23%
a daily-operated CSS if it
demonstrates this dependence 0
1989 1991 1993 1995 1998 2001 2004 2006
(especially in the case of hydraulic
operating mechanisms).
Preferential Subject for CIGRE Session
CIGRE TF13.00.1 CIGRE WG13 (A3).07 IEC PT
The number of CSS installations

ELECTRA No.163, 164 ELECTRA No.183, 185 ELECTRA No.264 IEC62271-302
Controlled switching, stare-of-the-survey Application guide Specification & Testing Technical report
has increased rapidly since the late Fig.3.3.41 CIGRE survey on worldwide installations of CSS
1990s due to satisfactory service Table 3.3.25 Potential CSS applications to UHV systems
performance. CSSs are often Applications Targets Benefits
specified for shunt capacitor and Capacitor energization Elimination of closing resistor
Voltage zero across CB
shunt reactor banks due to several Line energization Reduction of overvoltage
economic benefits including Capacitor de-energization

Line de-energization
Optimum arcing time Minimization of restrike probability
elimination of closing resistors and Elimination of reignition probability
Reactor de-energization Optimum arcing time
extension of maintenance interval Reduction of associated overvoltage
for nozzles and contacts. Transformer energization

Prospective core flux
identical to the residual flux
Reduction of inrush current and
associated overvoltage
According to the CIGRE survey Elimination of closing resistor
Rapid auto-reclosing Voltage zero across CB
shown in Fig. 3.3.41, Reduction of overvoltage
approximately 2,400 CSSs were

supplied and installed around the world in 2001, and more than 4000 units are presently
estimated to be in service. Before 1995, the number of installations was limited due to
technological immaturity, but it increased rapidly when effective compensation algorithms were
introduced using advanced sensors and reliable digital relay technologies. Today, 70% of the
installations worldwide are applied to capacitor banks; however, CSSs are not used for shunt
capacitors in Japan, as the amplitude of inrush currents is suppressed by fixed inductors,
originally intended to reduce the 5th harmonics of the power frequency. In some applications, the
use of a large series reactor may lead to severe TRV during fault clearing generated between the
capacitor bank and the series reactor.
In response to the need for flexible operation and control of switching equipment in transmission
and distribution systems, advanced controllers capable of controlled transformer energization
taking into account the residual flux in the transformer core [53],[54] and also capable of
controlled re-closing of uncompensated lines [55],[56] have been demonstrated to be effective in
suppressing switching transients in the field. Further development efforts will increase the
number of installations of CSS for these applications.
The installations include a wide range of rated voltages for gas circuit breakers (GCBs) of up to
550 kV for one break or 800 kV for two and four breaks. Hydro-Québec has extensive practical
experience in controlled switching for 765 kV shunt reactors using air blast (8 breaks) and SF6 (4
breaks) technologies, since commissioned in 1992. Since 1100 kV GCBs are composed of two
series 550 kV one-breaks, the effectiveness of the CSS can be visualized from the field
experience with 550 kV GCBs.
137
90
Mechanical variations bands
Voltage Zero or minimum (Deg)

Minimum Making Voltage (pu) 1.2 RDDS
0.0 1.0 75 0.7
1.0 0.2 1.2 0.8
Optimal Target for

0.4 1.4 0.9
0.6 1.6 60
0.8 0.8 1.0
1.2
45 1.4
0.6
0.4 30
24 deg
0.2 15
0
0 0.5 1.0 1.5 0 0.5 1.0 1.5 2.0
Rate of Decrease of Dielectric Strength (RDDS) Mechanical variations bands (ms)
Fig. 3.3.42 Minimum making voltage and optimal targets for voltage zero or voltage minimum
as functions of the RDDS and mechanical variations bands of GCB
Table 3.3.25 summarizes potential CSS applications for UHV systems, which include small
capacitive current de-energization and energization, small inductive current de-energization [57],
transformer energization and rapid auto-reclosing. Various technical and economical benefits
include elimination of closing and opening resistors and reduction of switching overvoltage. The
most beneficial application of CSSs at UHV would be for shunt reactor switching and
transformer energization. Line switching would be more of a technical challenge, especially for
shunt-compensated lines where the controller algorithm must determine the optimum energizing
instant considering the oscillating trapped charge on the line.
1100 kV two-break GCB typically has a grading capacitor (C1) with the values of 1000-2000 pF.
The capacitance (C2) of the middle part of two-break interrupter to the ground is estimated as
100-250 pF. Therefore the voltage distribution between breaks ranges 51-55%, which is given by
(C1+C2)/(C1+C1+ C2). Assuming the voltage distribution factor of 1.1 (55:45), the RDDS
requirement for making at the voltage minimum (RDDS=1, which corresponds to the maximum
slope of the system voltage at zero crossing.) is calculated as 186 kV/ms at 60Hz, which is almost
the same dielectric recovery performance requirement for 550 kV one-break GCB.
Figure 3.3.42 shows the dependence of the minimum making voltage on the RDDS and the
mechanical variations bands. The correlation can be applied to all the ratings including UHV
GCB. In case of RDDS=1, the minimum making voltage becomes around 0.4 p.u. using GCB
with the mechanical variations bands of +/-1.0ms. The optimal targets for voltage minimum are
also shown as functions of the RDDS and the mechanical variations bands.
3.3.10.2 Capacitive switching applications

The use of shunt capacitor banks in a UHV system is considered unlikely because the banks are
generally located in close proximity to the load areas. However, CSSs are technically applicable
for capacitive current switching, even though GCBs are required to possess high RDDS
performance so as to close at the minimum voltage, in addition to the minimum mechanical
variation bands. Modern GCBs generally provide a very low probability of restrike for capacitive
current interruption. Nevertheless, the probability of restrike can be further reduced by means of
the CSS, which ensures long arcing times especially effective for UHV GCBs after several fault
current interruptions.
138
3.3.10.3 Reactor switching applications
Controlled reactor switching can reduce the inrush current or the transient stresses. The making
target that minimizes reactor inrush current is the voltage peak. The associated switching
overvoltage in this case is generally low, but a steep voltage wave front may stress the reactor
insulation. Since it is impossible to achieve reduction of both inrush current and transient stresses
on the reactor energization with the same target, a compromise solution must be reached.
For application to UHV shunt reactor banks, GCBs are required to possess a certain arcing
window to interrupt small inductive currents without reignition. Figure 3.3.43 shows the rate of
rise of dielectric strength (RRDS) using 550 kV one-break GCBs, which corresponds to the
characteristics for 1100 kV two-break GCBs. Tests were carried out at breaking currents of 315
A in accordance with test duty 1 stipulated in IEC62271-110.6.115.
Small inductive current interruption phenomena can be classified into three categories depending
on the contact gap at the instant of current interruption: (a) Reignition-free as a result of
successful interruption because the dielectric withstand between contacts always surpasses the
transient recovery voltage after interruption; (b) Reignition as a result of dielectric breakdown;
and (c) Thermal reignition as a result of thermal interruption failure. Thermal reignition,
however, does not cause any prominent transients and can successfully interrupt the current at the
next current-zero. Accordingly, the periods for a reignition-free window (a) combined with the
thermal reignition window (c) can be chosen as the opening target. Controlled opening of shunt
reactor banks can eliminate the reignition overvoltage, which can potentially induce GCB
damage such as nozzle puncture. It also provides economic benefits such as reduced possibility of
reactor damage and extension of the number of interruptions before the nozzle and contact need
to be replaced, exceeding a factor of two relative to that required for any rating of GCB without
CSS.
3.3.10.4 Unloaded transformer energization

Transformer energization can create high-amplitude magnetizing inrush current of up to several
thousand amperes and may result in undesirable temporary overvoltages that cause inadvertent
operation of protection relays due to saturation of the power transformer iron core. High inrush
current also imposes severe mechanical stress on the transformer windings and may reduce the
life expectancy of a transformer exposed to frequent energization; for example, step-up
transformers in hydroelectric power plants are frequently switched to adapt to daily load
variations. Considering the importance
of securing long-term dielectric
Voltage (kV)
Rate of Decrease of
Dielectric Strength (RRDS) 1500
performance of UHV transformers, it 853kV for single pole of
1100kVGCB (550kV)
is preferable to reduce the mechanical Peak voltage of
Re-ignition 1000
and electrical stress from energization. reactor opeing test
Contact separation point 500

free of reignition
Inrush current will depend on (a) Reignition
magnetic characteristics of the

0
Contact separation point

transformer core; (b) processing causing reignition
instants of the circuit breaker; (c) Thermal

reignitio
n region Reignition-free window Reignition region
Thermal
reignitio
n region
electrical connections of the power Reignition region
（ms）
）
transformer; (d) residual flux; and (e) 12 10 8 6 4 2
Interruption
Time（
Target widow Avoidance region

electrical characteristics of the
“source” system. Fig.3.3.43 RRDS characteristic for 550kV GCB
139
Transient Flux
Flux
Residual Flux
Flux
Current
Voltage
Inrush
current
Energization
Severe inrush current is observed when a
high amplitude of transient flux is generated
at energization.
Fig. 3.3.45 Transformer bushing voltage sensor
Flux Flux
Stable Flux
residual
Flux
Current
Voltage
Inrush
current
Prospective Flux
Energization
Inrush current can be suppressed by
energization at the instant when the prospective
flux is equal to the residual flux of the core.
Fig.3.3.44 Magnetizing flux in a core of a Fig 3.3.46 Typical Measurement of the residual flux
transformer and corresponding magnetizing current
Figure 3.3.44 shows the dynamic magnetic flux and current behaviour when a transformer is
energized. Energization of a transformer with no residual flux in the core at peak voltage will not
cause any transients. However, the change in flux after energization depending on the residual
flux and the energization instant will generate greater saturation of the magnetizing current.
Therefore, the optimum targets should be adjusted taking into account the residual flux. The
inrush current can only be minimized by energization when the prospective normal core flux is
identical to the residual flux.
An innovative residual flux measurement was developed and its accuracy proven in the field by
integrating the voltage waveform after de-energization of the transformer as well as any CB
operations in case of fault clearing connected to the system.
In 2001, Hydro Québec has installed two prototype systems to control the optimum instant for
energizing high-voltage step-up transformers. Each system consists of a synchronous controlled
switching system with the new residual magnetic flux measurement system [53]. The system uses
the relationship between the magnetic flux “Φ” of an “N” turns coil and the applied voltage “E”,
expressed in this equation: Φ=∫E/N+Φr, where Φr is the residual magnetic flux. Measuring and
integrating the voltage of each phase just before and during de-energization of the transformer in
the steady state determines the residual flux, which is the end value of the integrated voltage.
With this method, only a small number of cycles are used for integration to prevent that even a
small offset error in the voltage measurement lead to a large error in the results. A novel
transformer built-in voltage sensor also has been developed to measure each phase voltage of the
transformer shown in Fig.3.3.45.
140
Phase U Phase V Phase W Phase U Phase V Phase W
500 500
Secondary VT Primary BUS

Secondary VT Primary BUS
Voltage (kV)
Voltage (kV)
0 (ms) 0 (ms)
0 10 20 30 40 50 0 10 20 30 40 50
- 500 - 500 Voltage drop: 0.32%
Voltage drop: 11.2%
300 300
Flux (P.U.) Voltage (V)

Flux (P.U.) Voltage (V)
0 (ms) 0 (ms)
0 10 20 30 40 50 0 10 20 30 40 50
- 300 - 300 Residual flux;
2 2 U:+8%, V:-37%, W:+29%
0 (ms) 0 (ms)
0 10 20 30 40 50 0 10 20 30 40 50
-2 Residual flux; -2
A
U:-3%, V:+31%, W:-28%
3 Delay closing
3
Current (kA)
Current (kA)
0 (ms) 0 (ms)
0 10 20 30 40 50 0 10 20 30 40 50
-3 Inrush current;
-3 Inrush current;
U:18A, V:20A, W:96A
U:1245A, V:2678A, W:1680A
Energization Phase V Phase U,W
instant Energization Energization
Fig. 3.3.47 (a) 3-phase simultaneous energization and (b) Controlled energization taking into account residual flux
Figure 3.3.46 shows a typical record of measured residual flux at the last opening, which is used
for adjustment of breaker characteristics and issuing a closing order for each phase. The results
show no inrush current or TOV on any phase. Figure 3.3.47 (a) shows a typical example of
voltage, current and flux in the core calculated in the case of three-phase simultaneous
energization. High inrush current ranging from 1245 to 2678 A is obtained due to magnetic
saturation in the cores, causing a voltage drop of up to 11.2% in the primary bus terminal. On the
other hand, figure 3.3.47 (b) shows a typical example of controlled transformer energization. The
controller chooses a phase with the maximum residual flux (Phase V) as the first phase to close.
The first phase is closed at the instant that the prospective normal core flux is identical to the
residual flux. The second and third phases (U and W) are closed 1.5 cycles later around the
voltage peak of the first phase. The inrush current is reduced to below 100 A and voltage
disturbance is effectively suppressed. By choosing the phase with the maximum residual flux
value as the first phase to be energized, both mechanical and electrical stress on the transformer
can be greatly reduced.
Most power transformers at the UHV level may not be energized or de-energized frequently;
however, the step-up power transformers in hydroelectric power plants can be switched often to
adjust to daily load variation. Therefore, a controlled transformer switching system taking into
account the residual flux can be used to effectively reduce the overvoltage caused by transformer
energization.
Furthermore, the sympathetic interaction phenomenon related to no-load transformer energization
sometimes appears when other power transformers are energized. The associated inrush current
may cause complete saturation in the iron core of the energized power transformers. Depending
on the severity of this phenomenon, the overcurrent protection relays or even the differential
protection relays of such on-load power transformers may operate inadvertently, causing a
complete shutdown of all transformers and subsequent system black-outs. [58], [59]
In the case of UHV-level power system expansion, performing no-load power transformer
energization might increase these phenomena. As this issue is of significant concern,
141
consideration should be given to the application of a CSS, which can greatly mitigate such
transient phenomena in the presence of UHV transmission systems/links.
3.3.10.5 Uncompensated and compensated line switching

Energization and auto-reclosing of long transmissions lines can cause undesirable overvoltages in
the transmission network so special overvoltage mitigation measures are employed to meet the
insulation coordination. The most common practise has been metal-oxide surge arresters that are
often combined with a closing resistor to improve reliability, but this approach is relatively
expensive. There is also a growing need of shunt reactors in order to mitigate the phase-to-ground
transient overvoltages due to line switching especially high-speed auto reclosing. The highest
phase-to-ground transient overvoltage may appear in the middle of the line, because they are
limited at the ends of the lines due to the application of shunt reactors, MOSA and CSS. Using
MOSA at the middle of the line can reduce these overvoltages.
The CSS can potentially reduce the re-closing transients and improve the reliability of restrike-
free performance. It can also provide economic benefits such as elimination of closing resistors
and reduction of the insulation level for surge arresters and transmission towers. For line
applications, GCB with a higher RDDS is generally preferable although operating scenario and
targeting strategies should be studied thoroughly. Idle time compensation is essential for drives
whose operating times have this dependence.
For UHV systems, the optimal resistance of closing resistors could vary depending on the length
of transmission lines. Furthermore, the effectiveness of closing resistors in reducing the
overvoltage may decrease in the case of very long lines, generally exceeding 200 km.
(a) Uncompensated line equipped with inductive potential transformer, where the trapped charge is relatively rapidly discharged
Voltage on a source side Voltage on a source side
Discharging Discharging
Voltage on a line side Voltage on a line side
Interruption Making (Optimum) Interruption Making

Making at a zero crossing of the source-side voltage is appropriate for line which are rapidly discharged such as those
quipped with inductive potential transformer. (RC<100ms typically)
(b) Uncompensated line equipped with capactive potential transformer, which do not affect the line discharging process
Trapped charge Voltage on a line side Trapped charge Voltage on a line side
Interruption Making (Optimum) Interruption Making
Making at a peak of the source voltage of the same polarity as the trapped charge is appropriate for line which do not
exist a leakage path of the trapped charge such as those quipped with capacitive potential transformer. The controller
should detect the line voltage polarity at the line charging current interruption instant.
Fig.3.3.48 Optimum making targets for uncompensated line switching

Consequently, using the CSS combined with MOSAs could be a viable solution for a UHV
transmission system with long lines.
142
(a) For higher degree of compensation, the voltage across the circuit breaker has (b) For lower degree of compensation, the voltage across the circuit breaker leads
a pronounced beat. to more complex voltage wave shape with a less pronounced beat pattern.
20 40 60 80 100 120 140 160 180 20 40 60 80 100 120 140 160 180
Voltage on a line side Voltage on a line side
20 40 60 80 100 120 140 160 180 20 40 60 80 100 120 140 160 180
Voltage across the circuit breaker Voltage across the circuit breaker
20 40 60 80 100 120 140 160 180 20 40 60 80 100 120 140 160 180
(ms)
(ms)
Target Target Target Target Target Target Target
window window window window window window window
Fig.3.3.49 Optimum making targets for compensated line switching
Strategies for different line configurations

are described below.
a) In the case of an uncompensated line with
an inductive potential transformer, the
controller can suppress the surge effectively
(less than 1P.U.) by controlled closing at
voltage-zero on the source-side because the
trapped DC charge is rapidly discharged
(typically less than 100ms). (See Figure
Fig. 3.3.50 500 kV transmission line arrangement
3.3.48)
b) In the case of uncompensated line with a
capacitive potential transformer, no leakage
path exists for the trapped charge. The
optimum target is voltage peak on the
source-side of the same polarity as the
trapped charge.
c) In the case of a compensated line, the
degree of compensation has a significant
effect on the line-side voltage. The voltage Fig. 3.3.51 Voltage profiles with a closing resistor or two
across the breaker show a prominent beat and three MOSAs
especially for a high degree of compensation
because the line oscillation frequency
typically falls in the range 30-50Hz. The
optimum instant is voltage minimum across
the CB; preferably during a period of the
minimum voltage beat [56] as shown in
Figure 3.3.49.
British Columbia Hydro has 500kV
transmission lines with a typically 300 km. Fig. 3.3.52 Three-phase reclosing overvoltage for various
Some systems are equipped with mid-line mitigation methods
series capacitors providing 50%
143
2.5pu
2.0pu
Energization withougt CSS
Energization with 1000ohm closing resister 1.8pu
Energization with CSS
500kV 60Hz
CB Transmission line (400km)
90mH
1pu=449kV Surge Surge
Reactor Arrester Arrester
40%compensation
1pu=449kV
1pu=449kV
Voltage (kV)
at source side　 Voltage (kV)
at source side　
Overvoltage :815kV (1.8pu)
Voltage (kV)
at line side Voltage (kV)
(0km from CB)　 at line side
(0km from CB)　
Voltage (kV)
Overvoltage :1105kV (2.5pu) (200km from CB)　
Voltage (kV)
(400km from CB)　
Voltage (kV)
across the CB　 Voltage (kV)
across the CB　
550 A rms
Current (kA) 550 A rms
through CB　 Current (kA)
through CB　
De- energization Simultaneous energization
De- energization Controlled energization
(a) Random energization (b) controlled energization

Fig.3.3.53 Reduction of overvoltage in application of controlled line switching
compensation. Economic constraints dictated that the bypass protection for this bank comprises a
MOSA rated at 2.2 p.u., 89 MJ, with no self-triggered or triggered gap. The line is equipped with
135 MVA/525 kV three-phase shunt reactors at each end of the line with a switchable reactor at
the receiving end.
The line will be operated with single pole reclosing and consequently the southern end shunt
reactor is equipped with a neutral reactor rated 1600 ohms for secondary arc damping. The
reactor at the northern end is grounded through a surge arrester to achieve a high grounding
impedance to optimise the secondary arc control. Figure 3.3.50 indicates the line equipment
arrangement [56].
Figure 3.3.51 compares the voltage profile along a line with conventional closing resistors to the
use of two and three MOSAs with staggered closing. The effectiveness of overvoltage control
due to the closing resistors decreases when the line length becomes longer than 150 km. It is
evident that with three MOSAs, only a small portion of the line experiences voltage exceeding
the design target, even with the non-optimum times inherent in staggered closing.
Figure 3.3.52 indicates the improvement that can be achieved by using line-connected MOSAs.
Two alternatives were considered: a MOSA at each line terminal and three MOSAs with the third
connected at mid-line. Note that the objective of reducing the overvoltage to less than 1.7 p.u. can
be achieved only with the use of three MOSAs plus controlled closing. Figure 3.3.53 shows
typical analyses of overvoltage reduction of controlled uncompensated line energization.
In another example, Tables 3.3.2 and 3 present some statistical simulations from switching
overvoltage studies on a 550 kV transmission line during the planning/design stage. This
transmission line is part of an interconnection tie between the two major grids of Brazil from 10
years ago (south/southeastern and north/northeastern grids). These two tables present the
144
analytical results of both phase-to-ground and also phase-to-phase transient overvoltage in the
line-receiving end and in the middle of the lines. [60]
Table.3.3.2 No-load line energization cases performed (550 kV Jaú→Miracema line)

Cases Switching overvoltage at ends Switching overvoltage in mid- MOSA Energy in line receiving
(p.u. - φ-n/φ-φ) point (p.u. - φ-n/φ-φ) end (p.u.)
Maximum Average Maximum Average Maximum Average
MOSA at both
1.95/2.25 1.91/2.15 2.50/2.40 2.18/2.14 0.48 0.25
line ends
PIR plus MOSA
1.90/1.90 1.66/1.65 1.95/1.85 1.63/1.59 0.10 0.01
at both line ends
CSS plus MOSA
1.90/1.95 1.83/1.88 2.05/2.15 1.87/1.83 0.15 0.04
at both line ends
Table.3.3.3 Automatic tripolar auto-reclosing cases performed (550 kV Jaú→Miracema line)

Cases Switching overvoltage at ends Switching overvoltage in mid- MOSA Energy in line receiving
(p.u. - φ-n/φ-φ) point (p.u. - φ-n/φ-φ) end (p.u.)
Maximum Average Maximum Average Maximum Average
MOSA at both
2.05/2.30 1.92/2.14 2.95/2.65 2.26/2.19 1.30 0.40
line ends
PIR plus MOSA
1.90/1.95 1.69/1.61 2.00/1.90 1.70/1.56 0.10 0.01
at both line ends
CSS plus MOSA
1.90/2.00 1.81/1.85 1.95/2.00 1.83/1.84 0.05 0.02
at both line ends
The results reveal that the general performance of the CSS is similar to PIR application, either
controlling phase-to-ground or phase-to-phase switching transient overvoltages, at both the
receiving end and middle of the line. A comparison of cases with/without CSS shows that
switching overvoltage can be mitigated at a considerable level in the middle of the line. It is also
obvious that the potential for phase-to-phase TOV mitigation exists.
The use of a more compact UHV OH line design yields expressive savings in capital-costs of the
line itself, due to the reduction of mechanical structural dimensions and weight. MOSAs and CSS
all together can potentially allow transmitting large amounts of powers perspective considering
recent increasing economic and environmental constraints.
3.3.10.6 Summary
The rapid increase of CSS application is ascribed to several factors including successful field
experience using the system with an effective compensation algorithm, the CIGRE proposal for
type testing recommendations and versatile operation and control of transmission systems due to
global changes in the electrical industry. Since CSSs can provide significant technical and
economic benefits including enhanced power quality and operation flexibility, they could be
incorporated into circuit-breaker control systems as a standard specification in the near future.
The advantages of CSSs will be more extensive for UHV systems, where experts place emphasis
on reducing switching overvoltage as much as possible, as insulation coordination is mainly
based on slow-front overvoltage and its cost consequences are considerable. Based on the
aforementioned considerations, there is no reason why CSSs cannot be applied at the UHV level,
145
as the behaviour of 1100 kV two-break circuit breakers can be directly extrapolated from single-
break 550 kV circuit breakers.
CSSs will provide many economic benefits when applied to shunt reactor banks and transformer
energization in the UHV system. In the case of processing at the voltage minimum, RDDS of at
least 186 kV/ms at 60 Hz (155 kV/ms at 50 Hz) is required for 1100 kV two-break GCBs, which
corresponds to RDDS = 1.
Step-up UHV transformers in hydroelectric power plants are frequently switched to cope with
daily load variations. Controlled transformer switching taking into account the residual flux is a
viable scheme for effectively reducing overvoltage caused by transformer energization. Since the
effectiveness of closing resistors in reducing overvoltage may decrease in the case of very long
lines typically exceeding 200 km, CSS combined with MOSAs could be one solution for long
UHV transmission lines.
The possibility of greater compactness related to the design of transmission towers and lines,
given by the application of CSS on UHV OH line circuit breakers, associated with shunt reactors
and MOA applications for TOV mitigation, should be thoroughly investigated during the
planning or engineering design stage of UHV systems.
As information technologies progress, it may become possible to use CSSs for fault current
interruption, uprating of modern and aged circuit breakers and compensated line auto-reclosing
with a minimum of surge arresters. Furthermore, various monitoring results of GCBs recorded in
the controller can be used for remote diagnostics and condition-based maintenance in order to
improve equipment reliability further and optimise maintenance practise.
146
3.4 DS, ES
3.4.1 DS capacitive switching
3.4.1.1 Introduction
Disconnecting switches normally have a capacitive current interrupting capability, however, there
are no standardized interrupting ratings for the capacitive current. For air-insulated disconnectors,
the interrupting current does not exceeded 2 A for the system operating voltages up to 800 kV.
For GIS disconnectors, the interrupting current ranges from 0.1 A at 72.5 kV and up to 0.8 A at
800 kV according to IEC 62271-102 [61]. Maintenance Team 42 (MT42) within IEC is
investigating this capability with particular reference to air-insulated disconnectors and how type
testing should be performed [62].
A very fast transient overvoltage (VFTO) is generated by disconnector (DS) operations during
switching of small capacitive currents due to the repetition of restriking between the disconnector
contacts, because the operating speed of DS is relatively slow (2-3m/s even in case of GIS DS)
compared with a circuit breaker. The maximum overvoltage is up to 3 p.u.

