Professional Documents
Culture Documents
Insulation Coordination For UHV AC Systems PDF
Insulation Coordination For UHV AC Systems PDF
COORDINATION
FOR
UHV AC
SYSTEMS
WG C4.306
TypethenameofTB
Members
E.Zaima,Convenor(JP),T.Koboyashi,Secretary(JP),J.Takami,AsistantSecrery(JP),P.C.
Fernandez(BR),D.Peelo(CA),Q.BuiVan(CA),W.Chen(CN),A.Sabot(FR),F.Gallon(FR),E.
Kynast(DE),A.Pal(IN),R.N.Nayak(IN),S.Malgarotti(IT),T.Yamagiwa(JP),E.Shim(KR),A.
Lokhanin(RU),P.Tlhatlhetji(ZA),A.Amod(ZA),C.vanderMerwe(ZA),U.Kruesi(CH),D.
Sologuren(CH),Y.Vachiratarapadorn(TH),A.J.F.Keri(US),A.Villa(VE),G.Carrasco(VE),H.Ito
(JP),T.Yokota(JP),Y.Shirasaka(JP),B.Richter(CH),U.Riechert(CH)
Coordinationwith
P.Zhou(CN),J.Lin(CN),Z.Li(CN),K.Uehara(JP),Y.Ishizaki(JP),S.Okabe(JP),M.Miyashita(JP),
H.Kajino(JP)
Copyright 2011
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.
Page 1
TypethenameofTB
Page 2
TypethenameofTB
INSULATION COORDINATION
FOR
UHV AC SYSTEMS
NoExtraCoverpageorblankpages
UseCIGREabbreviationsinsofaraspossible:StudyCommitteeSC,TechnicalBrochureTB,WorkingGroupWG
JointWorkingGroupJWG,TechnicalCommitteeTC
Photos:mustbeofreasonable definition (preferably 300 dpi);allfiguresandtablesmustbetitled,legibleandnumbered
withlegendsprovided.
NoCompanylogos
Table of Contents
1 Introduction .................................................................................................................6
2 Concept of recent practices on insulation coordination for the UHV and the 800kV
system ..........................................................................................................................8
2.1 Insulation coordination throughout substation and transmission line...................8
2.2 Reduction of insulation levels using overvoltage suppression measures .............16
References ..............................................................................................................19
3 Recent practice on insulation coordination for the UHV and the 800 kV system...........20
3.1 China UHV project .............................................................................................20
3.2 India UHV project ..............................................................................................29
3.3 TEPCO 1100 kV project .....................................................................................35
3.4 Brazil 800 kV project .........................................................................................46
3.5 China 750 kV AC Project....................................................................................60
3.6 India 800 kV project ..........................................................................................64
3.7 Korea 765 kV project.........................................................................................68
References ..............................................................................................................82
Page 3
TypethenameofTB
Page 4
TypethenameofTB
Page 5
TypethenameofTB
1 Introduction
Different countries in the world are planning and realizing UHV AC systems with operating voltages exceeding
800kV. When planning a new power system, in particular at a new voltage level, insulation coordination is one of
the most important subjects. The main task is the determination of stresses and the assessment of the strength of
the system and the equipment installed.
The general procedure of insulation coordination is described in IEC 60071-1 (2010). This standard does not give
precise advice regarding new voltage levels although it provides insulation levels for Um values of both 1100 kV
and 1200 kV. These insulation levels are based on both past experience available from former CIGR work that
also considered the 1000kV voltage level, and recent works in Japan, China and India. The research activities
within previous CIGR SC 33 in the topic of UHV transmission provided a good basis on overvoltages and air
insulation performance to make possible the design of air insulation for both 1100kV and 1200kV highest voltage of
equipment.
Since 1990s, metal oxide surge arresters have been applied to UHV substation design. Insulation coordination for
UHV has been changed based on these arresters throughout substation and transmission line. Also, gas insulated
switchgears (GIS, Hybrid-IS) have been generally applied to UHV substation design.
Considering the above issues, CIGR WG C4.306 has reviewed and discussed insulation coordination practice in
the UHV AC range taking into account the state-of-the-art technology, with special reference to surge arresters.
Such a review has been taken into account the accumulated knowledge of various CIGR working bodies, and
accomplished in collaboration with related CIGR SC A3 and B3 (WG A3.22, A3.28, B3.22 and B3.29).
Recommendation, for application guide IEC 60071-2 (1996) and IEC apparatus standards has been proposed.