The basic circuit for capacitive current switching is shown in Fig. 3.4.1. Recent research of
capacitive current interruption in atmospheric air has shown that capacitive current switching
using disconnectors is more complex than previously envisaged [63]. Successful current
interruption is dependent on the magnitude of the current (as determined by the value of the load
side capacitance CL) and on the ratio of the source side capacitance CS to CL. For high values of
CS/CL (CS/CL > 1), recovery after current interruption is mainly a dielectric event with shorter
arcing times and lower overvoltages. For low values of CS/CL (CS/CL < 1), recovery is a
thermal event due to restriking injected energy with longer arcing times and higher overvoltages.
The longer arcing times in the latter case means longer arc lengths that promote partial arc
collapses and even longer arcing times.
Fig. 3.4.1 Basic capacitive current switching circuit

The influence of the CS/CL ratio can be understood by reference to Figures 3.4.2 and 3.4.3.
When a restrike occurs, the voltages on the source and load sides of the disconnector will
equalize at the equalization voltage. At CS/CL > 1, the equalization voltage is close to the source
side voltage (A in Fig. 3.4.2) and at CS/CL < 1 is closer to load side voltage (B in Fig. 3.4.2). The
energy injected during equalization is not significant but that due to recovery to the source
voltage can be significant. The greater the difference between the equalization voltage and the
147
source voltage, the greater the energy injected to sustain the arc as shown in A and B in
Fig. 3.4.4.
1. 5
1. 5
1
1
0. 5
0. 5
0 0
- 0. 5 - 0. 5
- 1 - 1
load voltage
- 1. 5 (0.5 pu/div) - 1. 5
- 2
source voltage - 2
(0.5 pu/div)
- 2. 5
A Cs/Cl = 2.5 - 2. 5 2.3 pu B Cs/Cl= 0.04

670 6 75 68 0 68 5 690 6 95 7 00 13 90 1 395 140 0 1 405 1410 14 15 1420
Fig. 3.4.2 Restriking process without overvoltages (A) and with overvoltages (B) at interruption
of 2 A capacitive current with disconnector. Load side voltage (solid lines) and source side
voltage (dashed lines) are shown. Time: 5 ms/div. 1 p.u. = 245 kV (Traces courtesy of KEMA)
Fig. 3.4.3 Restriking process without overvoltages (A) and with overvoltages (B) at interruption
of 2 A capacitive current with disconnector. Load side voltage (solid lines) and source side
voltage (dashed lines) are shown. Time: 5 ms/div. 1 pu = 245 kV (Traces courtesy of KEMA)
Fig. 3.4.4 Restrikes in disconnector arc. Current: 2 A capacitive. 1 pu = 245 kV. A: CS/CL = 2.5; B: CS/CL
= 0.04. Upper trace: arc current; middle: voltage across disconnector blades; lower: energy in the restrike.
Time: 5 ms/div (Traces courtesy of KEMA)
Analysis of the restriking event shows that the equalization voltage UE is given by:
148
UD
UE = US −
1 + C S CL …(29)
where US is the source side voltage at restriking and UD is the voltage across the disconnector
prior to restriking. The maximum overvoltage on the load side UL is given by:
 UD 
UL = US + β 
 1 + C S CL  …(30)
where β is a damping factor of value less than 2.
The maximum current peak IL through the load is given by:
CL  1 
IL = U R  
LS  1 + C C L 
S
…(31)
where UR is the difference between US and UE. Equations (29), (30) and (31) clearly show the
interactive nature of capacitive current switching. Equation (31) in particular shows that the
source impedance needs to be accurately represented to demonstrate disconnector performance in
a type test.
In the case of GIS disconnectors, a slightly complicated circuit should be considered for
capacitive current switching. First, Ls is divided into Lss and Ls-GIS, and Cs into Css and Cs-GIS
as shown in Fig. 3.4.5. Here, Ls-GIS and Cs-GIS are inductive and capacitive impedance within the
GIS. However, Cs-GIS is quite small and then, the modified basic circuit for a full GIS will be as
shown in Fig. 3.4.6. In this basic capacitive current switching circuit, Css is usually much larger
than CL and then higher surge is generated. The Ls-GIS is also small because it is an impedance
within the GIS and, the restriking surge is of high frequency. Another point to be considered is
that the interrupting ability of GIS disconnecter is so high that the current is immediately
interrupted after a re-strike. A current that continues to flow for half cycle as shown in Fig. 3.4.2
(B) is seldom observed.
Disconnector
Lss Ls-GIS
U Css Cs-GIS CL
Fig. 3.4.5 Theoretical capacitive current switching circuit
149
Disconnector
Lss Ls-GIS
Disharge current
U Css CL
Fig. 3.4.6 Basic capacitive current switching circuit
3.4.1.3 UHV air-insulated disconnectors

The required withstand voltages (Urw) for the air-break DS in 1200 kV systems calculated in
accordance with the IEC 62271-1 practises are shown in Table 3.4.1. [64]
Table 3.4.1 Required withstand voltages for 1200 kV air-insulated disconnectors*

Lightning impulse (kV peak) Switching impulse (kV peak)
Phase-to-earth Across isolating distance Phase-to-earth Across isolating distance
2400 2400 (+980) 1800 1800 (+685)
* Note that actual rated voltages will be higher than these values.
The isolating distance across the open disconnector will probably exceed 10 m based on the
above withstand requirements. Quite apart from the obvious mechanical design constraints in
dealing with the associated blade lengths, in order to achieve the necessary blade speed to reach
the minimum contact gap for current interruption, as soon as possible, UHV disconnectors will
likely consist of two breaks in series.
The bus-work within UHV air-insulated stations will be the flexible (bundled) type and is
expected to have a charging current of about 0.002 A/m. Instrument transformers will be of the
optical type and will not contribute to the capacitive load. The current to be interrupted is
therefore unlikely to exceed 0.5 A, which corresponds to a bus length of 250 m. At such current
levels, current interruption tends to be a dielectric event. The arc will usually be of a “dancing”
nature, i.e. each restrike will be in a new location rather than in the previous arc channel, and
interruption is dependent on achieving a minimum contact gap capable of withstanding the
recovery voltage. The minimum contact gap is estimated to be at least 4 m.
3.4.1.4 UHV GIS disconnectors

Extrapolation of the test duty 3 specified bus-charging current in Annex F of reference [61]
suggest that the required bus-charging current switching capability of GIS DSs at UHV is in the
range 1 to 1.2 A. However, this remains to be confirmed once actual UHV GIS parameters
become known. A factor that will influence the DS interrupting capability is the perceived need
for opening resistors in order to limit the magnitude of the very fast transients (VFTs) generated
by restriking [65].
150
The information available is on GIS type DSs applied in Japan and China [66],[67]. A summary
of the bus-charging current switching duty for 1100kV DS in Japan is as follows,
a) Recovery voltage
The recovery voltage of bus-charging current switching for 1100kV DS is specified to be the
rated phase to earth voltage 635kV (=1100/ 3 kV) due to the same principle for 550kV
equipment in JEC-2310-2003 “AC disconnectors”.
b) Switching current
The switching current (I), which corresponds to the bus-charging current, is specified to be 0.5 A
derived from the following equation,
I = ωCV (32)
Where C is the load side capacitance, and is assumed to be 2000 pC, which is the maximum
value estimated by the actual UHV layout.
3.4.1.5 VFTO and its countermeasure

Very Fast Transient Overvoltages (VFTO) up to 2.8 p.u. were generated due to reignition during
the closing and opening operation of a disconnector switches (DS). A VFTO with high frequency
components may cause severe dielectric stresses to the substation equipment, especially the
windings of power transformers and potential transformers, because of its steep rate of change of
voltage. It will propagate through the GIS enclosure and raise the grounding potential of the
equipment to the earth. If a trapped charge remains on the grading capacitors of GCB, the
amplitude of the VFTO will increase due to the offset of the transient voltages, which may also
cause severe mechanical stresses to the substation equipment.
A switching resistor either in parallel or in series is often employed to reduce the amplitude of
VFTOs generated by DS switching. A DS with a switching resistor is normally required for
extended mechanical endurance performance due to frequent operations.
For 550kV systems, VFTOs are covered by the lightning impulse withstand level (LIWV).
However the VFTO for an 1100kV disconnector has to be studied and the proper countermeasure
here to be taken because the VFTO could be beyond LIWV, whose value has been determined
from the total insulation coordination study for 1100kV systems.
An introduction to the bus-charging current switching duty for the 1100kV disconnectors used in
Japan and China, the calculation results of VFTO analysis to estimate the effect of disconnector
with switching resistors fitted and the required specifications of the switching resistor on the
disconnector are described below.
3.4.1.6 Analysis of VFTO in full GIS

To confirm the effect of the switching resistor and to determine its value, the VFTO analysis was
carried out in Japan for the following substation arrangement and employed DSs with and
without switching resistors. The substation has a double-busbar scheme and consist of a GIS with
four feeder bays and four transformer bays. The switching condition of DS is the occurrence of
restriking with the remaining voltage of 2 p.u. between the disconnector contacts. There are five
types of DSs, a line-side DS, a bus-bar side DS with line feeder, a bus-tie DS, a bus-bar side DS
for the transformer bay and a bus-section DS. The resistor value for the DSs was varied to a
151
maximum 1000 ohms and the calculation for a disconnector without a switching resistor (2 ohms
of arc resistance), was also carried out for reference.
Figure 3.4.6 shows the single line diagram used for the VFTO calculation. Table.3.4.2 shows the
specific constant values used for the calculation. Fig.3.4.7 shows the calculation results.
Fig. 3.4.6 Single line diagram for calculation
Fig. 3.4.7 Calculation results (without switching resistor)
Table.3.4.2 Specific constant values for calculation

Items VALUE
GIS busbar Surge impedance 90 ohm
Propagation velocity 270 m/us
Transformer Capacitance 4600 pF
Lightning arrester Simulate as the minimum residual voltage V10kA=1555 kV
Disconnector Simulation for restriking arc Switching resistor 0 - 1000 ohm
Arc resistance 2 ohm
Restriking Source side :+1p.u.
condition Load side :-1p.u.
Transmission line Surge Impedance 230 ohm
From Fig. 3.4.7, it is understood that a maximum overvoltage of 3.25 p.u. occurred and the
overvoltage exceed LIWV when using a disconnector without a switching resistor on the one-
152
feeder bay plus (+) one-transformer bay condition. The overvoltage value goes on decreasing as
the resistance of the switching resistor increases, then saturates at approximately 1 p.u.
It is confirmed that the overvoltage value can be suppressed by using a switching resistor, for
example 1.13 p.u. (1015 kV) with a 500 ohm resistance and 1.07 p.u. (=961 kV) with 1000 ohms
Additionally, using the switching resistor is effective in increasing the overall reliability of the
dielectric capability by decreasing the induced voltage to the sheath and the control circuit. On
the other hand, increasing the resistance increases the voltage applied to the resistor, then
increase the resistor and increase the overall size of disconnector. The resistance of the switching
resistor is determined to be 500 ohms, which is effective for decreasing the overvoltage to 1.2
p.u. within the saturated region.
In Japan, field verification tests have been carried out where facilities are connected to a 550 kV
commercial network through the transformer. The disconnector surge tests were performed with
four different circuit configurations. Figure 3.4.8 shows an example of the VFTO generated by
disconnector operation [68]. It was confirmed that the maximum VFTO level could be
suppressed to less than 1.2 p.u. in the field. From the electromagnetic immunity (EMI) viewpoint,
the potential rise at the GIS tank was measured. It was confirmed to be no more than 10 kVp, and
there was no malfunction in the control and protection systems.
Fig.3.4.8 Example of waveform
The VFTO generated in 550 kV systems was evaluated and the results showed that it will not
exceed 2.8 p.u. The insulation coordination in the 1100 kV systems ensures that the VFTO level
due to disconnector switching will not exceed the LIWV level. A resistive making and breaking
method with a 500 ohm resistor was employed to reduce this VFTO level. The VFTO can
effectively be reduced to less than 1.3 p.u. due to the switching resistor [66]. The VFTO can be
successfully suppressed in 1100 kV systems by using DS equipped with a switching resistor. This
153
insulation coordination policy secures the VFTO less than the LIWV level of the substation
equipment.
3.4.1.7 Analysis of VFTO in hybrid GIS

VFTOs up to 2.8 p.u. were generated due to reignition during the closing and opening operation
of a disconnector switches (DS). A switching resistor either in parallel or in series is often
employed to reduce the amplitude of VFTO generated by DS switching.
Figure 3.4.9 shows a single line diagram for an1100 kV hybrid GIS planned in China [68], which
employs a one and half GCB arrangement for the first and future stages of the substation layout.
In EMTP analysis, the resistance of the GIS conductors and overhead lines were assumed to be
proportional to the square root of the dominant frequency of the surges considering the skin effect
under high frequency.
Fig.3.4.9 Single line diagram of 1100 kV Hybrid GIS planned in China
2500 2500
2442 kV (2.7 p.u.) 1922 kV (2.1 p.u.)
2000 2000
1500 1500
1000 1000
Voltage VS [kV]
Voltage VS [kV]
500 500
0 0
-500 -500 Configuration : Hybrid-GIS, RDS = 0 Ω
Configuration : Hybrid-GIS, RDS = 0 Ω Open : DS-11A, CB-12
-1000 Open : DS-11A, CB-12 -1000 Close :
Close : Line 1L in service at +1.0 p.u.
-1500 Line 1L in service at +1.0 p.u. -1500 Bus between CB-12 and DS-12A charged at -1.0 p.u.
Bus between CB-12 and DS-12A charged at -1.0 p.u. Operation : DS-12A
-2000 Operation : DS-12A -2000
-2500 -2500
0 5 10 15 20 0 5 10 15 20
Time t [µs] Time t [µs]
(a) Full-GIS (b) Hybrid-GIS

Fig. 3.4.10 Typical analysis of VFTO due to DS operation without a closing resistor
Fig. 3.4.10 shows typical results of VFTO analysis for DS operations without a switching
resistor. The maximum value of the VFTO is 2.7 p.u. (2442 kV) for a full GIS and 2.1 p.u.
(1922 kV) for a hybrid GIS. Since the results indicate that the VFTO may exceed the insulation
level (LIWV) of 2400 kV for a full GIS, a solution is required. Figure 3.4.11 shows typical
results of VFTO for DS operations with a switching resistor (500 ohms), which can reduce the
VFTO to less than 1.2 p.u.
154
Table 3.4.2 (a) VFTO peak after DS Fig. 3.4.2 (b). VFTO peak after DS
open without a closing resistor open with a parallel resistor
2500 2500
2000 2000
1066 kV (1.2 p.u.) 1026 kV (1.1 p.u.)
1500 1500
1000 1000
Voltage VS [kV]
Voltage VS [kV]
500 500
0 0
-500 Configuration : Full-GIS, RDS = 500 Ω -500 Configuration : Hybrid-GIS, RDS = 500 Ω
Open : DS-11A, CB-12 Open : DS-11A, CB-12
-1000 Close : -1000 Close :
Line 1L in service at +1.0 p.u. Line 1L in service at +1.0 p.u.
-1500 Bus between CB-12 and DS-12A charged at -1.0 p.u. -1500 Bus between CB-12 and DS-12A charged at -1.0 p.u.
Operation : DS-12A -2000 Operation : DS-12A
-2000
-2500 -2500
0 5 10 15 20 0 5 10 15 20
Time t [µs] Time t [µs]
(a) Full-GIS (b) Hybrid-GIS

Fig. 3.4.11 Typical analysis of VFTO due to DS operation with a parallel resistor
In China, the VFTO was evaluated at the 1100 kV Jindongnan GIS substation. In this system
configuration, the maximum VFTO attains 2.51 p.u. (2260 kV) after the DS operation without a
switching resistor as shown in Table 3.4.2. A DS equipped with an appropriate resistor can
successfully suppress the VFTO less than 1.27 p.u. (1144 kV). The optimal resistance is
estimated as 500-700 ohms in parallel [69].
The VFTO generated in 550 kV systems was evaluated in Japan, and the results showed that it
will not exceed 2.8 p.u. The insulation coordination in the 1100 kV systems ensures that the
VFTO level due to disconnector switching will not exceed the LIWV level. A resistor making
and breaking method with a 500 ohms resistor was employed to reduce this VFTO level. The
VFTO can effectively be reduced to less than 1.3 p.u. due to the switching resistor [68].
Field verification tests have been carried out in facilities connected to the 550 kV commercial
network through the secondary side of the transformer since May 1996. The total energized hours
reached more than 58,000 by the end of March 2007. Disconnector surge tests were performed
with four different circuit configurations. It was confirmed that the maximum VFTO level, could
be suppressed to less than 1.2 p.u. in the field. The potential rise at the GIS tank was measured
from the electromagnetic immunity (EMI) viewpoint. It was confirmed to be no more than
10 kVp, and there was no malfunction occurred in the control and protection systems.
155
In conclusion, the VFTO can be successfully suppressed in 1100 kV by using a DS equipped with
a switching resistor. The insulation coordination policy ensures a VFTO less than the LIWV level
of the substation equipment.
3.4.1.8 Requirement for the switching resistor of DS

The resistor for DS switching needs special duties on its dielectrics and thermally.
a) Dielectric capability
Fig.3.4.12 shows the calculation results of the voltage applied to the switching resistor. From the
results, dielectric capability of 1700 kV is sufficient, because the maximum voltage occurred by
restriking at +1p.u. of the source voltage and -1p.u. of the remain voltage on the load side for the
four line feeder and the four transformer feeder substation.
Fig.3.4.12 Calculation results (with switching resistor)
b) Thermal capability
The thermal capability of switching resistor should withstand the thermal stress of one CO
operation because the possibility of more than one CO operation within a few minutes, which
corresponds to the radiation time constant of the switching resistor, is estimated to be small.
3.4.1.9 Considerations for structure with switching resistor

An example of a GIS disconnector structure fitted with a resistor is shown in Fig.3.4.13. In the
Fig.3.4.14, when the movable and stationary electrodes start to separate, the charging current is
interrupted once because the current is small. However, a restriking arc generates because
insulation recovery characteristics is not sufficient (Step 1). As the distance between the two
electrodes increases, the restriking arc moves to the part between the resistor shield and the
moveable electrode (Step 2). The restriking repeats and current is completely interrupted when
the sufficient insulating distance is provided (Step 3). Inserting the resistor into a discharge path,
where the gap between electrodes is long and discharge voltage is high, is the principle to
suppress the overvoltage.
156
Fig 3.4.13 DS structure
Fig. 3.4.14 Surge suppressing concepts in switching process of 1100kV DS
3.4.1.10 Discussion
UHV capacitive current switching with DSs will require further review when MT42 completes its
task in late 2008 or early 2009. Of particular interest is how to address the influence of CS/CL in
type testing.
157
3.4.2 Bus transfer current switching
3.4.2.1 Introduction
Disconnector bus-transfer current switching ratings are stated in Annex B of reference [61]. The
rated bus-transfer current for both air-insulated and gas-insulated DSs is 80% of rated normal
current and normally not expected to exceed 1600 A. The rated bus-transfer voltages for DSs at
245 kV and above are shown in Table 3.4.3:
Table 3.4.3 Rated bus-transfer voltages for DS

Rated voltage (kV) Air-insulated disconnectors (Vrms) Gas-insulated disconnectors (Vrms)
245 200 20
300 200 20
362 200 20
420 300 20
550 300 40
800 300 40
Corresponding loop impedances are simply the bus-transfer voltage divided by the bus-transfer
current. Air-insulated DSs commonly include commutating contacts to achieve this capability.

Bus-transfer switching known as loop switching, is the commutation of current from one circuit
to a parallel circuit. With reference to the circuit shown in Fig. 3.4.15, as the disconnector opens,
the arc voltage builds forcing more and more of the current is into the parallel path. The recovery
voltage for this case is the power frequency open circuit voltage across the disconnector after the
current has transferred to the parallel path. The parameters to be considered in analyzing the
circuit in Fig. 3.4.15 are as follows:
Fig. 3.4.15 Equivalent circuit for bus-transfer switching
• Initial current before the disconnector is opened IS;

• Switch current iS as the disconnector is opening;
• Total current IT through both the disconnector and the parallel path; IT is a constant.
• Series and parallel impedances XS and XL; XS + XL = loop impedance XLOOP;
• Arc voltage ua;
• Open circuit voltage UOC across the disconnector as given by XLIT.
• Switching event is non-linear because ua is a non-linear quantity. We can write:
diL di
LL = L S S + ua …(33)
dt dt
158
and
iT = IT sin ωt
= i S + iL …(34)
,where L L = X L ω and L S = X S ω .
Equation (25) then becomes:
X L d(i T − i S ) X S di S
= + ua …(35)
ω dt ω dt
and differentiating the left term
X LIT cos ωt =
1
(X S + X L ) diS + u a …(36)
ω dt
Equation (26) is the basic equation that describes loop switching. Under initial steady state
conditions, the switch is closed and ua = 0:
1
X LIT cos ωt = (X S + XL ) d (IS sin ωt ) + 0
ω dt
= (X S + X L ) IS cos ωt …(37)
giving
XL (IT − IS ) = XSIS …(38)
Under final steady state conditions, the switch is open and IS = 0:
XLIT cos ωt = 0 + ua …(39)
i.e. the arc voltage equals the open circuit voltage across the switch. The condition to be met is
therefore that the arc voltage must rise to a value equal to the open circuit voltage across the
switch in order to totally transfer the current to the parallel path. The arc will actually become
unstable at an arc voltage lower than that given by Equation (39) and dependent on the power
input to the arc. The arc collapses and the arc voltage across the disconnect switch then jumps to
the open circuit value [62].
Equation (39) is also significant in that the circuit shown in Fig.3.4.16 can represent it. The
circuits shown in Figs. 1 and 2 are electrically equivalent and either one can be used for
calculation or test purposes. The circuit in Fig. 3.4.16 is that specified in reference [61].
159
XS + XL is
ua
XLI Tcos ω t
Fig.3.4.16 Bus-transfer switching circuit based on Equation (26)
The recovery voltage across the disconnector is the power frequency open circuit voltage and is
not a transient recovery voltage as implied in reference [61]. Once a current transfer has occurred,
the reverse is not possible by "reignition" or "restrike" since this would mean transfer to an
infinite impedance circuit.
3.4.2.3 Discussion
In the past loop switching with DSs has tended to be used in double-busbar layout schemes. Also,
there is some indication that UHV transmission systems may operate on a constant power flow
basis and the present upper limit of 1600 A bus-transfer current will need to be reviewed.
Likewise upper bus-transfer voltages of 300 V and 40 V air-insulated and gas-insulated DSs,
should also be reviewed based on the actual circuit layout dimensions. One example, however, is
that discussed in the appendix below for the Japanese UHV Project. That application is probably
unique to utilities in Japan and requires a bus-transfer voltage (recovery voltage) for 1000 kV
GIS of 300 V (margin included) at 8000 A. The corresponding TRV has a minimum rate of rise
of 400 V/µs again with a margin included.
3.4.2.4 Bus-transfer switching in Japanese UHV Project

A GIS double busbar configuration is used in the UHV project in Japan [66] and the bus-transfer
switching duty for 1100kV DS was determined by the following considerations.
1) Loop current
The rated loop current of an 1100kV DS is determined to be 8000A, which corresponds to the
rated current of the GIS. In Japan, it is general practise that the maximum loop current
corresponds to the rated current of the DS.
2) Recovery voltage
The recovery voltage (or bus-transfer voltage) of 1100kV DS (V) is derived from the following
simple equation:
V=ωL*d*I, where
L: inductance of the GIS busbar = 0.3µH/m
d: maximum loop length of circuit = 250m
I: loop current = 8000A (correspond to the rated current)
160
The rated recovery voltage is determined to be 300 V with allowance for a margin, which is the
same value as required for 550kV DSs in Japan.
3) TRV
The TRV of an 1100kV DS is derived from the following equation:
Vp= 2 ωZI, where
Z: surge impedance = 90ohm
I: loop current = 8000A
The rated TRV is determined to be 400V/µs again with allowance for margin.
Fig.3.4.17 Bus-transfer operation in double busbar four bus-tie system
3.4.3 Induced current switching for earthing switches (ES)