The task of CIGR WG C4.306 is divided into four main sections dealing with (see Figure 1.1):
Recent practice on insulation coordination for UHV system:
- Insulation coordination throughout substation and transmission line
- Reduction of insulation levels by application of higher performance surge arresters and other overvoltage
suppression measures
Overvoltage in UHV range: (especially focused on peculiarity to UHV AC system)
- Determination of stresses (TOV, switching overvoltage, lightning overvoltage and VFTO) by simulation tools
and verification by measuring results
- TOV due to load rejection and ground fault
- Switching overvoltages caused by closing and opening (fault-clearing) with ground fault overvoltage
- Lightning overvoltage caused by back-flashover and direct lightning, VFTO stress in GIS due to disconnector
switching (ref to CIGRE brochure "Monograph on GIS Very Fast Transients 1989)
Review on insulation coordination of air gaps in the UHV range:
- Phase-to-phase insulation
Selection of insulation levels:
- Coordination withstand voltages and safety factors for equipment
- Selection of insulation levels for equipment and transmission lines
More than 250 written technical contributions have been prepared by 29 experts from 15 countries during the
investigations.
Page 6
TypethenameofTB
Recent practice of
UHV insulation
coordination
1.2
Collaboration with
A3.22&28 and
B3.22&29
U r/ 3
Page 7
TypethenameofTB
Page 8
TypethenameofTB
Table 2.1.1 Standard insulation levels for UHV (IEC 60071-1 Ed.8)
Table 2.1.2 Standard rated switching impulse withstand voltage and minimum
phase-to-phase clearance for UHV (IEC 60071-1 Ed.8.1)
Page 9
TypethenameofTB
LIWV
(Substation)
SIWV
(Substation)
Switching Overvoltage
Page 10
TypethenameofTB
System analysis
Insulation characteristics
Performance criterion
Statistical distribution (+)
Inaccuracy of input deta (+)
(+) Effects combined in a co-ord ination
factor Kc
Test conditions
Test conversion factor Ktc
Standard withstand voltages
Rages of Um
Rated or standard insulation level : set of Uw
NOTE In brackets the subclauses reporting the definition of the term or the description of the action.
Sided bo xes refer to required input
Sided bo xes refer to performed actions
Sided bo xes refer to obtained results
Figure 2.1.2 Flow chart for the determination of rated or standard insulation level
in IEC60071-1
NOTE: The definition of some terms in above figure as given in IEC60071-1 is summarized as follows;
Urp: Representative overvoltages: Overvoltages assumed to produce the same dielectric effect on the insulation as
overvoltages of a given class occurring in service due to various origins. They consist of voltages with the standard shape of the
class.
Ucw: Co-ordination withstand voltage: For each class of voltage, the value of the withstand voltage of the insulation
configuration in actual service conditions, that meets the performance criterion
Urw: Required withstand voltage: Test voltage that the insulation must withstand in a standard withstand voltage test to ensure
that the insulation will meet the performance criterion when subjected to a given class of overvoltages in actual service
conditions.
Uw: Standard rated withstand voltage: Standard value of the rated withstand voltage as specified in this standard. The rated
withstand voltage is value of the test voltage, applied in a standard withstand voltage test that proves that insulation complies
with one or more required withstand voltages. It is a rated value of the insulation of an equipment.
Kc: Co-ordination factor: Factor by which the value of the representative overvoltage must be multiplied in order to obtain the
value of the co-ordination withstand voltage.
Ks: Safety factor : Overall factor to be applied to the co-ordination withstand voltage, after the application of the atmospheric
correction factor (if required), to obtain the required withstand voltage, accounting for all other differences in dielectric strength
between the conditions in service during life time and those in the standard withstand voltage test
Page 11
TypethenameofTB
Table 2.1.4 Classes and shapes of overvoltages, Standard voltage shapes and
Standard withstand voltage tests (IEC 60071)
2.1.3.1 T EMPORARY
OVERVOLTAGES
(TOV)
TOV includes healthy phase overvoltages due to transmission line ground faults and load rejections. In the case of
sudden load rejection on a heavily loaded, long line, such as a UHV system, the overvoltage is about 1.3 - 1.5 p.u.
This TOV is required not only to cover the peak voltage in the system, but also to cover the overvoltage generated
during their operation. Therefore power frequency withstand test was verified in both long time range and short
time range, because the voltage stress is different from both range as described in Chapter 8.4.