3.4.3.1 General
For earthing switches (ES) which are used in circuits with transmission line in parallel with the
energized other transmission line, a switching capability is required for the electromagnetically
induced current and the electrostatically induced current. (refer to Fig.3.4.18).
In Annex C of reference [61] an electromagnetic and electrostatic induced current and voltage are
standardized for 550 kV and 800 kV ES which are designated to be used in circuits with
relatively long lines or a high coupling to an adjacent energized circuit (class B earthing
switches). See Table 3.4.4.
Fig.3.4.18 Electromagnetically and electrostatically induced current switching
161
Table 3.4.4 Induced current and voltage for 550kV and 800kV earthing switches (class B)
Rated voltage Electromagnetic coupling Electrostatic coupling
(kV) Rated induced Rated induced Rated induced Rated induced
current (A rms) voltage (kVrms) current (A rms) voltage (kVrms)
550 160 20 25 25
800 160 20 25 32
For the 1100kV ES, the characteristics of the induced current should be estimated individually
because of the difference in the conditions, such as the rated current, the configuration of the
transmission towers and the length of the parallel transmission lines.
3.4.3.2 Duty for 1100kV Earthing switch (ES)

An 1100kV earthing switch (ES) is used for earthing the main circuit of feeders or main busbars
under de-energized condition. For actual operations, an induced current switching capability is
also requested for the earthing switch because some currents are induced from the adjacent
transmission line which is under operation. (refer to Fig. 3.4.18) The rated induced current
switching condition is determined from the analyses of the electromagnetically and
electrostatically induced current for various switching conditions of the earthing switch under the
single line operation of a double transmission line configuration.
1) Rated induced current switching condition
The rated induced current switching condition (Table 3.4.5) is determined from the analysis of
the electromagnetically and electrostatically induced current switching taking account of the
actual length of the 1100kV transmission line and the load current, and selected the severest duty
among the electromagnetically induced current.
Table 3.4.5 Rated induced current switching condition

Items Condition
Line feeder ES (Not applicable for main
Application of induced current switching duty
busbar, Tr feeder and maintenance ES
Induced current Switching current 1000 A
Recovery voltage 70 kVrms
Switching
TRV 160 V/us
condition
Electrostatic Switching current 40 A
induced current Recovery voltage 50 kVrms
Switching operation cycles 100 times
a) Electromagnetically induced current switching

The calculation was carried out by assuming a 200 km length transmission line and a load current
of 8000 A for one operating line with a double transmission line configuration and the other in a
de-energized condition. The well-known Carson-Pollaczek method was used for calculation and
the other detailed conditions are as follows,
The configuration of the conductors of the transmission line and towers are shown in Fig.3.4.19
and Table 3.4.6. The maximum current obtained from the calculation is shown in Table 3.4.8
together with the calculation results of a length of 250 km.
162
D1: 38.0 m
Grounding
Wire
D3: 31.0 m D2: 12.5 m

White
D5: 32.0 m D4: 17.5 m
Red
D3: 33.0 m D6: 17.5 m
Black
Circuit 2 Circuit 1
D8: 61.5 m
Fig.3.4.19 Analysis model of the tower
Table 3.4.6 configuration of conductors of transmission line for calculation

Overhead grounding wire Overhead Power line
2 2
Type and size of lines OPGW, 500 mm x 2 wires ACSR, 810 mm x 8 wires
Resistance of conductor 0.103 ohm/km 0.05 ohm/km (per wire)
Relative permittivity 5.0 1.0
Current for energized line 8000 A
Transmission line length 200 km (250 km for reference calculation)
Table 3.4.7 Switching condition for ES (for Electromagnetic induced current)

Condition of objective ES on de-energized
Condition of remote end ES on de-
Case No. line
energized line
A phase B phase C phase
Case 1 close close close
Case 2 open close close
Case 3 close open close
Case 4 close close open
Closed condition for all 3 phases
Case 5 open open close
Case 6 open close open
Case 7 close open open
Case 8 open open open
Table 3.4.8 Results of electromagnetically induced current switching for 1100 kV ES

Line length: 200 km Line length: 250 km
Current for energized line 8000 A 8000 A
Switching current 810 A 810 A
Recovery voltage 65.0 kV 81.3 kV
TRV 162 V/µs 162 V/µs
From the above results and the comparison with the standard value for a 550 kV ES (rated current
of 8000 A and line length of 200 km) in JEC 2310-2003 “AC Disconnectors”, the values of
switching a 1000 A current and 70 kV recovery voltage which were the same values for the 550
kV ES, were selected as the switching duty. For the TRV value, 160V/µs which was obtained
from the calculation results and was more severe than the standard value for a 550 kV ES was
163
selected. The calculation results of the 200 km line length were chosen because the length of 250
km line was too severe for taking into account of actual system condition.
b) Electrostatically induced current switching
The calculation for electrostatically induced current switching was carried out under the same
conditions as the electromagnetically induced current switching except for the following remote
end ES condition. For the switching conditions of the ES, the remote end ES on the de-energized
line was open condition for all three phases though it was closed for electromagnetically induced
current switching.
Table 3.4.9 Switching condition for ES <for Electrostatic induced current>

Condition of objective ES on de-energized line Condition of remote end ES on de-
Case No.
A phase B phase C phase energized line
Case 1 close close close
Case 2 open close close
Case 3 close open close
Case 4 close close open
Opened condition for all 3 phases
Case 5 open open close
Case 6 open close open
Case 7 close open open
Case 8 open open open
The calculations were carried out under above conditions shown in Table 3.4.9 and the maximum
current obtained from the calculation is indicated in Table 3.4.10 together with the calculation
results of 250 km of length. From the results, the values of switching current 40 A and recovery
voltage 50 kV, which were obtained from the results for 200 km line length and 8000 A rated
current, have been selected as the switching duty.
Table 3.4.10 Results of electrostatically induced current switching for 1100 kV ES

Line length: 200 km Line length: 250 km
Current for energized line 8000 A 8000 A
Switching current 31.8 A 40 A
Recovery voltage 45.4 kV 45.4 kV
c) Test duty of induced current switching

Because 100 times make-break operating cycles is specified for induced current switching in
JEC-2310-2003 “AC Disconnectors” [70], the same make-break operating cycles (100 times) was
adopted to 1100 kV ES.
3.5 Secondary arc current extinction, single-phase auto reclosing
3.5.1 General
Secondary arc induced by sound lines with the system voltage is observed when a single phase
grounding fault is generated. A single phase or multi-phase auto reclosing can be completed
successfully after the secondary arc extinction. System voltage, line configuration and length as
well as wind speed affect the secondary arc extinction time. The secondary arc generated in 345-
765 kV (EHV) OH-lines will not generally exceed 100 A and the arc is generally self-
extinguished within one second if the current is smaller than 40 A. However, secondary arcs with
higher current generated in UHV OH-lines are supposed not to self-extinguish.
164
In very long arcs typical for EHV/UHV insulation the non-uniformity along the arc length and
internal exchange of energy between parts of arc channel play more important role in secondary
arc extinguishing than its elongation, creating intermittent character of arc [69]. As the insulation
length is usually designed to be directly proportional to maximum operating voltage, and the non-
uniformity in longer arcs is more pronounced, the extinction of secondary current arcs in tests on
1000-1200 kV lines happened faster and at higher currents in comparison with observed in
operation of 330-750 kV lines [39].
Figure 3.5.1 shows motions of primary and secondary arc. Secondary arc ignites after primary arc
extinction by clearing the fault, and extends by elevation of itself and influence of wind and post
arc gas of primary arc. Existing conductive gas or atmosphere after arc extinction, it takes certain
time to recover insulation strength to withstand reclosing surge voltage.
The reclosing process is illustrated in Fig. 3.5.2. When secondary arc does not ignite or
extinguishes instantly, the time for reclosing is evaluated by only the recovery characteristic of
after primary arc extinction. The time for reclosing is evaluated by the longer one of recovery
time after primary arc or sum of secondary arc extinction time and recovery time.
extinction secondary floating of extinction extinction of

fault insulation
of primary arc post arc of post arc secondary
occurrence recovery
arc occurrence gas gas arc
primary arc secondary arc

extension of secondary conductive
post arc gas of arc by influence of atmosphere after
primary arc wind and floating of extinction of
post arc gas secondary arc
Figure 3.5.1 Primary and secondary arc motion
re-closing time*
(the longer time of below)
Insulation Recovery after

Primary Arc
Primary
Arc Secondary Insulation Recovery
Arc after Secondary Arc
time
fault clearing and extinction of
re-closing
extinction of secondary arc
primary arc
duration of post arc gas of
fault occurrence primary arc
Figure 3.5.2 Reclosing process after fault occurrence
(*The definition of reclosing time generally includes the time after fault detection.)
Fig. 3.5.3 shows the primary arc (10 kA peak) followed by the secondary arc from a staged fault
test on the BC Hydro 500 kV system. The initial intensity of the secondary arc immediately after
primary arc extinction and the gradual recovery is clearly evident. The primary and secondary arc
165
currents were measured by means of an optical current transformer and a typical trace is shown in
Fig. 3.5.4.
Fig.3.5.3 Secondary arc current behaviour (Courtesy of BC Hydro)
166
0.6
0.4
0.2
Current (kA)
0.0
-0.2
-0.4
-0.6
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35
Time (s)
Fig. 3.5.4 Primary and secondary arc currents (Trace courtesy of BC Hydro and NxtPhase Corporation)
Figure 3.5.5 summarizes the results of secondary arc tests including arc extinction time and
recovery peak voltage investigated in 330-1200 kV lines.
The arc extinction time of the secondary arc (with a current of 150 A and a peak recovery voltage
of 140 kV) is about 4 seconds for 1100 kV double-circuit lines of 210 km-length in Japan.
Russian experience showed that the 90 % values of the extinction times were evaluated as 0.7 s at
40 A peak, 1.2 s at 60 A peak and 2.6 s at 90 A peak. The secondary current peak of Russian
1200kV compensated lines may exceed 150A and the correlation of arc extinction time was given
[71]. A limited number of a single grounding fault tests conducted on the Russian 1200 kV lines
showed shorter arcing times as compared with those of 787 kV lines.
Arc extinction time of secondary arc
Gap distance: 5.5m, Primary arc: 10kA,
8
Extinction time of secondary arc (sec)
Arcing time: 0.1sec, Wind velocity: 2m/sec
7 160 A
765kV-240km compensated line

6 with modified 4 legged reactors (40A)
80 A
5 230-765kV compensated lines 1100kV-210km uncompensatd line
with 4 legged reactors (20A) (150A-140kV) 40 A
4 1200kV-500km compensated line
with 4 legged reactors (106A)
20 A
3 765kV-350km compensated line
with 4 legged reactors (64A)
2 765kV-90km compensated line 10 A
wo neutral reactor (38A-94kV) secondary
1 arc current
550kV-100km uncompensated line (30A-60kV)

0
0 50 100 150 200 250 300
1100kV-210km
line with HSGS Recovery voltage after secondary arc extinction (kV)
Fig.3.5.5 Arc extinction time of secondary arc

A four-legged shunt reactor equipped with an appropriate reactance at neutral point is often used
for shunt compensated EHV and UHV transmission lines. The arc extinction time of the
secondary arc with the current of 38 A and the peak recovery voltage of 94 kV is about 1.7
seconds for the Canadian 765 kV transmission lines without a four-legged shunt reactor,
167
however, it is reduced to less than 0.5 seconds with the application of 1500 ohm reactance at the
neutral point, resulting in a secondary arc current of 20 A. In Russia, the secondary current had to
be limited to below 90 A for 787 kV lines in order to limit the maximum dead time of single-
phase auto-reclosing (SPAR) less than 3.0-3.5 s.
In transposed lines a 4-legged reactor might present a simple solution to secondary arc
suppression. Its application either uses the existing insulation of the neutral point of shunt reactor
or requires its moderate increase. In Russian practise the insulation of 750 kV shunt reactor
neutral has class 35 kV (1-min withstand voltage 85 kV). To exclude permanent flow of possible
unbalance current through neutral reactor it is normally shunted by a circuit-breaker; and to limit
overvoltages on the neutral reactor it is equipped with a MOV.
The use of high speed grounding switches (HSGS) is one of the solutions, which immediately
extinguishes the secondary arc of the faulted phase grounded by closing operation of HSGS. For
UHV transmission lines, rapid auto reclosing operation within one second can be realised by
synchronized operations between HSGS and GCB. In general, HSGS will be applied because of
the following reasons:
− HSGS extinguishes considerably large secondary arc current due to the same phase
double-circuit grounding fault in the double-circuit 800 kV or UHV transmission
lines, which can not be extinguished in a second with four-legged reactor.
Especially, for the long untransposed transmission lines, it is difficult to reduce
secondary arc current to 40 A or smaller by four-legged reactors.
− In the short UHV lines of 100-200 km, it is not necessary to apply shunt reactors to
transmission lines to compensate reactive power. Applications of HSGS instead of a
four legged shunt reactor will be economically affordable in such short UHV lines,
HSGS have not been standardized yet. It is often customized to cope with longer arcing times
based on circuit breaker interrupter design although the TRV peak value is relatively lower. Arc
extension up to 80 ms has been developed.
3.5.1 Operating shunt reactors, four-legged shunt reactors

Utilities operate their shunt reactors in several ways, as they may be permanently connected to
one line’s end, may be connected in a switchable way to the line, or both fixed and switchable at
either end, or even connected to the substation rather than the line.
The usual way of operating the shunt reactors is to connect them before energization of the OH-
line in order to avoid too high power frequency voltages due to the Ferranti-effect (TOV). After
loading the line beyond a certain level, some or all shunt reactors are switched off. When the
loading drops under this level one or more shunt reactors are switched in again. During normal
system conditions, connecting or disconnecting shunt-reactors is based on the voltage profile.
Before unloading the line the shunt reactors are connected and may stay connected after de-
energization until the next energization.
In case of faults or problems in the system, the protection and/or transfer tripping will interfere.
Utilities that apply SPAR (single phase rapid auto-reclosing) will allow the faulty phase to be
tripped without the shunt reactor(s) connected, as the voltage on that phase is low anyway. Before
reclosing they automatically will connect the shunt-reactors to all phases. In this way, they are
prepared for the three-phase tripping if the SPAR was unsuccessful. Care has to be taken for
168
single phase ferroresonance, due to the capacitive coupling between the switched off phase with
its capacitance to earth in parallel to the shunt reactor (after arc extinction) and the healthy
phases. Among other solutions, a mitigation is to disconnect the shunt reactor temporary and
connect it after reclosing [3].
Utilities that apply TPAR (three-phase rapid auto-reclosing), have often a permanently connected
shunt reactor at least at one side. Anyway some utilities will delay the tripping until the shunt
reactors are connected, while others specify their circuit-breakers to clear healthy phase
capacitive currents under Ferranti-conditions. In Russia the air-blast circuit-breakers, used to
switch the EHV and UHV shunt reactors, are equipped with arc gaps across a part of the
interrupting chambers, so that at higher system voltages they are connected instantaneously.
Before three-phase reclosing the utilities connect their shunt reactors to the line.
Examples of the policies of several utilities are as follows:
HQ: both line ends switchable SRs (controlled switching; PIR on air-blast technology
and MOSA in combination with modern technology CB). At faults TPAR, but no
guarantee that SRs are connected -> CBs capable to switch out uncompensated lines
(Ferranti-effect).
AEP: prefer PIR above MOSA. For SPAR, AEP uses a special scheme to switch the shunt
reactors in order to extinguish the secondary arc current, as shown in the following
figure.
BC Hydro: one line end a fixed SR, other end switchable SR with fast insertion CBs capable to
switch out uncompensated lines.
Russia: either instantaneous insertion of SR through spark gaps or delay of healthy phase
switching until SRs are connected. Normal operation: connect SRs before switching
off. Automatic insertion in all phases of SR at an overvoltage for a certain period
SPAR with 3-phase insertion of SR before reclosing. Automatic insertion in all
phases of SR at a single pole or there-pole opening command to a line circuit-
breaker.
India: SRs fixed to the lines switchable SR at the busbars.
Turkey: all SR switchable (controlled switching, MOSA). Automatic insertion of SR at
higher operating voltage (within 1.5 s). SPAR, even with single phase disconnection
of SR when ferroresonance due to coupling to healthy phases may occur.
A modified reactor schemes has been applied to some 765 kV transmission lines in the US. Field
tests demonstrated a successful single phase reclosing. Field tests demonstrated a successful
single phase reclosing that if the secondary arc current were limited to less than 40 A rms, it
extinguished within 0.5 second as shown in Fig. 2.2.1
3.5.2 HSGS
3.5.2.1 Developments of HSGS for UHV system in Japan

A fast multi-phase reclosing system was employed to avoid loss of a double-circuit line. Because
of the large charging MVA of a UHV line, the electro-static induction from sound phases may
keep the secondary arc current in such a long period that the reclosing in a second will be
impossible.
169
The duration of the secondary arc in different fault conditions on six phases of a double-circuit
line was evaluated, and a reclosing system featured HSGS within one-second was developed by
TEPCO. With the connection of four-legged shunt reactors, reclosing within one-second is
difficult to realise for all the cases of faults of double-circuit lines.
The UHV experimental characteristics of insulation recovery after primary arc (4 kA, 0.1 sec)
interruption are shown in Fig.3.5.6. This data were carried out outdoors when the wind velocity
was generally lower than 2.5m/s, which corresponds to the wind velocity parameter on this study.
It takes around 0.6~0.9 sec for recovering insulation strength enough to withstand 1.6 p.u. On this
studying condition, possible surge voltage at reclosing is lesser than 1.6 p.u. Arcing on surface of
suspension insulator, which is a rare phenomenon, causes recovery time to be longer.
1.6 p.u.
~0.6sec ~0.9sec
Figure 3.5.6 Recovery characteristic after primary arc extinction [72]

1.2
Switching impulse flashover voltage (Vimp/VimpINF)
1.0
0.8
Arc Current = 100 [A]

0.6
Duration Time of Arc = 0.07 [sec]
0.4
0.2
Breakdown
Not breakdown
0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6
INF
Duration time between secondary arc extinction and switching impulse (Timp) [sec]
Figure 3.5.7 Recovery characteristics after secondary arc extinction [73]
When secondary arc at fault point would be persisted for some duration, there is an insulation
recovery characteristic after secondary arc extinction. The UHV experimental characteristic of
170
insulation recovery after arcing (100A, 0.07 sec) is shown in Figure 3.5.7 carried out outdoors
when the wind velocity was generally lower than 2.0m/s, which corresponds to the wind velocity
on this study. It takes around 0.5 sec for recovering the insulation strength before the fault.
Taking the above mentioned insulation recovery characteristics into consideration, operating
sequence of HSGS as shown in Fig. 3.5.8 will be applied in the UHV system of TEPCO.
①Transmission fault occurs primary arc CB
0sec HSGS HSGS
②Clear the faulted phase secondary arc
0.07sec
③HSGS at each end Clear secondary arc
　close to clear the
　secondary arc 0.27sec
④HSGS at each end open
0.8sec
⑤Circuit breakers reclose
1sec
Fig. 3.5.8 Fast reclosing using High Speed Grounding Switches (HSGS)
3.5.2.2 Delayed zero phenomena caused by successive faults

In addition to the ordinary induced current interruption, a delayed current zero phenomenon
caused by multiple lightning strikes (and therefore successive fault) was specified as a special
duty for TEPCO system as shown in Table 3.5.1. Although the probability of the occurrence of
this phenomenon would be very low, it was determined to achieve high reliability of the UHV
system. The phenomenon is explained as shown in Fig. 3.5.9.
The phenomenon considers a successive ground fault due to multiple lightning strikes or another
(second) phase ground fault is caused during an HSGS operation to interrupt the first ground
fault, At this instant, a larger current is induced by the second fault current and this additional
induced current may offsets the AC component of the originally flowing current (induced by
healthy phase) in the first phase, resulting in a delayed current zero. This condition continues
until the circuit breaker of the second faulted phase interrupts the ground fault current. Then the
HSGS should withstand a long arcing time up to 80 ms without a current zero.
Table 3.5.1 Specifications for HSGS (Delayed current Zero Condition)
Condition OHL length Interrupting Current Recovery voltage Rate of Rise
Electromagnetically 200 km 3500 A mean 170 kV peak 0.26 kV/µs
induced 40 km 7830 A mean 65 kV peak 0.46 kV/µs
Electrostatically 200 km 5800 A mean 570 kV* -
induced 40 km 7000 A mean 390 kV* -
Arcing time: 80 ms + minimum arcing time (with current zero missing)
* (1- cos) waveform
171
Fig. 3.5.9 Delayed current zero phenomena in HSGS
3.5.2.3 Field experiences of HSGS in KEPCO 765kV systems

In Korea, the 765 kV transmission system is to transport large capacity of electric power
generated in the east and west coasts to the Kyungin area and it is estimated that the normal
power will be several million kW. However, the system is using double-circuit steel tower in
order to save the land space and adopted the high speed multi-phase reclosing system for
preventing loss of the double circuits.
In the power system above 550 kV, the secondary arc does not extinguish in a short time induced
electro-statically from other phase or other circuit. Especially in double-circuit transmission lines
such as 765 kV transmission in KEPCO, it is desirable to install HSGS, since such conditions as
the capacitances between each phase of healthy phase and healthy circuit become very
complicated to be compensated and it is not only difficult but also not economic to apply the
four-legged reactors.
Figure 3.5.10 shows the KEPCO 765 kV transmission grid and locations of HSGS installations.
Figure 3.5.10 765kV Transmission Grid in KEPCO and HSGS applications
172
In the short 765 kV transmission lines less than 80 km, an HSGS installed at single end of the
transmission line can extinguish secondary arc current in 750 ms. Therefore, HSGS is installed at
both ends of the 765 kV transmission lines 80km or more and at a single end of those shorter than
80km.
Table 3.5.2 shows the operation experiences of HSGS in KEPCO. HSGS was operated at two or
three times per year per line. KEPCO have not experienced any miss-operations and
malfunctions.
Table 3.5.2 Operation experiences of HSGS in KEPCO
Substation Sinseosan ~Sinansung Sintaebaek ~ Singapyeong
Year (double circuit transmission line) (double circuit transmission line)
2003 12
2004 - 2
2005 2 8
2006 2 6
2007 6 6
Sum 22 22
3.6 Surge Arrester

3.6.1 Continuous operating voltage Uc, rated voltage Ur and TOV
capability
The selection of a suitable surge arrester first comprises the selection of continuous operation
voltage, Uc, and rated voltage, Ur.
Uc is determined from actual service voltage across the arrester, which for normal phase-ground
installation is equal to Um/ 3 . Ur is selected based on amplitude and duration of temporary
overvoltages. Normally arresters are not used to limit temporary overvoltages (TOV) since this
would in general result in enormous arrester energies. During TOV conditions therefore, in
general, the voltage across the arresters is assumed to be unaffected by the arrester.
Applying the same ratio Ur/Um as for 800 kV, i.e. 588/800= 0.735, rated voltages of 809 and 882
would be required for 1100 and 1200 kV systems respectively.
With the same relative protection levels as for the 800 kV arresters this would give minimum
LIPL and SIPL as per Table 3.6.1.
Table 3.6.1 Arrester data extrapolated from 800 kV
Maximum system voltage Extrapolated Rated voltage Extrapolated LIPL Extrapolated SIPL
1100 kV 809 kV 1940 kV 1569 kV
1200 kV 882 kV 2115 kV 1710 kV
Furthermore, applying the same protective ratios LIWV/LIPL and SIWV/SIPL as used for 800
kV result in insulation levels as per Table 3.6.2
Table 3.6.2 LIWV and SIWV of equipment calculated from protective ratios applied on 800 kV
systems multiplied with arrester protection levels extrapolated from 800 kV
Maximum system voltage Extrapolated LIWV of equipment Extrapolated SIWV of equipment
1100 kV 2677-2891 kV 1961-2134 kV
173
1200 kV 2919-3151 kV 2138-2326 kV
Extrapolation from 800 kV equipment thus results in much higher insulation levels than applied
for the UHV systems in Italy, Russia, Japan and China.
Limitation of TOV is vital to obtain an optimal protection by arresters. A lower TOV directly
affects possible protection levels since a lower arrester rated voltage could be selected. If a lower
arrester rated voltage is possible to use the protection level is improved correspondingly. To
improve it by adding more parallel block columns is less efficient.
As an example, for an arrester with multiple columns giving a protection level at 2 kA of 1.7 p.u.
doubling of the number of columns would give a reduction in protection level of only 2-3 %. On
the other hand if the rated voltage could be decreased by 10% the protection level also is reduced
by 10 %.
This is valid until the continuous operating voltage set a limit. Historically the factor Uc/Ur has
not been higher than 0.8 i.e. an Uc of 80 % of the “knee-point” on the voltage-current
characteristics has as maximum been applied for commonly used arresters. Actual ratio is in most
cases less since consideration of temporary overvoltages has required a higher rated voltage. The
maximum possible ratio Uc/Ur to use is mainly determined by the thermal performance of the
arrester housing and power losses of the metal-oxide blocks. (In the past also possible ageing of
the metal-oxide blocks was an issue as well.) The criteria for a maximum Uc are set in the
standards [74]. Energized at Uc and at maximum ambient temperature of 40 ºC (as per IEC) the
average temperature of the arrester shall not exceed 60 ºC if existing arrester standard [74] would
be applicable. Furthermore, thermal stability at Uc must exist after absorbing maximum decisive
energy followed by rated voltage as verified by operating duty tests according to the standard.
If the thermal performance of the arresters could be improved or metal-oxide blocks with lower
power losses used a higher ratio Uc/Ur would be possible. If temporary overvoltages are not
decisive an increased relative Uc would then be possible to make use of to obtain a lower
protection level.
However, the existing standard is here a hindrance to development. In the conditioning part of the
operating duty tests a power-frequency voltage not less than 1.2 times Uc shall be applied.
Practically this voltage must be below the “knee-point” otherwise the test is not possible to
perform since the test samples will be damaged by the power-frequency voltage.
Maximum TOV of 1.4 p.u. with duration 0.4 to 0.5 s is indicated for the 1100 kV UHV system in
China [75], [76] on the line side of the breakers and for this an arrester rated voltage of 828 kV
preliminary has been specified. However, to withstand this TOV a lower rated voltage is possible,
in the range of 740 to 775 kV. This will require a ratio Uc/Ur of 0.82 to 0.86 which also may be
possible to achieve depending on the thermal performance of the arrester design and power losses
of the metal-oxide blocks. At an initial stage in the design like for the current UHV systems it is
therefore unnecessary and even incorrect to specify the rated voltage of the arresters. To obtain
an optimum arrester only the stresses should be given as well as desirable protection levels. The
selection of a suitable rating thereafter should be left to the manufacturer.
In Japan sound phase overvoltage (TOV) of UHV system at one line grounding fault is expected
around 1.1 p.u. of maximum line to ground voltage. However, for the rated voltage of UHV
arresters in Japan, overvoltage factor of 1.3 p.u. was adopted considering the rare TOV, and in
line with effectively-grounded systems such as 550 kV, and then Ur was decided as 826 kV
(=1100 / 3 ×1.3).
174
Figure 3.6.1 shows the lightning surge operating duty test sequence according to Japanese
standard (JEC-217-1984). The rated voltage of surge arrester corresponds to applicable power
frequency voltage that an arrester can deal with lightning surges under the application of the rated
voltage without thermal runaway.
On the other hand, the maximum TOV duties exceeding the rated voltage, is caused by various
supposing factors, such as system conditions, generator constants and generator control systems,
and failure modes considered. Therefore, durability capability of surge arrester for the maximum
TOV is evaluated by the step 4 of stability evaluation test in JEC217-1984 standard as shown
Figure 3.6.2.
Number of Lightning Impulse
(Positive polarity)
1 2 3 4 5
Rated voltage Continuous Operating Voltage
1min Within 5min 30 min
Continuous Operating Voltage

Within 5min 1min
Within 5min 30 min
Rated voltage
6 7 8 9 10
Number of Lightning Impulse
(Negative polarity)
Fig.3.6.1 Lightning surge operating duty test
Step 1 Step 2 Step 3 Step 4
AC voltage test Lightning impulse Rectangular wave current and Temporary overvoltage
high current test thermal stability tests withstand test
Rated voltage Maximum AC TOV

µs
65kA 4/10µ 2000A-2ms Continuous
Continuous Continuous Operating Voltage
Operating Voltage Operating Voltage
cooling
cooling cooling
t(s) residual voltage t(s)
180 h 5min 2min. 1min. 30min. 30min.
measurement
Fig.3.6.2 Stability Evaluation Test according to JEC-217 1984
3.6.2 Residual voltages

Insulation levels selected for UHV systems are well below the levels in Table 3.6.2. Surge
arresters with improved protection characteristics and/or selected with lower relative rated
voltage with respect to the system voltage than for 800 kV therefore are needed. The number and
location of arresters in the substations must also be considered as well as other measures to limit
amplitudes and, in particular, steepness of incoming surges to the stations.
The arresters of the highest class as per the IEC standard are in the switching surge operating
duty test [74] subjected to a conditioning test with a first part comprising 20 8/20 current
impulses with peak value equal to the nominal discharge current i.e. 20 kA followed by 2 high
current impulses 100 kA 4/10. The large clearances on the UHV transmission lines to ensure
necessary switching surge withstand voltage may lead to that a higher nominal current than 20
kA should be introduced. For typically clearances in the towers of 6 to 7.3 m [75], [76] [77] the
175
withstand voltage for negative standard lightning impulse is approximately 3.6 to 4.5 MV.
Assuming a shielding failure at critical current a fast front surge with this amplitude will travel
towards the UHV station. Approximately the current through an arrester with protection level of
1620 kV (selected for 1100 kV systems in Japan and China) and furthermore assuming a surge
impedance of the line of 275 ohm will be 20 to 27 kA.
However, due to the design of the shielding of the line the probability of shielding failures may
be ignored.
2.2
Prospective overvoltage - p.u.
2
Load line 2.2 p.u.
1.8 Load line 2.0 p.u.
Load line 1.8 p.u.
1.6
SIPL 1.6 p.u.
1.4 SIPL 1.7 p.u.
SIPL 1.8 p.u.
1.2
1
0 1000 2000 3000 4000
Current - A
Fig.3.6.3 Load diagram for the range of prospective overvoltages 1.8 to 2.2 p.u. and U-I arrester
characteristics for SIPL 1.6 to 1.8 p.u.
Regarding suitable current amplitude to define the arrester protection level for switching surges 2
kA seems to be sufficient also for UHV applications. From Figure 3.6.3, arrester discharge
currents for prospective overvoltages of 1.8 to 2.2 p.u. and for voltage-current characteristics with
SIPL at 2 kA of 1.6 to 1.8 p.u. could be estimated. For e.g. a SIPL of 1.6 p.u. the discharge
currents range from 0.8 to 2 kA. Taking into account that prospective overvoltages most probably
will be controlled below 2 p.u. 2 kA as coordinating current will be sufficiently high.
In Japan, nominal discharge current for UHV Arresters has been determined to 20 kA from the
simple calculation formula, using spark-over voltage of air gap of transmission line which gap
length is about 6 m and surge arrester V-I characteristics. In addition to this, among the lightning
overvoltages analysis by EMTP program using the surge arrester V-I characteristics as shown in
Figure 3.1.1 and V20kA=1.8 p.u. (1620 kV), maximum discharge current was nearly 20 kA.
Switching impulse discharge current was 1kA or below from the result of switching surge
analysis.
These current values can be categorized as the current values of class 5 surge arresters specified
in Table 3.6.3 taken from [74] (20 kA for lightning and 2 kA for switching).
The lightning and switching insulation withstand test values are generally determined from basic
and simplified formula to multiply levels (LIPL and SIPL) with some factors which is very
convenient for general and relative evaluation methods of systems.
However, in order to decide the lightning and switching insulation withstand test values of UHV
systems, where high reliability and efficiency were considered as essential, detailed overvoltage
analyses have been used in Japan.
176
Table 3.6.3 Peak currents for switching impulse residual voltage test as per IEC 60099-4
Arrester classification Peak currents (A)
20 000 A, line discharge Classes 4 and 5 500 and 2 000
10 000 A, line discharge Class 3 250 and 1 000
10 000 A, line discharge Class 1 and 2 125 and 500
3.6.3 Long duration current impulse withstand

The line discharge test as per IEC [74] is used to verify energy capability of 10 and 20 kA
arresters. The highest class, i.e. class 5, is considered to cover line lengths up to 480 km. For a
588 kV arrester on an 800 kV system the test simulates the discharge of a 480 km line with surge
impedance 294 ohm charged to 2.16 p.u.. For commonly used protection levels the test results in
arrester energies of 5 to 7.5 kJ/kV rated voltage.
Transmission lines for UHV systems may be considerably longer than 480 km and thus
prospective arrester energies much higher. On the other hand it is vital at UHV to keep the
switching overvoltages as low as possible. Single-phase reclosure is e.g. considered as well as
preinsertion resistors and controlled switching which will reduce arrester discharge duties as well.
In Figure 3.6.4 arrester energies as function of prospective overvoltage i.e. the charging voltage
of the transmission line are given with the arrester protection level at 2 kA (SIPL) as parameter
for an 1100kV/1000 km line. The surge impedance of the line is assumed to 275 ohm. The
arrester discharge energy is significantly reduced if the prospective overvoltage without arrester
is limited.
20
Arrester discharge energy - MJ
18
16
14
12
10
8 SIPL 1.6 p.u.
6 SIPL 1.7 p.u.
4 SIPL 1.8 p.u.
2
0
1.7 1.8 1.9 2 2.1 2.2 2.3
Charging voltage - p.u.
Fig.3.6.4 Arrester discharge energy for 1000 km transmission line as function of overvoltage
factor (charging voltage). Arrester protection level at 2kA switching impulse (SIPL) as parameter
1100 kV GIS-arresters in Japan have been confirmed to have the surge suppressing, energy
absorption and thermal stability capabilities by switching surge duty test according to Japanese
standard (JEC-217-1984). The concept of this test was introduced considering that surge
arresters have to suppress various surge voltages and absorb their energies to protect the other
equipments, even if the other surge suppressing equipments and device do not work well. The test
sequence and test circuit are shown in Figure 3.6.5.
The primary purpose of switching surge duty test is to evaluate thermal stability of surge arrester
at Uc after absorbing the switching surge energy. Switching surge duty test supposes the duty of
the GIS-arrester at the receiving end of line, by resistor-closing failure of the circuit breaker. A
circuit breaker connects between peak voltage of rated voltage of arresters on the line and rising
177
voltage of reversed polarity of power source, and evaluates the thermal stability of arresters. That
is, without resistors, considering the most severe cases. Figure 3.6.6 shows an example of the
successful test record of this switching surge duty test.
Number of switching Number of switching
surge operating duty surge operating duty
1 2 3 4 5 6
Continuous Continuous
Pre-heat Pre-heat
5min Within 5min cooling 5min Within 5min

Group I Group II
Test Sequence
Test Circuit
Resistor reclosing failure is supposed .
Power source voltage : Rated voltage of SA

Line charging voltage : Peak value of rated
voltage of SA
Surge Impedance : 240 ohm
Line length : 250km
Line connection : Opposite polarity near 60
degrees during voltage rise
Test Circuit Energy per shot : 20MJ
Fig. 3.6.5 Switching Surge Duty Test according to Japanese Standards (JEC-217-1984)
The above duties, tested in Japan, are considered as very severe. Besides, the existent test
condition according to IEC [74] as shown in Table 3.6.4 seems necessary to be revised, because
for the surge arrester of UHV system, the test conditions are more severe, considering the system
characteristics such as low surge impedance and line lengths as also discussed earlier.
Fig. 3.6.6 Typical Oscillograms of Switching Surge Duty Test
Table 3.6.4 Parameters for the line discharge test on 20000 A and 10000 A arresters (as per IEC 60099-4, [74])
Arrester Line discharge Surge impedance of Virtual duration Charging voltage
classification class the line Z (Ω) of peak T (µs) UL (kv d.c.)
10 000 A 1 4,9 Ur 2 000 3,2 Ur
10 000 A 2 2,4 Ur 2 000 3,2 Ur
10 000 A 3 1,3 Ur 2 400 2,8 Ur
20 000 A 4 0,8 Ur 2 800 2,6 Ur
20 000 A 5 0,5 Ur 3 200 2,4 Ur
178
3.6.4 Lightning impulse discharge capability
The arrester energy from lightning may not be negligible either due to the high insulation strength
of the transmission line connected to the substation. Taking the example above with a 4.5 MV
full lightning surge with time to half-value on the tail equal to the median value for first negative
stroke current of 77.5 µs this results in a discharge energy of approximately 2.4 MJ for an arrester
with protection level of 1620 kV at 20 kA.
3.6.5 High current impulse withstand capability

The conditioning test in the standard with high current impulse of 100 kA 4/10 is more seen as a
check of the insulation strength of the metal-oxide blocks and not as realistic current due to e.g. a
close lightning stroke to the station. Such events must anyhow be avoided by proper shielding.
Therefore, it is no need to introduce even higher current impulses for UHV applications than the
existing 100 kA.
3.7 Radio Interference, Corona Losses

3.7.1 General Aspects
Radio-Interference (RI) and corona Audible-Noise (AN) are decisive factors for the design of the
measures to suppress corona discharges. The practise for corona noise reduction is basically the
same between each country; some differences of the conductor design of the line comes from:
- Target levels of RI and AN
- Location of evaluating point of RI and AN
- Required transmission capacity
Requirements for corona effects are mainly achieved by the configuration and design of multi-
bundle conductors used in EHV and UHV systems. In fact, the corona noise will be reduced with
an increase of the distance among phases, but it will be constrained by a limit for the OH line
compaction degree.
In the following, some aspects concerning the evaluation of the parameters related with the
corona effects (which includes 'radio interference', 'corona losses' and 'audible noise') are
presented.
The radio-interference-RI field values can be assessed by empirical relationships with the main
parameters of the installation, calibrated on the basis of the results of measurements performed on
existing installations. As an example, the following expressions developed by CIGRE-IEEE (and
then adopted by IEC-CISPR) can be used for pre-determining the RI field in the area surrounding
the line:
E = – 30 + 3.5 Emax + 12 r – 33 log D/20 + 5 [1 – 2 (log 10 f)2] + h / 300 (40)
Where,
E = RI field (dB/1µV/m)
Emax = electric field – max. value (kV/cm)
r = radius of the sub-conductor (cm)
179
D = distance from the line conductor (m)
f = frequency of the measurement (MHz)
h = height on the sea level (m)
Table 3.7.1 RI levels calculated at the frequency of 0,5 MHz and at 15 m
from the projection to ground of the lateral conductor bundle
Line features RIV levels (dB/1µV/m)
Umax (kV) kV/cm Conductor n/d/s Dry Light rainy Heavy rainy
765 14.9 (l) 6/31,5/450 43 60 67
16.1 (c)
1050 15.1 (l) 8/31,5/450 45 61 65
16.1 (c)
“l” = lateral phase; “c” = central phase; n = number of sub-conductors of the line bundle
2
d = conductor diameter (mm); s = sub-conductor section (mm )
The generation of RI noise is strongly affected by the atmospheric conditions, the geometric
characteristics and the superficial conditions of the line conductors. Table 3.7.1 gives an example
of the different RI values achievable under different weather conditions for different line
geometries and maximum electric field Emax on the conductor surface.
The definition of the RIV limits to ensure the protection of the radio-TV broadcasts is covered by
CISPR standards. In order to identify the limits the minimum signal level to protect is considered.
The definition of such a level falls under the competence of the national Authorities, depending
on the characteristics of the installations and the environmental conditions of the involved
regions. To this respect, the International Telecommunication Union - ITU has subdivided the
world territory according to three regions (1, 2, 3) and each region under three climatic zones (A,
B, C). For each of these zones the maximum level of noise naturally produced by the
environment has been identified.
Corona losses in HV installations due to corona effect depend mainly on the characteristics of
the conductor and on the electric field at its surface. Corona losses of a transmission line vary
with the atmospheric conditions, ranging from about 0.5 to 1.0 kW/km for clean and dry
conductors and some tens kW/km under heavy rain, snow or ice; under light rain, fog, dew
condition intermediate values result. Considering that the atmospheric conditions triggering high
losses levels generally exist for relatively short time periods, the corona losses of a line are on
average low, one order lower than the ohmic ones. The highest values of the corona losses is
anyhow an aspect to keep under control considering that corona losses in conjunction with the
maximum line load require an additional availability of energy to supply.
Audible noise from corona effect is usually noticeable for electrical installations rated above 700
kV. The line conductors give a higher contribution compared of that of the substations where the
design criteria of bus-bars and electrical components may keep the corona effect under control.
The climatic and environmental conditions of the area crossed by the line may strongly affect the
noise generated.
The highest audible noise levels are produced during heavy rain but in practise such a condition
is not considered the most crucial as the noise resulting from the rain itself may generally prevail
on the corona noise. As a consequence, the “wet” condition, i.e. during light rain, after rain, fog
or dew, is considered the condition likely to produce the most disturbance.
Usually, measurements are performed in broad band employing the filter type “A” and the values
are expressed in dB(A). Recommendations on how to evaluate the corona acoustic noise are
180
given in ISO 1996-1, ISO 1996-2. [77],[78] Maximum tolerability levels of the acoustic noise are
identified depending on the characteristics of the concerned area and the use of it.
Example of the different selected areas and the relevant Leqmax is given in Table 3.7.2. Leq is
increased by a 10 dB (A) when referred to the night time (from 10 p.m. to 7a.m.) in order to
consider the higher physiological sensitivity people have in this time interval.
Table 3.7.2 Maximum acoustic noise value allowed for different areas of concern
Area classification Leqmax (dB(A))
Daytime Night
I - protected 40 50
II - residential 45 55
III - mixed 50 60
IV - human activity 55 65
V - mainly industrial 60 70
VI - exclusively industrial 70 70
3.7.2 Eskom 765 kV transmission line and impact of RIV, CL and AN

on the conductor bundle design
Initial design studies in the early stages of the first Eskom 800 kV project (1979-80) indicated
that because the lines will be located at altitudes greater than 1500 m above sea level, a 6
conductor bundle will have to be considered to limit corona effects, i.e. RIV, AN and corona
losses CL.
Due to the uncertainties about air density correction factors to be applied to the AN, RIV and CL
generated by line bundle conductors at this voltage level, Eskom built a high altitude corona cage
to perform actual measurements relating to the mentioned corona parameters under dry and wet
conditions. [79]
For AN, a design target of 53 dBA (L50 wet) at 800 kV at the servitude boundary and mid-span
was selected. For RIV, experience based on the performance of 400 kV lines indicated that the
design limit for 800 kV should be 72 dB at 0.5 MHz for the L50 wet condition.
Corona losses of a few tens kilowatts per phase were predicted for wet conditions.
The optimization studies for the selection of the conductor bundle were done for a horizontal line
configuration and the following two possible bundles were considered.
6 x 400 mm2 (conductor diameter = 28.81 mm)
4 x 800 mm2 (conductor diameter = 38.1 mm)
The AN & RIV lateral distance profiles presented in Fig. 3.7.3 clearly show the better
performance of the 6 x conductor bundle over the 4 x conductor bundle and confirm by
measurement the validity of the initial design studies.
Corona Losses for centre phase, L50 under wet condition:
4 x Conductor = 70 kW/km/ph
6 x Conductor < 10 kW/km/ph
The final 6 conductor bundle used on both the first 436 km lines energized in 1987 was ZEBRA
ACSR with diameter of 28.6 mm arranged in a bundle with diameter 640 mm.
181
Fig. 3.7.3 AN and RI lateral distance profile for a 765 line
3.7.3 Italy’s 1050 kV experience
3.7.3.1 Radio Interference

In the following, some results achieved on the 1000 kV line of Italy’s Pilot Plant are presented.
dB /1µV/m
Frequency (MHz)
Fig. 3.7.4 Frequency spectrum of the 1000 kV line – Italian Pilot Plant
sub-conductor distance = 31.5 mm, Emax = 17.8 kV/cm [80]
Fig.3.7.4 shows the frequency spectrum measured at 20 m from the external phase. Fig. 3.7.5
shows the statistical distribution of the corona losses experienced under “dry” and rainy”
weather, respectively. Fig. 3.7.6 shows the mean value of the corona losses versus the rain
intensity, for the conductors bundle supplied at 730 kVrms and for two configurations, namely
“symmetric” and “asymmetric” (kasym.=1.5). The better behaviour of the asymmetric
configuration finds justification considering that under rain conditions with it may be convenient
to bring the lower sub-conductors nearer in order to reduce the electric field on their surface
where the water drops more easily locate.
182
Probability (%)
Rainy
Dry
Corona losses (kW/km)

Fig. 3.7.5 Corona losses for a bundle of 8 sub-conductors, d=31.5mm, s=450mm, Emax=17.8kV/m
Fig. 3.7.7 shows the variation of the corona losses value as a function of maximum electric field
on the conductor, under light (< 0.03 mm/min) and heavy (> 1mm/min) rain condition.
Fig. 3.7.8 shows the mean value of the “specific” corona losses, i.e. the losses (kW/km) referred
to the squared value of the radius of the sub-conductors (cm2) as a function of the mean value of
the electric field. The curve of Fig. 3.7.8 was developed by ENEL on the basis of the results of
several experimental investigations on the corona effects worldwide performed. It can be applied
for an estimation of the losses in the limit of the electric field values of practical concern.
To the purpose of providing examples of applicable formula in the following the relationship
developed by ENEL in the occasion of the research on the 1000 kV line for the Italian Pilot Plant
and validated by the comparison with the results of measurement is presented. The losses Po
(kW/km) can be evaluated according to the following formula a) or b) for the heavy rain
conditions (value 95% of the distribution for the rainy time) or light rain (50%), respectively:
log Po = 1.52 – 30 / Emax + 1,37 log n + 2.3 log d a) (41)
log Po = 1.04 – 41 / Emax + 1,13 log n + 3.8 log d b) (42)
183
where Emax = average of the maximum values of the electric field of the sub-conductors (kV/cm),
n = number of sub conductors, and d = diameter of the sub-conductor (cm)
3.7.3.2 Audible noise

In the following information is provided concerning:
The relationship of the corona audible noise versus the maximum electric field and the
geometric characteristics of the bundle under light and heavy rain conditions. (Fig. 3.7.9)
The change of the audible noise amplitude versus rainfall for the conductors bundle in its
symmetric and asymmetric (k=1.5) configuration. (Fig. 3.7.10)
Still, it should be considered that asymmetric conductors bundle or tubular conductors, even if
good solutions in reducing audible corona noise, may present, in the UHV range, practical
problems of realization (mechanical, related to vibrations, thermal).
Analytic formula for the audible noise have been identified on the basis of different experimental
investigations. As an example of calculation shown in the note, the audible noise at 15m from the

Heavy rain
Symmetric
Light rain
Asymmetric
Rainfall (mm/min) Emax (kV/cm)

Fig. 3.7.6 Corona losses vs rainfall intensity Fig. 3.7.7 Corona losses vs max. electric field
Specific Corona losses (kW/km2)
Rainy
Dry
Emean (kVrms/cm)
Fig. 3.7.8 Specific corona losses vs the electric field on the conductor surface
184
projection to ground of the lateral phase of the 1000 kV 8 bundle Dia.31.5 mm/450 mm
conductors bundle submitted to an electric field of 17.76 kV/cm results as follows:
under light rain condition (L50% value of the statistical distribution) = 54 dB(A)
under heavy rain condition (L95% value of the statistical distribution) = 58 dB(A).
µPa)
Audible noise (dB(A)/20µ
µ Pa)
Symmetric
Audible noise (dB(A)/20µ
Heavy rain
Asymmetric
Light rain
Rainfall (mm/min)
Emax (kV/cm)
Fig. 3.7.9 Audible noise versus rainfall Fig.3.7.10 Audio noise rainfall
Note: The empirical expressions for the acoustical noise (A), developed by ENEL during the investigations on the
Italian 1000 kV Pilot Plant. The expressions are valid for lines featured by V >700 kV, n>6, d=2 to 5 and installed at
sea level; an increase of 3dB(A) should be considered each 1000m rise.
Heavy rain: A=23-576/gmax+37 log(d)+10log(n) (43)

Light rain: A=30-863/gmax+57log(d)+7log(n) (44)
where:
gmax = mean value of the maximum electric field of the bundle conductors (kVrms/cm)
n, d = number and diameter (cm) of the bundle conductors, respectively
D = distance of the measurement from the conductor (m)
3.7.4 TEPCO 1100 kV experience
3.7.4.1 Basic concept of corona noise reduction

The measure to reduce corona noise is one of the important design issues for UHV substations,
especially AIS. Corona noise has two subjects. One is radio-interference (RI), which causes radio
disturbance, and the other is corona audible-noise (AN).
In Japan, the regulation for RI in AM radio broadcasting band (526.5kHz–1606.5kHz) is
relatively severe, and the government guideline requires to take some measures in case that the
audience suffer with a SN ratio (Signal to Noise ratio) of less than 20dB(µV/m). Therefore, the
design of transmission lines has been based on the SN ratio of 20dB(µV/m) or above.
On the other hand, the radio noise generated in a substation is not only emitted to its surrounding,
but also propagates through the transmission line and then be emitted from the line. Accordingly,
the countermeasure for corona noise of substations focused on RI, and its suppressed level is
coordinated with that of transmission lines.
185
The design concept of UHV substations will be same as that of 550kV substations, which have a
lot of AIS and have many field experiences of corona noise reduction.
3.7.4.2 550kV substation

The radio-interference electric field intensity right below the 550kV double-circuit transmission
line (4-bundle conductors with ACSR410mm2) in rain condition is around 60 dB (µV/m).
Accordingly, the allowable RI level of substations has been set for 60 dB (µV/m).
With regard to the AN of substations, disturbances have not been caused up to now because
corona discharges are suppressed from the RI requirement, and no special considerations have
been taken.
3.7.4.3 UHV (1100kV) substation

Since UHV substations adopt fill-GIS, the countermeasure
will be limited to the entrance of transmission lines. The RI
level of UHV transmission lines are around 60 dB (µV/m),
thus 60 dB (µV/m) will be adopted as the allowable RI level
of UHV substations same as the 550 kV. The entrance of
UHV Test station (Fig.3.7.11) was designed to satisfy this
level.
With regard to AN, it was observed in the UHV Test Station
as shown Fig.3.7.12, and considered in the conductor design
of transmission lines as followings. However, no special
considerations in substations will be taken because the noise
from transformers caused by magnetic vibration of iron core Fig.3.7.11 Entrance of
is predominant. UHV Test station
Fig.3.7.12 Frequency analysis result of audible-noise measured at the entrance of UHV Test
Station (Applied voltage is 664kV=1100kV/ 3 x 1.046)
3.7.4.3.1 Conductor design of UHV transmission line

For UHV transmission lines, 8-bundle conductors with 810 mm2 ACSR have been adopted
(partially 8-bundle conductors with 610mm2 are adopted in mountainous area).
186
These designs were determined taking into consideration not only RI but also AN, because they
become a predominant factor of corona disturbances when the system voltage increases.
The AN of 550 kV transmission lines which have long-term operating experiences is around 50
dB (A), and problems have not been caused. On the other hand, complaints have been generated
in overseas transmission lines with about 55 dB (A).
Therefore the allowable level for UHV transmission lines was determined to be 50 dB (A) or
below in light rain condition in the vicinity right under the transmission lines.
Although random-noise is predominant in AN, corona-hum-noise (pure sound; with power
frequency component and even-number multiple component of power frequency) was also
reduced as much as possible, because such noises often cause discomfort once it is perceived.
Table 3.7.3 Estimated values of corona noise of UHV and EHV transmission line
550 kV 1100 kV Reference
Circuit Double-circuit Double-circuit Sub-conductor distance; 40 cm,
Conductor 410 mm2x4 810 mm2x8 symmetric configuration
G max 15.5 kVrms/cm 14.2 kVrms/cm At 1.0E (maximum voltage/ 3 )
Gp 14.9 kVrms/cm 14.0 kVrms/cm
RI 61 dB (µV/m) 60 dB (µV/m) Under heavy rain condition
Random- 45 dB (A) Under light rain condition
48 dB (A)
noise [49 dB (A)] (L50 value of the statistical
AN Corona-hum- distribution, rainfall; 0.5 mm
34 dB (A)
- /hour)
noise [38 dB (A)]
Note: G max is the maximum electric field on conductor surface, and G p is that in the bottom of conductor
where water droplets are formed at rain condition and from which corona is mainly generated. These values
show roughly calculated ones from the following approximate equations that assume complete three-phase
transposition.
 r π
V 1 + ⋅ 2(n − 1) ⋅ sin 
G max =
 S n ‥‥(45) S
D
n ⋅ r ⋅ ln 

 re 
 r π 
V 1 + ⋅ 2(n − 1) ⋅ sin ⋅ cos θ 0  G max
Gp =
 S n  ‥‥(46) θo Gp
D
n ⋅ r ⋅ ln 

 re 
where,
V: Applied voltage to earth, n: Number of sub-conductors of the bundle,
S: Sub-conductor distance, D: Phase to phase distance,
r: Radius of sub-conductor, re: Equivalent radius of the bundle
n n −1
 
 
 S 
re = n ⋅ r ⋅  
 2 ⋅ sin  π   ···· (47)
 
 n  
187
Even under the same electric field, as the diameter and number of conductor increase, corona discharge
tends to extend more easily, thus both RI and AN will also increase.
AN values are estimated ones based on actual measurement data obtained from a corona cage in the Central
Research Institute of Electric Power Industry (CRIEPI) in Japan. The values enclosed in parentheses refer to
noise values when spiral wires are wound around those conductors to prevent wind noise. [81]
3.7.4.3.2 Practical design of substation equipment

For bus-bar designs, minimum conductor size and number of conductors are determined to meet
the RI level of 60 dB (µV/m) under rain condition. Regarding substation equipments and
insulators, to suppress corona noise below that of busbar by an appropriate arrangement of
shielding-rings is a fundamental practise. In actual design, corona noise level varies depending on
the structure of connections with conductors. Therefore, the corona tests as shown in Table 3.7.4
are often performed though it is not standardized in Japan. [82]
The equipments of UHV Test Station such as gas bushing, V-shaped suspension insulator were
designed to satisfy the corona test’s requirement and verified at manufacturer’s test site before
shipment.
Table 3.7.4 Typical procedures of corona test for substation equipment
Test item Condition Test voltage Allowable level
There should be no apparent
Visual corona test Dry 1.2E (to earth)
visual corona or audible-noise
RIV (Radio-Interference -
Wet 1.0E (to earth) RIV: 90 dB (µV) or below
Voltage) measurement test
* 1.0E = Maximum voltage/ 3
** RIV is measured by the following circuit.
Line trap
Coupling capacitor
R L =150 ohm
Zs
RL
Noise measuring
apparatus
Testing transformer Equipment for test
Fig.3.7.5 Basic circuit of RIV (Radio-Interference-Voltage) measurement (NEMA Standard [80])
Note
[1] Test voltage
Test voltage of the visual corona test is set to be 1.2 E, considering,
(1) Single phase test is conducted, and as the result, potential gradient on the conductor surface considered
to be less than that of 3-phase arrangement
(2) Actual operating conditions (effect of relative air density, etc.)
The radio-interference voltage measurement test is carried out under wet conditions.
Since 1.2 E is too-strict requirement, 1.0 E is adopted. This level was determined through experiences
considering that it would be appropriate to comprehensively evaluate by both test of visual corona and radio-
interference.
188
[2] Allowable level of RIV (Radio-Interference-Voltage)
The RI level of transmission lines is expressed by radio-interference electric field intensity dB (µV/m). On the
other hand, RIV is measured in dB (µV) for substation equipment usually. The difference between radio-
interference electric field intensity and RIV is approximately 30dB from various test data on transmission
lines. Consequently, the allowable level for RIV can be determined to be 90 dB (µV) by adding 30 dB to 60 dB
(µV/m) that is above-mentioned RI level.
3.7.5 BPA 1200 kV experience

At the end of the seventies (1977-1978) BPA performed long-term measurements in the 10 km
joint right-of-way of the three-phase 8x41 mm 1200 kV prototype line and the single-phase 2x41
mm 500 kV line running parallel, with the aim, among others, of assessing the long-term
electrical and corona performance of UHV transmission systems and comparing the performance
of the two line configurations at the RI, AN and television interference in roughly the same
weather and over the same time period.
Table 3.7.5 gives a synthesis of the major values achieved (with all levels corrected to single
circuit).
For further comparison, estimated values for two other single-circuits BPA-500 kV line
configurations are given. The ozone concentration was measured. No increase was detected that
could be attributed to the corona on the conductors.
Table 3.7.5 Corona effects quantification

System voltage (kV) 1200 500
Operating voltage (kV) 1150 525
Sub-cond. bundle n, Φ (mm) 4 x 41 2 x 41 1 x 63.5 3 x 33.1
Electric fieldI (kV/m) 7.3 7.3 6.7 8.5
RI (dB /1µV/m) II
dry 46 52.5 61 46
rainy 65 70 81 68
III
AN dB(A) 53 56.5 62,5 46
TV interf. (dB /1µV/m) IV 13 7.3 6.7 8.5
I max. value at midspan
II quasi-peak L50 level measured 15m from outside phase, CISPR Meter, 0.5 MHz
III weighted L50 levels measured respectively at 15m from outside phase
IV quasi-peak L50 level measured 40m from outside phase, CISPR Meter, 75 MHz
Additional specific remarks are as follows.
3.7.5.1 Radio Interference

With the line energized at 1200 kV the values were 48 dB and 67 dB under dry and rainy
condition, respectively.
At the operating voltage of 1150 kV and 525 kV the 8x41mm line produced the same RI levels as
the 3x33.1mm 500 kV configurations, and is about 6 dB and 15 dB quieter than the 2x41mm and
1x63.5mm configuration, respectively.
189
3.7.5.2 Audible noise
With the line energized at 1200 kV under rainy weather the levels measured at 15 m and 150 m
from the lateral phase were 55 dB and 47 dB, respectively.
The level at 1100 kV was 4.5 dB lower than that at 1200 kV.
Under rainy weather, the level of the1200 kV line operating at its full voltage was 5 dB quieter
than the twin 2x41mm 500 kV line operating at 540 kV.
The single 1x63.5 mm and 3x33.1 mm 500 kV lines operating at 525 kV were 7.5 dB noisier and
9dB quieter, respectively, than the 1200 kV line operating at its full voltage.
At 1150 kV, the 1200 kV 8x41 mm line was 7 dB noisier, 3.5 dB quieter and 9.5 dB quieter than
the 3x33.1 mm, 2x41mm and 1x63.5mm 500 kV lines, respectively, operating at 525 kV.
3.7.5.3 Electric Field

With the 1200 kV line energized at its full voltage, the maximum electric field measured 1m
above ground at mid-span was 7.6 kV/m for a line clearance of about 24.4 m.
Electric field strengths and associated effects (such as short-circuit currents on vehicles and
fences) were nearly the same as those found under 500 kV lines with equivalent electric fields.
Due to increased clearances required for 1200 kV to keep the same maximum electric field from
500 kV lines, the attenuation rate for the 1200 kV line must be less than 500 kV lines.
3.7.5.4 TV interference
With the line energized at 1200 kV under rainy weather and the antenna at 40 m from the outside
phase the L50 TVI levels at 75 MHz for Peak, Quasi-peak and Field Intensity measurements
were 28 dB, 15 dB and –2 dB, respectively.
Under rainy weather the 1200 kV line was 13 dB, 19 dB and 9 dB quieter than the twin 2x41mm,
the single 1x63.5mm and the single 1x33.3 mm 500 kV line, respectively.
3.7.6 Russian 1150 kV experience

Designing the 750 and 1150 kV lines was preceded by measurements of corona losses and TV-,
radio- and audible noises on experimental short prototype lines and then was followed by field
measurements on erected lines in operation. [71] The maximum gradient on conductor surface in
1150 kV bundle has not to exceed 90% of starting gradient of total corona, thus providing some
margin for possible voltage increase over rated one, lower position of the conductors in the mid-
span and reduced air density on altitudes up to 500 m. Radio noise at 100 m from the line was
prescribed in levels of 48-30 dB for frequencies, respectively, from 0.15 MHz going up to 1000
MHz. Acoustic noise at distance 300 m has not to exceed 55 dB (A) during daytime and 45 dB
(A) at night.
On bundled phases of 750 and especially 1150 kV lines corona losses in bad weather could be
very high. As an effective measure to reduce them a temporary reduction of operating voltage
was recommended. As an adviser to Interconnected pool dispatchers a system of on-line
monitoring corona losses was developed and put in operation on several 750 kV lines.
190
References
[1] J. Takami, S. Okabo, “Observational Results of Lightning Current in Transmission,
Towers”, IEEE-PD, Vol.22, No.1, Jan ’07, pp. 547-556
[2] Gu Dingxie, et al., “Study on Overvoltage and Insulation Coordination for 1000 kV AC
Transmission System”, International Conference of UHV Transmission Technology 2006,
Beijing, ACTECH037
[3] CIGRÉ Technical Brochure 336, December 2007, “Changing Network Conditions and
System Requirements, Part II The impact of long distance transmission”, CIGRE WG
A3.13
[4] A. Mercier, et al., “Transformer Controlled Switching taking into account the Core
Residual Flux”, CIGRÉ SC 13 Session 2002, Report 13-201
[5] A.K. Lohkanin, V.S. Rashkes, et al., “Temporary Overvoltages and their Influence upon
the Insulation Level of Equipment”, CIGRÉ SC 33 Session 1990, Report 33-209
[6] “Electric power transmission at voltages of 1000 kV AC or ± 600 kV Dc and above”,
Électra 122, pp. 41-75,
[7] Lin Jiming, et al., “Limitation of overvoltages in 1000 kV pilot project in China”,
IEC/CIGRE UHV Symposium 2007, Beijing, Report 2-1-3
[8] A.L.J. Janssen, et al., “Changing Network Conditions and System Requirements – Studies
performed by CIGRÉ WG A3.13”, CIGRE SC A3 Colloquium 2007, Rio de Janeiro,
Report PS 2-1
[9] E. Zaima, et al., “System Aspects of 1100 kV AC Transmission Technologies in Japan:
Solutions for Network Problems Specific to UHV AC Transmission System and
Insulation Coordination”, IEC/CIGRE UHV Symposium 2007, Beijing, Report 2-1-2
[10] C. Neumann, C. Weck , “Planning of UHV systems with respect to insulation
coordination based on CIGRE experience”, IEC/CIGRE UHV Symposium 2007, Beijing,
Report 2-1-1
[11] E, Zaima, et al., “System Design and Insulation Coordination for UHV Transmission
System in TEPCO”, International Conference of UHV Transmission Technology 2006,
Beijing, ACTECH006
[12] Du Shuchun, et al., “Study on Insulation Coordination for 1000 kV UHV AC
Transmission System”, International Conference of UHV Transmission Technology 2006,
Beijing, ACTECH007
[13] CIGRÉ Technical Brochure 85 (1994), “UHV Technology”
[14] K.-H. Weck, “Standardization of insulation withstand levels for UHV systems in IEC TC
28, Insulation coordination”, IEC/CIGRE UHV Symposium 2007, Beijing, Report 5-4
[15] Bao-jiu Li, et al., “Study on main design principles for 1000 kV AC substation”,
[16] IEC Standard 60071-1 Insulation Coordination, Part 1: definitions, principles, rules
[17] IEC Standard 60071-2 Insulation Coordination, Part 2: application guide
191
[18] Liangeng Ban, “Estimation of VFTO for GIS and HGIS of China 1100 kV UHV pilot
project and suppressing countermeasures”, IEC/CIGRE UHV Symposium 2007, Beijing,
Report 2-3-4
[19] R. Hemmi, et al., “Evaluation of VFTO and Its Reduction…”, International Conference
of UHV Transmission Technology 2006, Beijing, ACMAN008
[20] Y. Yamagata, et al., “Phase-to-Ground and Phase-to-Phase Sparkover Characteristics of
Insulation at the Entrance of UHV Substation” IEEE Transaction Power Delivery, Vol.17,
No.1 (2001)
[21] Y. Yamagata et al., “Development of 1100 kV gas circuit breakers and their verification
tests”, IEC/CIGRE UHV Symposium 2007, Report 2-4-3
[22] H. Hamada et al., “Severe duties on high-voltage circuit breakers observed in recent
power systems”, CIGRE Session 2002, 13-103
[23] IEC 62271-100 “High-voltage switchgear and controlgear – Part 100: High-voltage
alternating-current circuit-breakers”, Edition 1.2, October 2006
[24] Lin Jiming,et al., “The Transient Characteristics of 1100 kV Circuit Breakers”,
IEC/CIGRE UHV Symposium Beijing July 2007, 2-4-4
[25] CIGRE WG A3.22, “Technical Requirements for Substation Equipment exceeding
800kV”, CIGRE Session 2008, A3-211 (to be published)
[26] IEEE C37-06-200x, Draft: Standard AC High-Voltage Circuit Breakers Rated on a
Symmetrical Current Basis - Preferred Ratings and Related Required Capabilities for
Voltages above 1000 Volts (under discussing)
[27] D. D. Wilson, “Series compensated lines-voltages across the circuit breakers and
terminals caused by switching”, IEEE Transactions on Power Apparatus and Systems,
Vol. PAS-73, pp. 1050-56, May-June 1973.
[28] F. ILICETO et al., “TRVs across Circuit breakers of series compensated line. Status with
present technology and analysis for the Turkish 420 kV grid”, IEEE Transactions on
Power delivery, Vol. 7, No. 2, April 1992
[29] F. ILICETO et al., “TRVs across Circuit breakers of series compensated lines. Analysis,
design and operation experience in the 420 kV Turkish Grid”, CIGRE Session 2002, 13-
109
[30] B.L. Avent et al., “Application of 500 kV circuit breakers on transmission line with MOV
protected series capacitor banks”, CIGRE Session 2002, 13-107
[31] Khodabachian et al., “TRV and the non-zero phenomenon in Hydro-Québec’s projected
series-compensated system” CIGRE Session 1992, 13-303
[32] Holmberg et al., “The Stöde Thyristor Controlled Series Capacitors”, CIGRE Session
1998, 14-105
[33] J. Redlund et al., “A new fast protective device for high voltage series capacitors”, IEEE
No 1, pp.4244-0493, 2006
[34] Bui Van et al., “Control of switching Overvoltages and Transient Recovery Voltages for
Hydro-Québec 735 kV Series compensated Transmission System”, IPST’07, June 4-7,
2007
192
[35] WG 13.01, “Surge impedance of overhead lines with bundle conductors during short-line
faults”, ELECTRA 17, April 1971
[36] T. Shimato et al., “Evaluation of interruption capability of gas circuit breakers on large
time constants of DC component of fault current”, CIGRE session 2002, 13-104
[37] CIGRE Technical brochure No.304, Guide for application of IEC 62271-100 and IEC
62271-1, October 2006
[38] William D. Stevenson, Jr., “Elements of Power System Analysis”, 2nd edition, McGraw-
Hill Book Company, Inc.
[39] CIGRE WG A3.22, IEC/CIGRE UHV Symposium 2007, Beijing, Report 4-1
[40] E. Colombo, et al., “Testing Challenges for UHV equipment”, IEC/CIGRE UHV
Symposium 2007, Beijing, Report 4-1
[41] Lin Jiming, et al., “The Transient Characteristics of 1100 kV Circuit Breakers”,
[42] A.L.J. Janssen et al., “Changing Network Conditions and System Requirements – Studies
performed by CIGRÉ WG A3.13”, CIGRE SC A3 Colloquium 2007, Rio de Janeiro,
Report PS 2-1
[43] CIGRE WG A3.13, “Changing Network Conditions and System Requirements, Part II
The impact of long distance transmission”, CIGRÉ Technical Brochure 336, December
2007
[44] CIGRE Technical Brochure 305 “Guide for application of IEC 62271-100 and IEC
62271-1 – Part 2 : Making and breaking tests”. October 2006.
[45] IEC 62271-110, Edition 1: High-voltage switchgear and controlgear – Part 110: Inductive
Load Switching, 2005
[46] CIGRE WG13.07, “Controlled Switching of HVAC Circuit Breakers: Guide for
Application”, Part 1: ELECTRA No.183, pp.43-73, Part 2: ELECTRA No.185, pp.37-57,
1999
[47] IEC 62271-110, High-voltage alternating current circuit-breakers - Inductive load
switching
[48] E. Zaima et al., “Reignition surges at reactor current interruption in cable-system GIS”,
IEEE Tr .on Power Delivery Vol.5, No.2, pp.947-, 1990
[49] C. V. D. Merwe., “On site switching test conducted in a 765kV GIS”, Six international
symposium on high voltage engineering, New Orleans LA, USA 32-28, 1989
[50] J.A. Bachiller, et al., "The Operation of Shunt Reactors in the Spanish 400 kV Network -
Study of the Suitability of Different Circuit Breakers and Possible Solution to Observed
Problems", CIGRE Session 1994, 23-106
[51] Gu Dingxie, et al., "Study on Overvoltage and Insulation Coordination for 1000 kV AC
Transmission System", International Conference of UHV Power Transmission
Technology, 2006
193
[52] CIGRE WG13.07, “Controlled Switching of HVAC Circuit Breakers: Planning,
Specification and Testing of controlled switching systems”, ELECTRA, No.197, pp.23-
733, 2001
[53] A. Mercier, et al., CIGRE 2002 Session 13-201, “Transformer Controlled Switching
Taking into Account the Residual Flux”
[54] H. Ito, et al., CIGRE 2005 Colloquium A3&B3-209, “Application of Controlled
Switching System for Transformer Energization taking into account a Residual Flux in
Transformer Core”
[55] Fröhlich, et al., “Controlled closing on shunt reactor compensated transmission lines”,
Power Delivery, IEEE Transactions on Volume 12, April 1997, pp.734–740
[56] B. L. Avent, et al., CIGRE 2002 Session 13-107, “Application of 500kV Circuit Breakers
on Transmission Line with MOV Protected Series Capacitor Bank”
[57] T. Kobayashi, et al., "Application of Controlled Switching to 500 kV Shunt Reactor
Current Interruption", IEEE Transactions on Power Delivery, Vol. 18, No. 2, April 2003
[58] Hayward, C. D., "Prolonged inrush currents with parallel transformers affect differential
relaying", AIEE Trans., Vol. 60, pp. 1096-1101, Jan. 1941
[59] Bronzeado, H. S. and Yacamini, R., “Phenomenon of sympathetic interaction between
transformers caused by inrush current”, IEE Proc. Science, Measurements and
Technologies, Vol. 142, No. 4, July 1995
[60] P. Fernandez, et al., CIGRE 2005 Tokyo Colloquium A3&B3-208 “Use of controlled
switching system in power system to mitigate transients. Brazilian experience”
[61] IEC 62271-102: High-voltage switchgear and controlgear - Part 102 Alternating current
disconnectors and earthing switches
[62] IEC 62271-xxx, TR (future publication): Capacitive current switching capability of
disconnectors
[63] D.F. Peelo, R.P.P. Smeets, S. Kuivenhoven and J. Krone, “Capacitive Current
Interruption in Atmospheric Air.” Paper No. 106, CIGRE SC A3 and B3 Joint
Colloquium, Tokyo 2005
[64] L. Stenström, “Calculation of required insulation withstand voltages for UHV systems per
IEC procedure.” IWD A3-07 (WG22) 132
[65] G. Dingxie, Z. Peihong, X. Muhong, W. Sen, D. Min and L. Ying, “Study on Overvoltage
and Insulation Coordination for 1000 kV AC Transmission System.” International
Conference of UHV Power Transmission Technology, 2006
[66] T. Kobayashi, et al, Basic design/specifications of GIS for UHV AC and its verification
test at site, IEC/CIGRE UHV Symposium Beijing 18-21 July 2007, Session 2-3-3
[67] Y. Yamagata, et al, “Field test of 1000kV gas insulated switchgear”, CIGRE Session 2000,
13-209
[68] R. Hemmi, et al, Evaluation of VFTO and its reduction by parallel resistor during
switching operation of disconnecting switch in future UHV gas insulated substation, UHV
seminar, pp. 460-466, China 2006
194
[69] Ban Liangeng, et al, Research on VFTO for 1000 kV UHV GIS of China, UHV seminar,
pp. 250-257, China 2006
[70] JEC 2310-2003: AC Disconnectors

[71] N.N. Belyakov, et. al., “Application of single-phase auto-reclosing in a complex EHV
network containing 1200 kV transmission lines”, CIGRE Session 1990, Paper 34-207
[72] Y. Goda et al., “Insulation Recovery Time after Fault Arc Interruption for Rapid Auto-
reclosing on UHV (1000kV Class) Transmission Lines,” IEEE Transactions on Power
Delivery, Vol. 10, No. 2, April 1995
[73] K. Sunabe and T. Inaba, “Recovery Characteristics after the Interruption of Fault Arcs in
UHV-DC Transmission Lines”, T. IEE Japan, Vol. 110-B, No.2, p.138-146, 1990
[74] IEC 60099-4, Edition 2.1, 2006-07 “Metal-oxide surge arresters without gaps for a.c.
systems”
[75] China 2006, ACTECH006, ”Study on Insulation Coordination in 1000kV UHV AC
Transmission Project”
[76] China 2006, ACTECH007, ”Study on Overvoltage and Insulation Coordination for 1000
kV AC Transmission System”
[77] ISO1996-1, 2003 Acoustics – Description, measurement and assessment of environmental
noise – Part 1: Basic quantities and assessment procedures
[78] ISO 1996-2, 2007 Acoustics – Description, measurement and assessment of
environmental noise – Part 2: Determination of environmental noise levels
[79] AC Britten et al, “Radio interference, Corona losses, Audible noise and Power Frequency
Electric Fields as factors in the design of Escom’s 765 kV lines”, CIGRE Open
Conference on EHV Transmission Systems, October 1987, South Africa, No 3
[80] NEMA, “Method of Measurement of Radio Influence Voltage [RIV] of High-Voltage
Apparatus”
[81] Y.Sawada, et al., “Akagi 1000kV project in CRIEPI – Corona test facilities and results up
to 1983” CIGRE Session 1984, 36-05
[82] M.Isozaki et al., “Special design and arrangements on UHV AC conductors and
subconductors taking reduction of environmental impact into consideration”, IEC/CIGRE
UHV Symposium 2007, Beijing, Report 2-2-4
195
4 Technical Specifications of 800 kV Substation Equipment
4.1 800 kV Power Transformers
The 800 kV transmission systems have been adopted in several countries including Canada, the
United States, Brazil, Venezuela, South Africa, Korea, China and India. Table 4.1.1 lists the main
specifications for 800 kV class power transformers that have been investigated to date through
ongoing surveys.
Table 4.1.1 Major items for 800 kV power transformers (for substation)
Item Specification
Country Canada, USA, Brazil, Venezuela, Korea, China, India, etc.
Utility HQ, AEP, NPA, FURNAS, EDELCA, KEPCO, SGCC, PG, etc.
Type Auto-transformer
Phase Single
Rated Voltage (kV) 735/ 3 , 765/ 3 , etc.
Insulation strength: BIL (kV) 1950, 2050
Rated Power (MVA) 1000/3, 1500/3, 1650/3, 2000/3, etc.
Short circuit impedance (%) 10.6, 14, 15, 18, etc
Cooling OFAF, ODAF, etc
Sound level dB (A) 80, 83, 85, 91, etc
Additional information is given in Table 4.1.2, which lists detailed specifications for a KEPCO
project as an example
Table 4.1.2 Detail specification of power transformers in KEPCO project

Classification 765 kV Transformers
Type Single phase auto-transformer (2 divided tanks)
Rated voltage 765/ 3 / 345/ 3 / 23kV
Single phase 400 (200x2) 520 (260x2) 666.7 (333.35x2)
Rated capacity
Three phases 1,200 1,560 2,000
Cooling method FOA1 FOA2 FOA3
Cooler type
Cooling system
(self-cooling function : no)
Insulation strength (BIL) 2,050 kV
(BSL) 1,500 kV
Short circuit impedance 18 percents
Amplitude of voltage regulating +-7 percents
Voltage regulating type OLTC (VFVV type)
OLTC
Tap position Common winding (neutral point side)
Number of tap 23 (including rated voltage tap)
Oil-Gas bushing
Bushing type (BIL)
(Enclosed type) (2,050 kV)
Connection of equipment GIB
Noise level 85 dB
Preventive diagnosis system Yes
The main points obtained from the survey results are as follows: The transformers range from
1000 to 2000 MVA in bank capacity, which is very large compared with that for 500 kV class
196
transformers. The main type of cooling applied is either OFAF (forced) or ODAF (direct), and is
a major characteristic of large-capacity transformers. Although not indicated in the table, some
transformers are equipped with an on-load tap changer on the neutral-point side of the windings.
So, another feature of very high voltage transformers is on-load tap changers.
4.2 800 kV Voltage Transformers (VT), Current Transformers (CT)
4.2.1 800 kV VTs

Table 4.2.1 lists the technical specifications of 800 kV class voltage transformers adopted in
South Africa, Canada, Russia, Korea and China, obtained from the world survey conducted by
CIGRE WG A3.22. Inductive- and capacitive-type VTs are applied to 800 kV systems. VT
specifications are unique compared to those for other equipment because they include rated
secondary voltage, accuracy, and burden. Secondary voltage ranges from 110/ 3 to 100 V,
similar to that in the 500 kV specifications. Accuracy ranges from 0.2 to 1.2% for measuring
applications and 3 to 6% for protection applications.
Table 4.2.1 Specifications for 800 kV VT

Country South Africa Canada Russia Korea China
(Nominal voltage) (765kV) (735kV) (750kV) (765kV) (765kV)
Classification Conventional ITs Conventional ITs Conventional ITs Conventional ITs Conventional ITs
Capacitive type
In case of large network
Type Inductive type Inductive type Inductive type Inductive type
automatism, inductive
type is required
Standard BS3941 (1975) CSA CAN3-C13.1 Russian standard IEC186 (1996) IEC60044-2
Lightning impulse 2100 kV, 2310 kV 2100kV with GSOA,
2100kV 2250kV 2100kV
voltage for chopped wave 1950kV with MOV
Switching impulse 1550kV with GSOA,
NA 1550 kV 1425kV 1550kV
voltage 1425kV with MOV
Power-frequency 950kV with GSOA,
960kV 830 kV 975kV 975kV
Rated primary voltage 765kV / 3 765kV / 3 750kV / 3 765kV / 3 765kV / 3
Rated secondary
110kV / 3 120kV / 3 100V 115-115kV / 3 100kV / 3 , 100V
voltage
Rated burden 50/50VA 200 VA 300VA/200VA/1000VA 200/200VA 50/50/50VA
0.6 WXY 1.2Z (for
measuring), 3PZ (for
Accuracy class 0.2/0.2 0.5% / 0.2% / 3% 1.0/3P 0.2/0.5&3P/3P
protection purpose 5 to
120%), 6PZ for U >150%
Temperature -50 to 40 degree Celsius
NA NA NA NA
characteristics ambient temperature
Voltage Factor 1.2/Continuous - - 1.2/Continuous, 1.5/30S 1.5/30S
Service Experience
23 years 40 years - 5 years Manufactured in 2007
(including on-site experience)
4.2.2 800 kV CTs

Table 4.2.2 lists the technical specifications of 800 kV class current transformers in South Africa,
Canada, Russia and Korea. All CTs are the iron core type. CT specifications are unique compared
197
to those for other equipment because they include rated secondary current, accuracy, burden,
excitation characteristics and transient characteristics. The rated secondary current is generally
specified as 1 A. Accuracy ranges from 0.2 to 1.2% for measuring applications and 1% for
protection applications. CTs used for protection applications are classified into two types with
specified excitation characteristics or specified transient characteristics. The rated short-circuit
current is 40 to 50 kA.
Table 4.2.2 Specifications for 800 kV CT

Country South Africa Canada Russia Korea
(Nominal voltage) (765 kV) (765 kV) (750kV) (765 kV)
Classification Conventional ITs Conventional ITs Conventional ITs Conventional ITs
Iron core type Iron core (hair pin type) -
Type Iron core type Iron core (ring) type
(Except for Line Protection) oil type
Standard BS3938 (1973) CSA CAN13-C13 Russian standard IEC185 (1996), JEC 1201
2100 kV, 2310 kV for 2100kV with GSOA,
Lightning impulse voltage NA NA (2250kV)
chopped wave 1950kV with MOV
NA 1550 kV NA (1425kV)
voltage 1425kV (with MOV)
950kV with GSOA,
Power-frequency voltage NA 830 kV NA (830kV)
830kV with MOV
(1) 4800/3200A,
(1) 2000/ (2) 4000:1A, (1) 8000/6000/4000/3000/
Rated primary current (2) 2400A, 1000-4000A
(4 secondary windings) (2) 2000A
(3) 4800/3200/1600A
Rated secondary current 1A 1/1/1/1/5A 1A 1A
(1) 15 VA for protection, (1) 25VA for protection,
Rated burden (1) NA, (2) NA, (3) 15VA -
(2) 30 VA for measuring (2) 30VA for measuring
(1) class X, (2) class X, (1) 10 TPY for 1 A for protection (1) 1% (4 relay outputs), (1) 1T for protection,
Accuracy class
(3) 0.2 (at 1600A tap) (2) 5 TPY for 5A for measuring (2) 0.5% (one measuring outputs) (2) 1.2 (at 2000A tap) for measuring
Secondary winding (1) 9.6ohm (at 4800A tap) (1) 32.8ohm (at 8000A tap)
- -
resistance (2) 5.0ohm, (3) NA (2) NA
(1) 4800V (at 4800A tap) (1) 9050V (at 8000A tap)
Knee-point e.m.f - -
(2) 550V, (3) NA (2) NA
Maximum exciting current (1) 0.042A (at 4800A tap) (1) 0.08A (at 8000A tap)
- -
at knee-point e.m.f (2) 0.05A, (3) NA (2) NA
Continuous secondary
(1) 2A, (2) 2.1A, (3) 2A 1 A and 5A for protection - (1) 2A, (2) 2A
current
Rated primary short-
50kA 40kA - 50kA
circuit current
Primary time constant
NA 150 ms - 180ms
Permissible time to accuracy limit
Service Experience
23 years 40 years - 5 years
4.3.1 Dielectric requirements

A comparison of dielectric requirements for the circuit breakers of existing 800 kV systems
shows that the basic values are well in line with IEC 60694, and correspond to revised version
IEC 62271-1 to be published soon. The requirements are listed in Table 4.3.1.
198
Tab.4.3.1 Specifications of dielectric requirements for 800 kV circuit breakers in comparison with IEC
Canada Brazil S. Africa USA
IEC 60694 Russia China
HQ Furnas ESKOM AEP
Draft 62271-1 750 kV 800 kV
765 kV 765 kV 800 kV 800 kV
Rated Voltage kV 800 800 765 800 750 800 800
GSOA /
GIS / AIS
- MOV
Lightning impulse withstand voltage
Phase-to-earth
and between kV 2100 2100 2100 2100 / 2100 2100 / 1800 2100 2050
phases
across open
kV 2100 + 455 2100 + 455 2100 + 650 2100 + 455 2255
switching device
Switching impulse withstand voltage
Phase-to-earth
1425 1425 /
and across open kV 1425 1550 1550 / 1425 1550 / 1425 1425
(Draft 1550) 1600 dry
switching device
Between phases kV 2420 2420
across open 1550 /
kV 1100 + 650 1175 + 650 1300 + 650 1100 + 650
switching device 1870 dry
Power-frequency withstand voltage
Phase-to-Earth kV 830 830 870 960 950 / 830 830 960
across open
kV 1150 1300 1400 1150 960
switching device
Additional remarks
325 ohm or 300 ohm of
400 to 600
Closing resistors --- line surge 450 ohm two duty
ohm
arresters cycle rating
In particular, the LIWV (BIL: basic insulation level) of 2100 kV is nearly the same for all
specifications. The AEP requirement is slightly less at 2050 kV due to the application of
ANSI/IEEE standards that generally uses different values. Russia used a reduced LIWV of 1800
kV when using metal oxide arresters (MOV), which may have been overly optimistic when it was
specified such a long time ago. The GSOA (gap silicium oxide arresters) specification is in line
with IEC and the other utilities.
The 1425 kV switching impulse voltage to earth in the current standard is not followed by all of
the utilities. Some require the next higher level of 1550 kV. The need for an increased level has
been taken into account in the revised standard
The withstand capability across open circuit breakers must be tested using a combined voltage
test, impulse voltage against power frequency voltage, according to IEC test procedures.
ANSI/IEEE specifications did not use the combined voltage test and therefore the voltage
requirements of AEP cannot be directly compared with the others.
Concerning the lightning impulse test, the power frequency fraction of the test voltage according
to IEC is only 70% of the peak line-to-earth value (450 kV). The background to this is the low
occurrence probability of the highest impulse value together with the short time period of power-
frequency peak value with opposite polarity. The exception is Brazil, where a 100% peak value
of 650 kV is used, presumably because of the extremely high rate of lightning flashes.
For the combined voltage test with switching impulse, all the utilities follow the IEC procedure
for using the full power-frequency peak value against a reduced switching impulse voltage. The
different switching impulse values are in line with the line-to-ground requirements. An
explanation for the reduced impulse level in this test is given in CIGRE Guide TB 304.
The requirements for the power-frequency withstand voltage to earth differ widely. The
minimum is 830 kV (IEC value, corresponding to 1.8 p.u.) and the maximum is 960 kV (nearly
199
2.1 p.u.). Since the high values cannot be related to service circumstances, they may be proof of a
special high withstand capability as a quality control measure. The power frequency requirements
across open circuit breakers are also extremely high in some cases. Hydro-Québec stipulates a
value of 1300 kV (exceeding the IEC value) to be consistent with the 1829 kV (1293 kV rms
value) specified for line charging current interruption. The reason for Russia’s use of 1400 kV is
not known.
IEC recommends 2.5 p.u., with the maximum requirement exceeding 3.0 p.u. for 1400 kV. The
ANSI/IEEE recommends 960 kV (2 p.u.), the same withstand voltage level as to ground.
For the limitation of switching overvoltage, closing resistors were commonly used on old-type
circuit breakers. Today, many utilities are using MOAs to reduce switching overvoltage. The
specified resistance varies between 300 and 600 ohm. Opening resistors are specified only by
Furnas for application to circuit breakers to switch shunt reactors (1000–4000 ohm).
4.3.2 Making, breaking and switching test requirements

Tables 4.3.2 and 3 show a comparison of circuit-breaker specifications related to 800 kV for
terminal fault test duties for short-line faults, out-of-phase and capacitive current switching tests.
Hydro-Québec stipulates a special class of circuit breakers for series-compensated lines. The
requirements are also defined in the two tables.
4.3.2.1 TRV for terminal faults

The majority of specifications closely follow the requirements of IEC 62271-100 [1],[2] with the
exception of those from Furnas (Brazil) and AEP (United States), which stipulate higher values
for the first-pole-to-clear factor (kpp).
A first-pole-to-clear factor of 1.3 is specified in IEC 62271-100 for terminal fault test duty T100
as it covers the values that can be obtained in the case of three-phase to ground faults in 800 kV
networks that have effectively earthed systems (see the chapter on TRV for terminal faults).
The same kpp value is required for T60 and T30, but Amendment 3 to the IEC standards will
require a factor of 1.5 for T10 in order to cover TRV obtained during three-phase long-line faults.
The corresponding peak value of TRV at T10 will be 1499 kV, based on an amplitude factor of
1.7×0.9 with kpp = 1.5 or 1.76 with kpp = 1.3, and will be close to the value currently specified by
Furnas. The amendment prepared by IEC is in accordance with the present recommendations of
CIGRE WG A3-19, which has studied the implications of three-phase line faults and determined
that TRV in this current range would be covered if an amplitude factor of 1.76 is adopted.
The requirements for Furnas’ 800 kV networks were defined in the early 1980s, including a first-
pole-to clear factor of 1.5 for terminal faults. At that time, Brazil’s network consisted of several
weakly connected subnetworks. Although each subnetwork could be considered as solidly
earthed, the network as a whole could not. Furthermore, the 800 kV network would not be a
meshed network but rather a radial one, consisting of 3 parallel circuits, each 895 km long, and
4 substations. Thus, the planning engineers at that time, in consideration of such an unfamiliar
high voltage level in Brazil, decided to adopt conservative requirements derived from system
simulations under severe operation and emergency conditions.
AEP’s specification is based on ANSI/IEEE C37-09-1979 [3],[4],[5], which has a first-pole-to-
clear factor of 1.5 in order to also cover three-phase ungrounded faults. It should be noted that
when this standard was revised in 1999, kpp of 1.3 was specified for this rated voltage, as in IEC
200
62271-100. The present revision prepared by IEEE proposes a choice between 1.5 and 1.3,
intended to cover or not cover, respectively, the case of three-phase ungrounded faults. In the
case of T10, the same TRV peak value introduced by IEC in Amendment 3 to 62271-100 would
also appear in the revised IEEE standard, with an alternative value retained at the draft stage with
a higher peak of 1610 kV (corresponding to an amplitude factor of 1.64 with kpp = 1.5 or 1.89
with kpp =1.3) that is close to the 1649 kV specified by AEP corresponding to an amplitude factor
of 1.68 with kpp = 1.5 (the possibility of having three-phase ungrounded faults must be assumed
to have this TRV peak).
In spite of the existing differences in TRV peak at T10, there is agreement between all
specifications on the rate of rise of recovery voltage, which is approximately equal to 7 kV/µs.
4.3.2.2 DC time constant

The standard time constant of 45 ms for the DC component is specified in China, while higher
values ranging from 67 to 88 ms are specified for other networks. These last values are
comparable to the 75 ms given in IEC 62271-100 for special cases and rated voltages equal or
higher than 550 kV.
4.3.2.3 Single-phase fault tests

Only Eskom specifies IEC 62271-100 requirements for single-phase faults. Double-earth fault
tests are not applicable in effectively earthed systems.
4.3.2.4 Short-line-fault
The values in all specifications are covered by the requirements in IEC 62271-100.
4.3.2.5 Out-of-phase
The specifications require an out-of-phase factor of 2.0 and a TRV peak as stipulated in IEC,
with the exception of Brazil, where a significant TRV peak is required, considering that the
voltage can be phase-separated by 180°.
For series-compensated lines in the case of system separation (extreme contingencies), Hydro-
Québec specifies that a circuit breaker that may be exposed to severe out-of-phase stress: (3.3 ×
1.0 × peak system voltage) = 2000 kV, interrupted current = 5 kA, industrial frequency recovery
voltage = 1400 kV.
4.3.2.6 Capacitive current switching

A voltage factor of 1.4 covers the specified values given in Table 4.3.2, with the exception of a
factor of 1.5 specified in Russia. The same factor was previously stipulated in ANSI/IEEE, but
has since been lowered to 1.4 [6] taking into account the actual synchronism of modern circuit
breakers. A line-charging current of 900 A is generally adopted in the specifications, with the
exception of Furnas and AEP specifying 1500 and 1800 A, respectively.
Hydro-Québec specifies the restrike probability class: C2 for a maximum line length of 310 km
and overvoltage on non-faulted phases of 1869 kV (2.8 p.u.). In the case of system separation
(extreme contingencies), Hydro-Québec stipulates a recovery voltage of 2200 kV and a restrike
probability corresponding to C1 class for this special duty.
201
Table 4.3.2 Specifications for terminal fault of 800 kV circuit breakers compared with IEC
Canada HQ South
Brazil USA
IEC 765 kV Africa Russia China
Furnas AEP 800
62271-100 special class for ESKOM 750 kV 800 kV
series-compensated 765 kV kV
lines 800 kV
Transient Recovery Voltage of Terminal fault
1.4 at T100
First-pole-to-clear-factor: kpp p.u. 1,3 1,3 1,5 1,3 1.5 at T30
1,3
T10
Amplitude factor: kaf p.u. 1,53 1,53 IEC 1,53
TRV peak value: uc kV 1299 1299 1464 IEC 1299 1649
Time: t2 or t3 µs 186 186 t3 = 280 IEC 186 t3 = 230
Rate-of-rise: u1/t1 or uc/t3 kV/µs 7 7 IEC 7 7,17
T30
TRV peak value: uc kV 1308 1308 IEC 1299 1589
Time: t2 or t3 µs 262 IEC 260 t3 = 310
Rate-of-rise: u1/t1 or uc/t3 kV/µs 5 5 IEC 5
T60
First reference voltage: u1 kV 637 637 983 IEC 637
Time: t1 µs 212 212 470 IEC 212
TRV peak value: uc kV 1274 1274 1464 IEC 1274 1508
Time: t2 or t3 µs 1272 1272 t2 = 2800 IEC 1272 t2 = 735
Rate-of-rise: u1/t1 or uc/t3 kV/µs 3 3 IEC 3
T100
Amplitude factor: kaf p.u. 1,4 1,4/(1,81) IEC 1,4
First reference voltage: u1 kV 637 637 983 IEC 637
Time: t1 µs 318 318/(640) 940 IEC 318
TRV peak value: uc kV 1189 1189/(1920) 1380 IEC 1221 1189 1408
Time: t2 or t3 µs 1272 1272(960) t2 = 2800 IEC 1272 t2 = 1470
Rate-of-rise: u1/t1 or uc/t3 kV/µs 2 2 IEC 2,8 2
DC time constant ms 45 75 88 67 45
Special time constant ms 75
Auto-reclose time s 0,3 0,3 0,3 0,3 0,3 0,3
GIS: 2,5
Making current peak 50Hz / (100kA )/
p.u. 2,5 / 2,6 2,7 2,75 2,5
60Hz AIS: 2,5
(125 kA)
Making current peak at 2,7 at 120
p.u. 2,7 2,7
special time constant ms
bundle of 4 x 686 Only three Single and
Additional remarks mm2 conductors phase auto- three phase
Note (1) reclosing auro-reclosing
Note : Special breaking test sequences are requested by HQ for series-compensated line CB.
Terminal fault with current of 15 kA (37% of rated capability of CB), peak TRV is 2040 kV and RRRV is 2 kV/µs.
Test with delayed zero current crossing, peak TRV is 1800 kV with RRRV of 2 kV/µs. Interrupted current ≈ 1 500 A.
202
Table 4.3.3 Specifications for short-line fault, out-of-phase and capacitive current switching tests of 800 kV
circuit breakers compared with IEC
Canada HQ 765 kV Brazil South Africa USA
IEC Russia China
Special class for series- Furnas ESKOM AEP
62271-100 compensated lines 750 kV 800 kV
765 kV 800 kV 800 kV
Short-line fault tests
IEC
Z Ω 450 450 450 450
AF 1,6 1,4 1,6 1,6

t dl µs 0,5 0,5 0,5
Out-of-phase making and breaking tests
Voltage factor p.u. 2,0 2,0 /(2,45) IEC 2,0 2,0
U1 kV 980 1307 980
t1 µs 636 1880 636
Uc kV 1633 1633 /(2000) 2144 1992 1633
AF p.u. 1,25 1,25 2,0 1,25
t2 µs 1272/ (1559) 11000 1,54
u1/t1 kV/µs 1,54 1,54 2,25
Breaking current x Irated 0,25 0,25/(5 kA) 0,25 0,25
Capacitive current switching tests
1.2 GIS: 1,2 1,2 or
Voltage factor p.u. 1,4 1,4 1,5 1,4
1.4 / 1,7 AIS: 1,4 1,4
540 -
Capacitive current A 900 900 1500 900 900 1800
990
Recovery Voltage kV 1829 /(2200)
Restrike probability class C1 or C2 C2/ (C1)
Frequency Hz 66
Line length km max 310 300-550
4.3.3 Requirements for Inductive Load switching

Hydro-Québec stipulates special tests to verify the suitability of CBs for controlled switching
applications. The test results are also used to set controller parameters, including minimum arcing
time to avoid restrike, statistical scatter, operating time variations with ambient temperature,
control voltage, idle time, mechanical wear, etc.) [7]
Table 4.3.4 Specifications concerning inductive load switching in comparison with IEC
South Africa 800
IEC 62271-110 Canada HQ 765 kV
kV
TRV peak voltage kV 1241 1241 IEC

Time to peak (t3) for
Load Circuit 1/ Load µs 296/525 208/294
Circuit 2
Rated pressure except for test duty
Test conditions Rated and minimum pressure
No 4
Mechanical endurance
M1 or M2 M2
class
Controlled switching Yes (opening and closing)
4.4 DS, ES
The WG A3.22 cannot collect sufficient technical specifications on the 800 kV DS and ES.
However, there are no specific subjects for 800 kV DS and ES to be discussed within the WG.
203
4.5 High-Speed Grounding Switch (HSGS)
4.5.1 General
Korea’s 765 kV transmission system delivers large amounts of electric power from generation
facilities on the east and west coasts to the Kyungin area. Normal power is estimated at several
million kW. The system employs double-circuit transmission towers in order to reduce land use
and adopts a high-speed multi-phase reclosing system to prevent a route failure from causing
further outage.
However, in power systems exceeding 500 kV, when an arc earthing fault occurs the faulty phase
is separated by the CB, and the secondary arc induced by electrostatic and electromagnetic
induction from other phases or the other circuit (because of double circuits) does not disappear in
a short time, high-speed reclosing (within 1 second) cannot always be achieved. In this case, it is
necessary to ensure quick reclosing after extinguishing the secondary arc by means of installing a
proper suppression device to avoid the risk of system separation accident due to decreased
stability. Especially in 2-circuit transmission lines, it is desirable to install a high-speed
grounding switch (HSGS), since conditions such as those related to reactance capacity between
each healthy phase and healthy circuit become very complex and it is not only difficult but also
costly to apply the square shunt reactor.
0 1 6 9 15 21 34 40 60[cycle]
0 0.01667 0.10 0.15 0.25 0.35 0.567 0.667 1.0[s]
Both Preceding Following Secondary Preceding Following Analysis

Analysis Fault
starts occurrence CBs
trip
HSGS
close
HSGS
close
arc HSGS HSGS ends
extinction open open
▲ ▲ ▲ ▲
Fig. 4.5.1 Schematic sequence of HSGS operation after a fault occurrence
4.5.2 Design considerations of HSGS

A. Review of probability of multiple lightning occurrences
The LPATS calculated value of probability of multiple failures on a 765 kV transmission line is
one (1) time/147 years based on the measurement status during a 2-year period. The value is one
(1) time/91 years (per 100 km) in the case of applying a correction coefficient using measurement
data from Japan.
204
B. Review of forest fires
According to data from the Forestry Office, the height of flames at the occurrence of a forest fire
in Korea is 20–30 m. The height of 765 kV transmission lines from the ground is as shown in
Table 4.5.1.
Table 4.5.1 Height of 765 kV transmission line from ground
Classification Normal (m) Lowest (m) Type of steel tower
Upper arm 63 - 68 60 - 65 Suspension, strain
Middle arm 46.5 - 49 43.5 - 46 "
Lower arm 30 27 "
It is therefore estimated that a fault (following fault) at the occurrence of a forest fire may arise at
2 phases of the lowest arm (each phase at the lowest arm of 2 circuits) of both sides of the tower.
Accordingly, in 765 kV transmission lines, a 2-phase ground fault of the lowest arm has the
highest probability of occurrence during a forest fire and requires a review
Table 4.5.2 Specifications of HSGS duties
Specification
Breaking
Utility Consideration Rated voltage Transient Remark
current by
recovery
[kV] electromagnetic
voltage
induction
BPA Not considered 550 700 78 kV rms Under operation
TEPCO Considered 1,100 7,000 900 kV peak Under verification
700 kV peak
KEPCO Considered 800 8,000 Under operation
1.3 kV/µs (rrrV)
4.6 Surge Arresters

The survey on 800 kV equipment shows that arresters are specified for a rated voltage of 588 kV
with only a few exceptions. With a normal power frequency overvoltage capability for metal
oxide arresters of at least 1.1 times the rated voltage for 1 s, this further implies that the arresters
are selected to withstand a temporary overvoltage of at least 1.4 p.u.
Furthermore, most of the utilities appear to have adopted the IEC philosophy for specifying the
continuous operating voltage, Uc, according to the actual service voltage and not as a factor of
the rated voltage. However, there are some exceptions such as a required Uc of 490 kV, which
would indicate a system voltage as high as 849 kV. A nominal discharge current of 20 kA and
line discharge class (LDC) 5 as per IEC [8] are also common requirements.
With the protection levels generally required for the 800 kV arresters, LDC 5 results in an energy
requirement of approximately 5-7.5 kJ per kV rated voltage per single impulse. As stipulated in
the IEC standards, two impulses must also be thermally withstood, which thus gives 10-15 kJ/kV
as rated voltage. However, several utilities specify even higher energy requirements of up to 13
kJ per kV rated voltage for single impulse capability and thus 26 kJ per kV thermally. LDC 5
requirements are normally met by arresters in a single block column. However, such a high
requirement as 26 kJ/kV rated voltage thermally requires a design with multiple columns of metal
oxide blocks.
205
The required protection levels given at the nominal lightning impulse current of 20 kA with
waveform 8/20 µs and at switching impulse current of 2 kA with waveform 30/60 µs vary
somewhat between different utilities, as shown in Figs. 4.6.1 and 4.6.2.
1600 1350
1300
1550
1250
1500
LIPL at 20kA - kV
SIPL at 2kA - kV
1200
1450
1150
1400
1100
1350
1050
1300 1000
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Utility Utility
Fig. 4.6.1 Range of required lightning surge Fig. 4.6.2 Range of required switching surge
protection levels (LIPL) at 20kA for 800kV protection levels (SIPL) at 2kA for 800kV
arrester arrester
The standard insulation level for 800 kV equipment is 1675-2100 kV for lightning and 1300-
1550 kV for switching impulses. The survey shows that the requirements for external insulation
strength of the arrester are at the upper range of these insulation levels. In addition, the arrester
standard stipulates that the external insulation shall have an LIWV of at least 1.3 times the
residual voltage at a nominal discharge current of 20 kA and an SIWV of at least 1.25 times the
residual voltage at 2 kA switching impulse. These factors include an altitude correction factor of
1.13, valid for 1000 m above sea level. As shown in Figs. 4.6.3 and 4.6.4, the requirement for
lightning is fulfilled in all cases, but for switching surges, a factor as low as 1.1 is used. This may
be due to the arresters, in the particular case, being installed at low altitudes and therefore an
atmospheric correction factor was not considered necessary.
The insulation level for most 800 kV equipment is LIWV of 1950 or 2100 kV and SIWV of 1425
or 1550 kV. Taking the lowest protection levels for any of the 800 kV arresters, this gives a
protection ratio of 1.38-1.49 for lightning and 1.25-1.36 for switching surges.
Pollution performance of metal oxide surge arresters does not appear to be of great concern for
most utilities. Only the creepage distance is specified, and pollution tests are generally not
required.
Service experience appears to be excellent. Failure of an 800 kV arrester is a very rare case.
206
1.80 1.35
1.60 1.30
1.40
1.25
1.20
1.20
LIWL/LIPL
SIWL/SIPL
1.00
1.15
0.80
1.10
0.60
1.05
0.40
0.20 1.00
0.00 0.95
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Utility Utility
Fig. 4.6.3 Range of ratios external insulation Fig. 4.6.4 Range of ratios external insulation
strength for lightning surge (LIWV) to strength for switching surge (SIWV) to
protection level (LIPL) for 800kV arrester protection level (SIPL) for 800kV arrester
4.7 Shunt Reactors
4.7.1 General
Table 4.7.1 lists the main technical specifications for shunt reactors (SRs) presently in use by
utilities on EHV power systems. Members of CIGRE Working Group A3.22 collected the data
contained in this table.
The primary reason for applying shunt reactors to EHV power systems is the need to absorb line
charging MVar and consequently reduce the Ferranti effect. Different approaches are used by
utilities in terms of reactive power compensation. Some utilities use permanently connected SRs
to compensate for reactive power supplied by the line and the switchable SRs directly at the
substation busbars for voltage control. This is the case in India’s power grid. Most utilities that
have contributed to the survey use shunt reactors at both ends of the line, except for BC Hydro,
which may use a shunt reactor at one end only for shorter lines. Three-phase reactor banks range
from 133 to 330 MVar for power systems of 550 to 800 kV.
SRs are used in transmission OH lines for two main purposes: (a) switchable SR at busbars or
tertiary windings of power transformers for steady-state voltage control issues; and (b) fixed SR
at OH line ends to deal with switching transient overvoltage (TOV) requirements, related to line
closing and/or opening operation (including fast auto-reclosing issues).
Utilities in Russia also use SRs to reduce switching and lightning surges. To ensure that the SR is
always connected when the line is switched, utilities in Russia use a spark gap in parallel with the
breaking chambers. In addition to the spark gap, there is also a system-wide fast switching
automatism to switch on the SRs. Hydro-Québec also uses a system-wide automatic scheme to
switch the shunt reactors to control the voltage in the case of significant system disturbances.
Most utilities that responded to the survey are using a single-phase reclosing scheme (SPR),
except for Hydro-Québec, which is using three-phase reclosing only on its 735 kV lines. For
utilities using SPR, a neutral reactor at the shunt reactor bank is always used to extinguish the
secondary arc within an acceptable delay. In cases where the lines are not transposed, a special 4-
legged reactor bank is used by AEP (USA), as described in Section 3.5.2, to provide secondary
arc suppression. For utilities applying SPR, there is of course a need to properly isolate the
neutral of the reactor bank. The insulation levels are given in Table 4.7.1.
207
Table 4.7.1 Main characteristics of UHV shunt reactors
Country/utility Canada/ Russia Canada/BC- USA/AEP Brazil/Furnas South Africa India/Power
Hydro-Quebec Hydro /ESKOM Grid
Nominal/Maximum 735/765 787 550 765/811 kV 765kV/800kV 765 800
operating voltage(kV)
Capacity of 3-phase Type 1 : 155 330 135 Type 1 : 150 R1:330MVAr/800kV 133 240
group, (Mvar) Type 2: 330 Type 2: 300 R2:150MVAr/800kV
R3:180MVAr/69kV
connectedto 800kVTr
tertiarywinding
Positioned on line ends At both ends Yes One or both ends None, or one at each line One or both ends dependent Both ends At both ends
dependent on line terminals, dependent on online length
length line length (please refer to the brazilian
projects inChapter2)
Switchable All switchable with Yes Some fixed Some are fixed (switched BusReactorswitchable. Line Yes Bus Reactor
CB equipped with (switched with the with the line) and some reactorsarepermanently switchable. Line
controlled switching line) and remainder are switchable. connected.Tertiarywinding reactors are
device (CSS) switchable with CB reactorsareswitchable. permanently
equipped with CSS connected
Automation for fast Yes Yes (spark gaps plus No No No No
switching automation
Application of 4-legged No Yes Yes Yes, on some No Three single phase
reactor lines. reactors, NGR
provided on neutral
Permissible TOVs, L-G, 1.33 p.u. for 10s 1.30 p.u for 20 s., 1.25 p.u. for 10s Line trips for over 1.06 1.30p.u. inlessthan10sec 1,25 p.u for 60 s ?
p.u. 1.88 p.u for 1 s- pu/ 5 minutes (forpowerfrequencyTOV) 1,75 p.u. for 1 s
Level of switching 1.8 p.u. 2.1 (1.8)* 2p.u.onolder linesand 2.0 pu 2.0pu 1550 (reactorinsulation
surges on line end, L-G, 1.7p.uonlinesbuilt referredto 765 level)
p.u. after1990. kV*sqrt(2)/sqrt(3)basis
Power frequency 1 min Specifiedonlyfor 900 (750)* Specifiedonlyfor 1.6 pu 920kV

withstand test voltage, L- bushings:880kV(2 bushings:620kV (Specifiedonlyfor bushings)
G, for line side, (kV rms) p.u.)
Same for the neutral 34 kV 85 185 kV 550 kV LIWV Notavailableat the moment 550 kV LIWV
(Iwilltryto find some
informationonthis issue)
Power frequency 1-hr Specifiedonlyfor 635 Actualtest sequence: 690kV1-hr Notavailable(Iwilltryto find
withstand test voltage, L- windings:750kV, 1 540kV(1.7p.u.)for 800kVEnhanced some informationonthis
G, for line side, kV hour. (1,7p.u.) 7200cycles followed issue)
byone(1)hourat 476 (AppliedVoltage: ANSI
kV(1.5p.u.)allwith C57.21)
partialmeasurements.
Switching surge 1550 kV 1675 kV 1175 kV 1700 kV SIWV: 1550kV 1425 kV
withstand test voltage, L- (1425)* AC voltage:870kV(1.88E)
G, kVp Combinedtest forswitching
implusepowerfrequency:
1300+650kV
Lightning impulse 1950 kV 2250 kV 1550 kV 2050 kV LIWV: 2100kV 1950 kV
withstand test voltage for (1950)* Combinedtest for lightning
internal insulation, L-G, impulsepluspower
frequency:2100+455kV
on full wave
Same on chopped wave 2145 kV 2400 kV Not required 2255 kV Notavailableat the moment
(kV) (2100)* (Iwilltryto find some
informationonthis issue)
*
Note: Values not within parentheses are for Russian equipment developed prior to 1985, protected by zinc oxide gapped
arresters and actually installed on 1150 kV transmission lines; Values within parentheses are for equipment newly developed and
protected with MOVs.
4.7.2 Additional information related to Russia’s 787-1200 kV shunt

reactors
The common practise in Russia is to place shunt reactors on 787 kV line ends (usually on both
ends). All reactors are switchable. Their circuit breakers are equipped with specially designed air
gaps in parallel to arc quenching chambers, which ensures extremely fast (of the order of
microseconds) connection of the reactor to the line upon detection of switching or lightning
overvoltage, if the reactor was switched off in the previous operational mode. Flashover voltages
of these gaps are coordinated with the insulation level of the equipment protected. In addition,
208
there is an automation that commands switching-on of all three poles of the shunt reactor circuit
breaker:
• Immediately, if reactor current is detected in even one of the poles;
• Immediately, if a command is issued to open the line circuit breakers, be it one-pole or three-
pole line operation;
• If a relay overvoltage protection device detects the appearance of TOV, with a time delay
corresponding to the value of increased voltage.
Shunt reactors on the line ends, air gaps on the shunt reactor circuit breakers and the
abovementioned automatics and relay protection are mandatory elements of overvoltage
protection in Russia’s 787 kV transmission system. It should be noted here that the air gaps
provide line connection not only for the shunt reactor, but also for the shunt reactor arresters, thus
reducing not only TOVs, but also lightning and switching overvoltages. The system was such a
success for 787 kV transmission that it was transferred to 1200 kV transmission.
209
References
[1] IEC 62271-100 High-voltage switchgear and controlgear - Part 100: High-voltage
alternating-current circuit-breakers (October 2006)
[2] IEC 17A/768/CDV - Draft amendment 3 to IEC 62271-100 (November 2006)
[3] ANSI/IEEE C37.09-1979 - Test Procedure for AC High-Voltage Circuit Breakers Rated on
a Symmetrical Current Basis (October 1979)
[4] IEEE Std C37.09-1999 - Standard Test Procedure for AC High-Voltage Circuit Breakers
Rated on a Symmetrical Current Basis (June 1999)
[5] IEEE PC37.06 -D8.2 Draft: Standard AC High-Voltage Circuit Breakers. Rated on a
Symmetrical Current Basis – Preferred Ratings and Related Required Capabilities for
Voltages above 1000 Volts (April 2006)
[6] IEEE Std C37.09a – IEEE Standard Test Procedure for AC High-Voltage Circuit Breakers
Rated on a Symmetrical Basis – Amendment 1 : Capacitance Current Switching September
2005).
[7] IEC 60694 – Common specifications for high-voltage switchgear and controlgear standards
(January 2002)
systems”
210
5 Technical Specifications of Substation Equipment 1000 kV
and above
5.1 UHV Power Transformers
Russia has been operating its 1200 kV transmission system since the 1980s, and China and Japan
have been engaged in ongoing projects related to 1100 kV transmission systems. Table 5.1.1 lists
the major specifications for the UHV transformers used in these projects. The main points
determined from the survey are as follows: The transformers range from 2000 to 3000 MVA in
bank capacity, which is much greater than that for 800 kV transformers. Transportation
restrictions can greatly influence decisions on structure, transportation weight and size of UHV
transformers. Cooling-type transformers, either OFAF or ODAF, are also applied as a major
characteristic of large-capacity transformers. Although not indicated in the table, some
transformers are equipped with an on-load tap changer on the neutral-point side of the windings
to regulate the high-voltage-side windings. So, another feature of very high voltage transformers
is on-load tap changers.
India is planning future projects related to a 1200 kV transmission system. This survey would be
applied to consider the detailed specifications.
Table 5.1.1 Major items for UHV power transformers (for substation)
Item Specifications
Country Russia Italian China Japan
Utility - - SGCC TEPCO
Type Auto-transformer
Phase Single
Um (kV) 1200 1050 1100 1100
Rated Voltage (kV) 1150/ 3 1000/ 3 1050/ 3 1050/ 3
2550
BIL (kV) 2250 2250 1950
(2250)
Rated Power (MVA) 2000/3 1200/3 3000/3 3000/3
Short circuit impedance 12.5 percents 15 percents 18 percents 18 percents
Cooling - - OFAF ODAF
Sound level dB(A) - - 75 65
Additional information is provided in Tables 5.1.2 and 5.1.3, which list detailed specifications for
Italy and Japan’s individual projects.
Table 5.1.2 Detail specification of power transformers in Italy’s project

Auto-transformers
Rated voltage (kV) 1000/ 3 / 400/ 3 / 12.2
Rated power (MVA) 400/400 (single-phase unit)
Rated LIWV (kVp) 2250/1300/95
Rated SIWV (kVp) 1800/consequent
Short circuit voltage (percent) 15
Table 5.1.3 Detail specification of power transformers in TEPCO project
211
Item Specifications
Primary 1050/ 3 kV
Rated
Secondary 525/ 3 kV
voltage
Tertiary 147 kV
Rated capacity 3000 MVA/3 × 3
Tertiary capacity 1200 MVA/3 × 3
Primary tapping range ± 7% (27 taps)
Impedance 18 %
Test LIWV Primary 1950 kV
voltage AC Primary and Secondary 1.5E (1 hour) - 3 E (5 min) - 1.5E (1 hour)
5.2 UHV Voltage Transformers (VT), Current Transformers (CT)

5.2.1 UHV VTs
Table 5.2.1 lists the technical specifications for UHV voltage transformers in Japan, Italy and
Russia, obtained from the world survey conducted by CIGRE WG A3.22. UHV transmission
systems utilize optical- or capacitive-type VTs, which have secondary voltage ranging from
110/ 3 to 100 V, similar to that specified for 800 kV VTs. Accuracy ranges from 0.2 to 1% for
measuring applications and 1 to 3% for protection applications. Reported applications for optical-
type VTs include photoelectric sensors and electronic circuits in UHV systems. Their temperature
and frequency characteristics are summarized in the table.
Table 5.2.1 Specifications for UHV VT
Country Japan Italy Russia
(Highest voltage) (1100 kV) (1050 kV) (1200 kV)
Classification Non conventional ITs Non conventional ITs Conventional ITs
Capacitive type with secondary
amplification: Two secondary
Type Capacitive and Optical type Capacitive type
taps: one for measurement and
one for protection
Standard - - Temporary Russian specification
2900kV with GSOA,
Lightning impulse voltage 2250kV 2250 kV
2400kV with MOV
2100kV with GSOA,
Switching impulse voltage NA 1675 kV
1800kV with MOV
1.5E(30min.)- 3E (1min.) 1150kV (1min with GSOA),
Power-frequency voltage 910 kV (1 min)
-1.5E(30min.) 1100kV (1min with MOV)
Rated primary voltage 1100kV / 3 1050kV / 3 1150kV / 3
Rated secondary voltage 110V / 3 100kV / 3 100V
Rated burden 1VA 30 VA 300/600VA
class 0.2 for measuring (better
Accuracy class 1T 0.5% / 3%
than) 3P for protection
±0.2% (Summer: +20~+40º C)
Temperature characteristics - -
±0.5% (-20~+40º C)
±5% at fifth harmonics
Frequency characteristics - -
±10%at seventh harmonics
Service Experience
10 years as a field test 3 years At full voltage about 3 years
212
5.2.2 UHV CTs
Table 5.2.2 lists the technical specifications for UHV current transformers in Japan, Italy and
Russia. UHV CTs have either an iron core or an air core. The type with an air core is employed in
order to avoid iron core saturation caused by increased short-circuit current with large DC time
constant. CTs with an iron core have a rated secondary current of 1 A. Secondary voltage of 20 V
is specified for CTs with an air core because they have a voltage output, in principle. Accuracy
ranges from 0.2 to 1% for measuring applications and 1% for protection applications. CTs with
specified transient characteristics are used for protection applications where the rated short-circuit
current is 50 kA.
Table 5.2.2 Specifications for UHV CT
Country Japan Italy Russia
(Highest voltage) (1100 kV) (1050 kV) (1200 kV)
Classification Conventional ITs Non conventional ITs Conventional ITs Conventional ITs
bus-bar side:
3-iron core
(protection)
core 1: TPY
Iron core type
Type Iron core (ring) type Air core (ring) type core 2, 3: TPX
(3-step cascade)
line-side:
2-iron core
(1meas.+1prot.)
core 1: TPX
Temporary Russian
Standard JEC-1201 - -
specification
Lightning impulse 2900kV with GSOA,
NA NA 2250 kV
NA NA 1675 kV
Power-frequency 1150kV (1min with GSOA),
NA NA 910 kV (1 min)
voltage 1100kV (1min with MOV)
bus-bar side: 1250 A
(1) 8000A for line protection, (2) 8000A for 8000A for busbar line-side:
Rated primary current 2000-4000A
measuring, (3) 4000A for TR protection protection core 1: 1250 A
core 2: 1200 A
Rated secondary
1A 20V 1A 1A
current /voltage
bus-bar side:
core 1: 5 VA
(1) 60VA for line protection, (2) 25VA for core 2,3: 30 VA 5P30 (1) 40VA for protection
Rated burden 2～6k Ω
measuring, (3) 25VA for TR protection line-side: (2) 20VA for measuring
core 1: 30 VA 5P30
core 2: 20 VA
bus: 5P(1) 5P30(2,3) 1% (4 relay outputs)
Accuracy class 1.0 ±1.0% (at 400～8000A)
line: 5P30(1) 0.2(2) 0.5% (one measuring outputs)
Secondary winding
(1)(2) less than 40Ω (3) less than 20Ω less than 10Ω - -
resistance
Secondary winding
NA less than 10Ω - -
reactance
Rated primary short-
50kA 50kA - -
circuit current
Primary time constant/
(1) 100msec/1.5cycle, 30msec/∞
Permissible time to accuracy 100msec/ 1.5cycle - -
(3) 100msec/6cycle
limit
Service Experience
(including on-site 10 years as a field test 10 years as a field test 3 years At full voltage about 3 years
experience)
213
Considerable research on UHV systems is carried out in Russia, Italy, USA, Japan, China, and
India. A commercial UHV network is presently operating in Russia and one is planned in China
for the near future and in Japan and India in the foreseeable future.
CIGRE WG A3.22 conducted a world survey on the specifications adopted for UHV substation
equipment. Basic specifications for circuit breakers in Russia, Italy, Japan, and China’s UHV
projects are listed in Table 5.3.1. Each project’s TRV duties without opening resistors are shown
in Table 5.3.2, and those with opening resistors are shown in Table 5.3.3. TRV duties without
opening resistors are not specified in Japan’s projects.
Table 5.3.1 Basic Specifications for circuit breakers of UHV projects

Italy China
Country Russia Japan
(Pilot Plant) (Pilot Plant)
Highest voltage for
1200 1050 1100 1100
equipment (kVrms)
Rated short circuit
63 kA 50 kA 50 kA
current (kA)
Rated nominal current 8000 (bus bar)
8000
(A) 6000 (line bay equipment)
2900 (with GSOA), 2250
LIWV (kV) 2250+600 2250+900, 2400+900
2400 (with MOV) 2250+1100 2 / 3
2100 (with GSOA),
SIWV (kV) 1675+600 1550 1675+900, 1800+900
1800 (with MOV)
1150 (1min with GSOA), 1.5E (30min.)- 1.5pu(30min)+
PFWL (kVrms) 1100 (1min with MOV) 910kVrms 3 E(1min.)- 3 pu(1min)+
2000 (between opened contact) 1.5E(30min.) 1.5pu(30min)
NA Closing/opening Closing/opening
Closing/Opening resistor 500 ohm resistor 700 ohm
resistor Insertion time= 10ms for
making, 35ms for breaking
3-phase: O-0,3s-CO-60s- Auto-reclose time: Auto-reclose time:
Operating sequence CO; 0.3sec 0.3sec
1-phase: O-1s-CO-60s-CO
Short-time withstand 2.5 p.u. 2.7 p.u. (50Hz)
Making current peak
current: 158kAp
FPCF 1.5 at T30 1.3 1.1 1.3
150ms (DC component 45ms
DC time constant at opening=80% , di/dt Special time
(ms) at breaking =28A/µs constant:120ms
(equivalent to 63kA)
Service experience 10 years as a filed test
214
Table 5.3.2 TRV Duties for circuit breakers of UHV projects (without resistor)
Country Russia Italy (Pilot Plant) China** (Pilot Plant)
T100s: Uc=1518kV, T10: Uc=1790kV, tp=200ms, T10: Uc=1786kV (AF=1.53), Uc/t3=7kV/µs, t3=255µs
u1/t1=2.8kV/µs Uc/t1=9,0kV/µs T30: Uc=1786kV (AF=1.53), Uc/t3=5kV/µs, t3=357µs
T60: Uc=1520kV, tc=800µs, T60: Uc=1751kV (AF=1.5), u1/t1=3kV/µs, t2=1752µs
TRV for BTF u1/t1=3,3kV/µs u1=876kV, t1=292µs
without breaking T100s: Uc=1370kV, tc=1000µs, T100s: Uc=1635kV (AF=1.4), u1/t1=2kV/µs,
resistor u1/t1=3,3kV/µs t2=1752µs, U1=876kV, t1=438µs
T100a: Uc=998kV, tc=1045µs, T100a:Uc-=0.98p,u, Uc+=0.55p.u, t1=438µs, U1-
u1/t1=2,7kV/µs =0.74p.u, U1+=0.76p.u, t2=1752µs, u1/t1=1.48kV/µs
(minor loop), u1/t1=1.52kV/µs (major loop)
(Source-side) u1=943kV t1=285µs, Uc=1370kV,
tc=1000µs, u1/t1=3,3kV/µs
TRV dutiy for SLF
(Line-side)
without breaking L60: u0=378kV, ul=604kV, tl=120µs, ul/tl=5,0kV/µs
resistor L75: u0=236kV, ul=377kV, tl= 60µs, ul/tl=6,3kV/µs
L90: u0= 94kV, ul=151kV, tl= 20µs, ul/tl=7,6kV/µs
Out-of-phase TRV: Uc=2841kV, Uc=1890kV, tp=150µs, Amplitude factor: 2.0
without breaking u1/t1=2.25kV/µs Uc/t1=12,6kV/µs TRV: Uc=2245kV, u1/t1=1.54kV/µs, t2=1750-3500µs ,
resistor Breaking current: 16kA Breaking current: 12.5kA
voltage factor: 1.5 Breaking current=900A Voltage factor: 1.4
Capacitive Capacitive u1=198kV t1=200µs Uc=1614kV Capacitive current:1200A
current switching current:1160-2030A tc=8ms u1/t1=1kV/µs
(line= 400-700km)
** TRV duty of China’s project needs to be reviewed within the WG.
Table 5.3.3 TRV Duties for circuit breakers of UHV projects with opening resistor
Country Italy (Pilot Plant) Japan
Breaking Unit-with resistor Main contact :T16(8kA):Uc=1385kV, t2=461µs,
T10: Uc=798kV, tp=565µs, Uc/t1=1,4kV/µs Uc/t1=3kV/µs , T30 –T100:Uc=1385kV,
TRV for BTF
T60: Uc=1261kV, tc=3150µs, u1/t1=2,5kV/µs u1=990kV,u1/t1=3kV/µs, t2=1485µs
with breaking resistor T100s: Uc=1232kV, tc=1025µs, u1/t1=3,1kV/µs Resistor contact: Uc=1200kV, u1=550kV,
T100a: Uc=860kV, tc=1045µs, u1/t1=2,2kV/µs u1/t1=3kV/µs, t2=550µs
Main Breaking Unit with R Main contact :
(Source-side) u1=756kV, t1=274µs, Uc=1192kV, Z=450 ohm, AF=1.4
tc=1010µs, u1/t1=2,7kV/µs (Conductor: 810mm sq. 8-conductors)
TRV dutiy for SLF with
(Line-side) Power Source side: T100*0.9
breaking resistor L60: u0=378kV, ul=378kV*, tl=120µs,ul/tl=3,2kV/µs
L75: u0=236kV, ul=236kV*, tl= 60µs, ul/tl=3,9kV/µs
L90: u0= 94kV, ul=94kV*, tl= 20µs, ul/tl=4,7kV/µs
with resistor: Uc= 942kVb (TRV critically damped), Main contact , Breaking current=12.5kA
tp=500µs, Uc/t1= 1,9kV/µs u1=1200kV, Uc=2160kV, u1/t1=2kV/µs,
Out-of-phase Auxiliary BU: u1=732kV t1=150µs, Uc=1197kV, t2=1800µs
with breaking resistor tc=2700µs u1/t1=4,9kV/µs, Operating sequence: Resistor contact :
CO Breaking current=2kA, u1=1750kV, Uc=2160kV,
u1/t1=2kV/µs, t2=1800µs
Main contact
Breaking current=1kA, TRV peak=900kV
Resistor contact :
Condition 1: Breaking current=1kA, TRV
Capacitive current peak=2515kV（1.4pu）,Waveform: (1-cos)
switching Condition 2: Breaking current=0.1kA, TRV
peak=2100kV（1.17pu）, Waveform: (1-cos)
Condition 3:Breaking current=0.6kA, TRV
peak=2000kV（1.11pu）, Waveform: sin
rrrV=0.34kV/µs
5.4 DS, ES
Table 5.4.1 lists the disconnector specifications adopted in Russia, Italy, Japan, and China, as
indicated in the world survey conducted by CIGRE WG A3.22.
215
Table 5.4.1 Specifications for disconnectors of Russia, Italy, Japan, and China
Italy China
Country Russia Japan
(Pilot Plant) (Pilot Plant)
Um ; Highest
voltage for 1200 1050 1100 1100
equipment (kV)
Rated short circuit
63 50 50
current kA
2250
2900 (with GSOA)
LIWV (kV) between contacts: 2400+900
2400 (with MOV)
2250+1100 2 / 3
2100 (with GSOA), 1800kV, 1675+900kV
SIWV (kV)
1800 (with MOV) 3
1.5E (30min.)- 3 E(1min.)-
1150 (1min with GSOA) 1.5E(30min.) 1100
PFWL (kVrms)
1100 (1min with MOV) AC voltage for between 1100+635
contacts : 1265kV rms (1min.)
Closing/Opening
Resistor
Test voltage: 300Vrms Test voltage: 400V
Bus-transfer current
rrrV = 400V/μsv Test current: 1600A
switching tests
Test current: 8000A
Test voltage: 635kV Test voltage: 635kV
Bus-charging
Current: 0.5A (Load side Current: 0.5A
switching tests
capacitance : 2000pF )
Opening and closing resistor
Withstand voltage = 500ohmv, Operation = CO
and energy tests of Voltage across resistor for
closing/opening thermal capacity :
resistor 1700kV(Lightning impulse)
Energy Absorption: 25kJ
5.5 High-Speed Grounding Switches (HSGS)

Table 5.5.1 lists the specifications for UHV high-speed grounding switches (HSGSs) in Japan.
Table 5.5.1 Specifications for HSGS of Japan
Country Japan
Electromagnetic induced
current=5800 Arms, TRV=640 kVp, Uc/t3 = 0.79 kV/µs, (OHL length=70-200 km)
current=7000 Arms, TRV=410 kVp, Uc/t3 = 1.15 kV/µs, (OHL length=40-70 km)
Induced current
switching tests
Electrostatic induced
Current=1200 Ap, TRV=900 kV (1-cos) , (OHL length =70-200 km)
Current= 520 Ap, TRV=700 kV (1-cos) , (OHL length =40-70 km)
Operating sequence =C-(θ) O (θ= 0.5s)
HSGS shall have a interruption window more than 80ms for Delayed current Zero
phenomena.
TRV: Electromagnetic induced
Induced current current=3500 Amean, TRV=170 kVp, Uc/t3 = 0.26 kV/µs, (OHL length= 200 km)
switching tests for
Delayed current Zero current=7830 Amean, TRV= 65 kVp, Uc/t3 = 0.46 kV/µs, (OHL length= 40 km)
phenomena Electrostatic induced
Current=3500 Amean, TRV=570 kV (1-cos) , (OHL length =200 km)
Current=7830 Amean, TRV=390 kV (1-cos) , (OHL length =40 km)
Long arcing time: 80 ms with current zero missing (special case)
216
5.6 Surge Arresters
A survey on arrester requirements for the UHV pilots and projects in Japan, China, Italy and
Russia provided general information on such issues as rated and continuous operating voltage,
protection levels and energy requirements. However, many important arrester parameters are not
completely addressed in available papers.
5.6.1 Continuous operating voltage, Uc, rated voltage, Ur, and TOV
capability
Continuous operating voltage is generally specified as Um/√3, which follows IEC procedures.
Rated voltages are given in Table 5.6.1.
Table 5.6.1 Um and arrester rated voltages for UHV pilots and projects
UHV project in Russia Italy Japan China
Um –kV 1200 1050 1100 1100
Rated voltage (Ur) - kV 800 750 826 828
In China’s projects, the specified maximum temporary overvoltage is 1.4 p.u. with a duration of
0.4-0.5 s on the line side of the breakers. [1], [2] In Japan, arresters are designed for a TOV
durability test with 1.55 p.u. for 0.17 s. [3] However, normal TOV of 1.1 p.u. is expected at a
single line-to-ground fault, and 1.3 p.u. is considered rare. In Russia, the maximum phase-to-
ground voltage at a ground fault reaches 1.4 p.u. With the given TOV values, it should be
possible to select a lower rated voltage for the UHV system in China and Russia. A rated voltage
of 740-775 kV would be possible in a case such as Italy’s pilot plant.
5.6.2 External insulation strength

LIWV of 2250 kV is specified for the GIS arresters in Japan, [3] which is the same as for the
GIS, but no specific requirements are found for arresters used in other cases. Presumably, the
arrester insulation requirements are the same as for other equipment. Strictly following the
existing IEC standard [4] for 10 and 20 kA arresters with a rated voltage of 200 kV and above,
the lightning impulse withstand voltage shall be ≥1.3 times the lightning impulse protection level,
and the switching impulse withstand voltage shall be ≥1.25 times the switching impulse
protection level. These factors include an altitude correction factor, Ka, of 1.13 that takes into
account an altitude of 1000 m and is calculated from the formula Ka = em(H/8150) setting the m-
factor equal to 1.
For UHV arresters, the factor for switching impulse withstand voltage should be reconsidered
since the m-factor significantly reduces the withstand voltage levels in the UHV range.
Otherwise, the requirement for arrester insulation will be higher than that for other equipment for
which an m-factor is considered.
5.6.3 Residual voltages and classifying currents

For lightning protection in Japan, China and Italy’s systems, characteristic value at 20 kA is
selected as the coordinating current. In Russia, the value at 15 kA is specified. LIPL of 1.8 p.u. is
commonly used except in Italy’s pilot project, where a protection level of 2.1 was applied. This
most likely reflects the state of surge arrester technology at the time. The actual protection level
of the delivered arrester was 2.0 p.u. [5] In Russia, higher values were used at an earlier stage
217
when only SiC gapped surge arresters were available. In Japan, 4 parallel columns with high-
performance ZnO elements, which have seen extensive field use as 550 kV GIS arresters, have
achieved the low LIPL. The V-I characteristics of the 1100 kV arrester are shown in Fig. 3.1.1 of
the chapter 3.
The switching surge protection level is around 1.6 p.u., except for Italy where 1.69 p.u. is
adopted. Coordinating currents are specified as 2–3 kA. For Japan’s project, the current for
switching surge protection is not officially specified at present. However, the actual protection
level of the arrester at 2–3 kA would be around 1.6 p.u.
5.6.4 Energy requirements

The required energy capability for arresters in Italy’s project is specified as only 3–4 kJ/kV. [6]
However, the actual capability of the arrester is 17 MJ, [7] presumably due to the use of a
multiple-column arrester to obtain as low a protection level as possible. High energy capability is
then automatically obtained.
In both Japan and China, TOV is mentioned as the deciding factor in energy requirements. This is
obvious in Japan’s case with a TOV of 1.55 p.u. with a duration of 0.17 s considering the arrester
characteristics specified. [3] However, for China, the calculated arrester energy at 1.4 p.u. of 0.4 s
with approximately the same arrester characteristics as in Japan is reported to be 8.6 MJ, [2]
which seems very high.
Energy resulting from switching surges is calculated as only 3.26 MJ. Based on simulations,
20 MJ is specified in China and 55 MJ in Japan. In China, thermal stability after two energy
discharges of 20 MJ is required. The required test procedure used to verify thermal stability for
Japan’s GIS arrester is shown in Fig. 5.6.1.
Step 1 Step 2 Step 3 Step 4
AC voltage test Lightning impulse Rectangular wave current and Temporary overvoltage
high current test thermal stability tests withstand test
Rated voltage Maximum AC TOV

65kA 4/10µs 2000A-2ms Continuous
Continuous Continuous Operating Voltage
cooling
cooling cooling
2sec
115 C-70h 5min 2min. 1min. 30min. residual voltage t(s) 30min.
measurement
µs and 2000A-2ms current duties are intended for test of one ZnO column.
Note 1: 65kA-4/10µ
Note 2: 2000A current duty is intended for evaluation of scattering of current sharing to 4 ZnO columns.
Fig. 5.6.1 Stability Evaluation Test according to JEC-217 1984
5.6.5 Lightning impulse discharge capability

Discharge duties relevant to lightning were not addressed. The reason for this may be that
lightning is considered negligible compared with the anticipated stress from TOV and switching.
However, the high insulation strength of the transmission lines may give significant arrester
energy in the case of shielding failure. [5].
218
5.6.6 High current impulse withstand capability
In the IEC standards, the conditioning test in the switching surge operating duty test comprises 2
high-current impulses with an amplitude of 100 kA and waveform of 4/10 µs.
Japan’s UHV arresters have been tested using high-current impulses of 65 kA, the same as
Japan’s standard value for 10 kA arresters considering the current to be reduced by the
4×parallel-column arrangement; however, for China and Italy’s projects, 100 kA is specified.
This most likely reflects the higher applicability of the standards compared to prospective arrester
stress.
5.6.7 Pollution performance

Pollution does not pose a threat to GIS arresters. However, such risks as external flashover,
partial heating of the active part and internal partial discharge tend to increase for AIS arresters.
A certain creepage length on the insulators is normally specified for AIS arresters. In Russia,
21600 mm is specified and in China around 25 mm per kV (system voltage). Specific tests for
verifying AIS arresters under polluted conditions are not covered in available papers.
5.6.8 Mechanical strength

Information on mechanical strength is not presently available. However, due to the size of UHV
arresters, AIS in particular, the mechanical requirements are demanding. Different solutions
comprising high-strength fiberglass tubes with external silicone rubber insulation for pedestal
mounting or molded designs with silicone rubber insulation for suspended mounting have been
suggested [5].
5.6.9 Short-circuit rating

Italy’s project is the only one for which a short-circuit current rating of 63 kA is specified. The
other projects also have requirements on short-circuit performance, but they have not yet been
published in available papers. Considering the huge size of UHV arresters and possible severe
consequences of an arrester exploding due to overload, performance under short-circuit
conditions is an important issue. Furthermore, AIS arresters are most probably located very close
to other equipment in order to achieve the desirable protection, which of course makes it even
more important that a possible arrester failure occurs in a “controlled” way.
5.7 Shunt Reactor

EHV/UHV shunt reactors have retained their position as a reliable, less expensive source of
consumption of excessive reactive power from transmission lines. In comparison with shunt
reactors connected to the tertiary winding of an EHV/UHV transformer, the EHV/UHV shunt
reactor provides two important benefits: it is more effective because it is not placed beneath the
reactance of the transformer, and it can limit TOV when connected to the line end. Practically,
connecting EHV/UHV shunt reactors to the line is necessary if the line length exceeds
approximately 300 km at 750 kV or 200 km at 1000–1200 kV. Such shunt reactors must be
switchable, and proper automation or spark gaps (used in Russia) are needed to ensure fast
connection to the line for limiting switching overvoltage and TOV at possible line faults or
switching errors. In combination with FACTS technology, EHV/UHV shunt reactors provide
219
good control of planned and fault regimes. Last year, magnetic-controlled shunt reactors
appeared on the market, but this technology has not yet been applied directly to 750 kV or higher
voltages and it is mainly limited to lower voltages or to transformer tertiary winding connection
schemes.
Although similar to EHV/UHV transformers in many design details, EHV/UHV shunt reactors
have some specific features. First, they may be switched approximately 100 times more
frequently than transformers with a correspondingly higher number of switching overvoltages
applied to their insulation. Second, when placed on the remote end of the switched line they
experience more severe TOVs than transformers. Third, they require more of a linear
magnetization curve then transformers in order not to become a powerful source of higher
harmonics, which frequently leads to a gap in their magnetic core and magnetic shields
preventing overheating of their tank. Four, if used in 4-legged reactor schemes for secondary arc
suppression, they may need a somewhat increased level of insulation for their neutral.
Unfortunately, less attention is paid to the requirements and design of UHV shunt reactors
compared that that for the transmission lines, circuit breakers, arresters or transformers, so
available information is very limited. In Russia, 1150 kV shunt reactors were produced and put in
operation and UHV shunt reactors will be an integral part of the 1000–1200 kV systems under
development in China and India.
The main characteristics of UHV shunt reactors in Russian, China and India are summarized in
Table 5.7.1. [7], [8], [9], [10]
Table 5.7.1. Main characteristics of UHV shunt reactors

Country Russia China India
Nominal/Maximum operating voltage, kV 1150/1200 /1000 1200/1250
Capacity of 3-phase group, MVA 900 600, 720 or 960 500 or 660
(preliminary estimate) (preliminary estimate)
Positioned on line ends? Yes Yes Yes
Switchable? Yes Yes Yes
Automation for fast switching? Spark gaps plus ?
automation
Application of 4-legged reactor Yes Yes
Permissible TOVs, L-G, p.u. 20 s-1.30, 5 s-1.35 0.5 s-1.4 1.35
Level of switching surges on line end, L-G, p.u. 1.8 (1.6) * 1.7
Power frequency 1-min withstand test voltage, L- 1100 (1000) * 1100 (5 min)
G, for line side, kV ef
Same for the neutral 120 ? ?
Power frequency 1-hr withstand test voltage, L-G, 900 950
for line side, kV
Switching surge withstand test voltage, L-G, kVp 2100 (1800) * 1800
Lightning impulse withstand test voltage for 2550 (2250) * 2250
internal insulation, L-G, on full wave, kVp
Same on chopped wave 3200 (2550)* 2475
Note: * Values not within parentheses are for Russian equipment developed prior to 1985, protected by zinc oxide gapped
arresters and actually installed on 1150 kV transmission lines; Values within parentheses are for equipment newly developed and
protected with MOVs.
220
References
[1] China 2006, ACTECH006, ”Study on Insulation Coordination in 1000kV UHV AC
Transmission Project”.
[2] China 2006, ACTECH007, ”Study on Overvoltage and Insulation Coordination for 1000
kV AC Transmission System”.
[3] IEC/CIGRE UHV symposium Beijing 18-21 July 2007, paper 2-5-1. “Development of
1100 kV AC GIS-arrester and the Verification Tests”
systems”
[5] IEC/CIGRE UHV symposium Beijing 18-21 July 2007, paper 2-5-1. “MO surge arresers
for systems above 550 kV – Experience and challenges for the future –“
[6] A3-07(WG22)141 “Italien 1000 kV project” by E. Colombo.
[7] Reference Book on HV Electric Installations, I. Baumshtein and S. Bazhanov, Editors,
Energoatomizdat Publishing House, Moscow, 1989, 768 pp. (in Russian)
[8] Gu Dingxie et al, “Study of Overvoltage and Insulation Coordination for 1000 kV AC
Transmission System”, IEC-CIGRE UHV Symposium, Beijing, July 18-21, 2007
[9] Shen Hong et al., Reactive Power Characteristics and Compensating Measures of UHV
Transmission Systems, Ibid
[10] Du Shuchum et al., “Study of Insulation Coordination in 1000 kV UHV AC Transmission
Project”, Ibid
221
6 Conclusions and Future tasks
Technical specifications and service experience from all major UHV projects and trials have been
reviewed in conjunction with relevant data from a number of commercially operated 800kV
systems. From this review it is clear that there a number of technical topics which are particularly
challenging in the UHV range and that, in many cases, simple extrapolation of assumptions from
lower voltages is not appropriate. Several distinctive phenomena have been identified including
a prominent Ferranti effect, large DC time constant of fault current, severe TRVs and slow front
overvoltages, prolonged secondary arc extension and reduced first-pole-to-clear factor. All of
these require study in more detail if UHV standardisation is to be successful.
To date the following conclusions can be drawn:
(1) Optimal insulation coordination to reduce the construction costs of UHV systems by
applying compact transmission towers can be realised using high-performance
MOSAs. The design of insulation coordination by means of accurate computer-
aided simulations is common practise for such projects.
(2) The application of MOSAs can reduce the amplitude factors of TRVs for terminal
faults, transformer limited or secondary faults and out-of-phase switching. Where
such mitigation is applied circuit-breakers specifications could be modified
accordingly however such reduction in TRV is not necessarily possible for so called
long line faults where the TRV is generated at both sides of the circuit-breaker
terminals.
(3) Circuit-breakers with opening resistors in combination with MOSAs can effectively
suppress the TRV peak values and rates of rise for terminal faults, long line faults,
transformer limited faults and out-of-phase switching. Where such mitigation is
employed specifications should take account of the reduced requirements however
without the application of opening resistors, careful specification of kaf and kpp or
even special test duties will be required.
(4) Several distinctive phenomena such as large DC time constant in the fault current,
severe TRVs and secondary arc extension are created by the design of UHV
transmission lines. Their effects on the specified duties of equipment should be
considered in greater detail prior to standardisation.
(5) A range of DC time constants associated with fault conditions have been specified
for UHV projects past & present, often exceeding the 75ms value specified in IEC
for 550kV systems. Using the common assumption that the influence of the high DC
component on the T100a test duty can be evaluated by the energy of the last major
loop before the interruption, slope of current at the time of interruption and TRV
characteristics, a value of 120ms appears to be appropriate to meet most UHV
requirements. This value will be recommended as the special time constant for
asymmetrical currents in test duty T100a based on international investigations.
(6) The line surge impedance for L90 and L75 duties could be reduced for rated
voltages of 1100 kV and above on the basis that bundle contraction is unlikely to be
completed when interruption occurs (within 50-80ms from fault initiation). This
results from UHV transmission lines having 6 or more, large cross-section
conductors, strung at high tensions.
222
(7) UHV shunt reactor switching overvoltages and mitigation measures (such as
controlled switching) remain a topics for deeper investigation and development.
(8) Rapid auto-reclosing can be realised with either a high speed grounding switch or a
four-legged shunt reactor scheme to suppress the secondary arc. A dead time of 0.3
seconds as commonly used for rated voltages up to 800 kV can be applied to UHV
operating sequences.
Having reached these conclusions, WG A3.22 will continue to work towards developing a robust
basis for UHV standardisation for substation equipment within the scope of SC A3 by pursuing
the following activities:
• Collect further service experience from 800 kV systems and the latest testing experience
from the UHV projects to ensure that “lessons learnt” are captured and disseminated.
• Study the effect of line length (especially long lines) on temporary, transient and
switching voltages and overvoltages (TOV, SFO, FFO, TRV).
• Compare the data collected to date with new information from the Chinese (and maybe
other) projects.
• Extrapolate the detailed studies and testing experience performed in Japan to address
more general requirements.
• Consider, in a more general context, the physical effects and application possibilities of
elements such as high grade MOSAs, opening resistors, MOVs in parallel with arcing
chambers, controlled switching, etc.
• Consider, in a more general context, the effects, application possibilities, and relative
merits of methods of secondary arc suppression (e.g. 4 legged shunt reactors, HSGS,
switchable shunt reactors).
• Make specific recommendations for generally applicable values of key parameters for
incorporation into future standards. As a minimum recommendations will be made with
respect to first pole to clear factor(s), amplitude factors(s), peak factor/damping of
transients, DC time constants, the need to cater for HF-components in SC-current, line
surge impedance, out-of-phase switching conditions, effects of phase angle under stable
condition, rate of rise of recovery voltage for transformer limited faults, TRV peak for
long line faults, requirements for high speed grounding switches and for MOSAs, and
capacitive (line) switching parameters.
• Review the service experience with 800 kV networks and equipment and make
recommendations for the revision of the existing standards for EHV equipment if these
are found to be deficient.
• Continued liaison with other Study Committees & Working Groups to ensure technical
consistency within pre-standardisation recommendations.
This work will be reported in future technical brochures.
223

362 Technical Requirement of 800 KV Substation and Above

Uploaded by

Copyright:

Available Formats

362 Technical Requirement of 800 KV Substation and Above

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

362 Technical Requirement of 800 KV Substation and Above

Uploaded by

Copyright:

Available Formats

What is the purpose of Working Group A3.22?

What is the purpose of Working Group A3.22?

What are some specific issues and technical challenges mentioned for UHV AC systems?

What are some specific issues and technical challenges mentioned for UHV AC systems?

362

Field experience and technical specifications of Substation equipment up to 1200 kV

H. Ito (Convenor), A. Janssen (Secretary),

ISBN : 978-2 85873-049-0

3 Specific topics ........................................................................................................................77

5 Technical Specifications of Substation Equipment 1000 kV and above .............................211

6 Conclusions and Future tasks...............................................................................................222

ACSR Aluminium Conductor Steel Reinforced

BPA Bonneville Power Administration (USA)

CCGT Combined Cycle Gas Turbine

EHV Extra High Voltage

FFO Fast Front Overvoltage

HSGS High Speed Grounding Switch

IEC International Electrotechnical Commission

KEPCO Korea Electric Power Corporation (Korea)

LDC Line Discharge Class

MAIS Manœuvres Automatiques d’Inductances Shunt

NCI Non-Ceramic Insulator

PD Optical potential devices

TCSC Thyristor-Controlled series Capacitor

UCTE Union for the Coordination of Transmission of Electricity in continental

VEI Russian Electrotechnical Research Institute (Russia)

1.2 Specific phenomena peculiar to UHV AC systems

1.3 Summary of future tasks by equipment type (SC A3 only)

1.3.2 Disconnectors (DS), earthing switches (ES) and high speed

1.3.3 Metal oxide surge arrester (MOSA)

1.3.4 Instrument transformers

2.2.1 Features of American Electric Power’s 800 kV transmission

2.2.2 Overview of field experiences and the evolution of AEP’s 800 kV

Conventional 4 Legged Modified 4 Legged

Figure 2.2.1 Modified reactor schemes

2.2.3 Specific issues and technical challenges for the operation of

2.3 IREQ/Hydro Québec project

2.3.1 Features of Hydro Québec 765 kV transmission system

Fig.2.3.1 Hydro-Québec 735kV series-compensated transmission systems

2.3.2 Evolution of Hydro Québec 765 kV transmission network

2.3.2.1 Actual 765 kV transmission network

2.3.2.2 765 kV substation layout

2.3.3 Overview of field experience and the evolution of Hydro-

2.3.4 Specific issues and technical challenges for the operation of

Table 2.3.1 Hydro Québec specifications

2.4 Korea’s project

2.4.1 Features of KEPCO 800 kV transmission system

2.4.2 800kV GIS substation

Fig.2.4.2 KEPCO 800 kV GIS substation

Table 2.4.2 Switching overvoltage (1p.u.=800 2 / 3 kV)

Fig.2.4.3 KEPCO insulation distance of substation outgoing part

2.4.4 Technical specifications of 765 kV power transformer

2.4.5 800 kV Gas insulated switchgear

2.5.1 Features of Furnas’ 800 kV transmission system

Fig.2.5.1 FURNAS main assets

2.5.2 Overview of field experience of 800 kV long distance

2.5.3 Evolution of Furnas’s 800 kV AC Transmission Systems

Fig.2.5.4 Detailed view of the Furnas transmission system for

2.5.3.1 Application philosophy for 800 kV Shunt reactor (SR)