Page 12
TypethenameofTB
OVERVOLTAGES
(S WITCHING
OVERVOLTAGES )
The duration of wave front is about a few-hundred microseconds, such as the overvoltage in opening / closing
transmission lines and ground fault. This switching overvoltage has much influence on the insulation design of
towers, thus switching overvoltage is particular important for UHV systems because of the saturation effects of the
air insulation distance on the switching impulse strength.
As shown in Figure 2.1.4, for the 1100 kV voltage level, the flashover voltage of air-insulated gaps for switching
overvoltage has a tendency to saturate. Therefore, extremely high tower is required for air insulation. On the
contrary, to reduce the construction cost of UHV system, switching overvoltages can be reduced by adopting circuit
breakers with closing and/or opening resistors and higher performance arrester. Figure 2.1.4 shows the relation
between air insulation distance and switching overvoltage, and Figure 2.1.5 shows the comparison between the
double-circuit tower design based on 2.0 p.u. and 1.7 p.u.
2000
1500
(550kV-Switching OV:
2.0pu)
1000
500
3m
0
0
6m
9m
10
Figure 2.1.4 Relation between air insulation and switching impulse withstand
voltages
ExpectinTra
g the
nsmission
ap plicaline
tion of 550k V
construtech
ctednology
this time
Page 13
TypethenameofTB
OVERVOLTAGES
(L IGHTNING
OVERVOLTAGES )
Lightning strokes terminating on UHV transmission lines can generate overvoltages of several MV depending on
the front-steepness of the overvoltage and the height of the tower. Shielding failures as well as back-flashovers
have to be taken into account.
Lightning overvoltage is the predominant factor for substation equipment design. Therefore, lightning overvoltages
in the UHV substation are highly suppressed for size reduction within a rational level by installing several higher
performance surge arresters at adequate locations.
2.1.3.4 V ERY
(VFTO)
The GIS disconnector, when switching a charging current, repeats restriking and generates VFTO, which can reach
up to approximately 3.0 p.u. At a UHV substation, lightning overvoltages are effectively suppressed by higher
performance surge arresters. Disconnector switching overvoltages are likely to exceed the lightning overvoltage if
no measures are taken to limit them. Therefore, the resistors can be a suppression measure for the VFTO.
Page 14
TypethenameofTB
TR
SW
LIWV SIWV
2250 kV 1800 kV
2400 kV 1800 kV
SW
LIWV SIWV
2400 kV 1800 kV
LIWV SIWV
TR
2250 kV 1800 kV
SW
2400 kV 1800 kV
(c) Jingmen substation (China)
LIWV SIWV
TR
1950 kV 1425 kV
SW
2250 kV 1550 kV
(d) Shin-Haruna testing site (Japan)
Figure 2.1.6 Substation designs and corresponding insulation levels (LIWV and
SIWV) of China and Japanese projects (TR: Transformer, SW: Switching equipment in
above captions)
Page 15
TypethenameofTB
Italy
1050
GIS
1800
2250
2250
Japan
1100
GIS
1620
1950
2250
China
1100
GIS, Hybrid-IS
1620
2250
2400
India
1200
AIS
1700
2250
2400
Figure 2.2.1 Dead tank-type higher performance surge arrester (Japan, China)
Page 16
TypethenameofTB
arresters (China)
arresters (India)
Italy
1050
MOSA
Closing &
opening R
(500 )
Japan
1100
MOSA
Closing &
opening R
(700 )
1.7
1.6 / 1.7
China
1100
India
1200
MOSA
Closing R
(600 )
MOSA
Closing R
(600 )
1.7
1.7
Page 17
TypethenameofTB
1.83
1.60
1.52
1.42
Italy
1050
GIS
110
Japan
1100
GIS
500
China
1100
GIS
Hybrid-IS
500
None
India
1200
AIS
None
Page 18
TypethenameofTB
R esistor
M ovable
contact
References
[1] Eiichi Zaima, C.Neumann, Insulation Coordination for UHV AC Systems based on Surge Arrester Application
(CIGRE C4.306), IEC-CIGRE Second International Symposium on Standards for Ultra High Voltage Transmission
[2] Guagfan Li, Jianbin Fan, The experience of UHV substation facilities in China, International Conference on
Development of 1200kV National Test Station
[3] IEC 60071-1 Ed. 8.1, Insulation co-ordination - Part 1: Definitions, principles and rules, 2011
[4] IEC 60038, Standard voltages, 2009
[5] CIGRE brochure No.400, Technical Requirements for Substation exceeding 800kV, 2010
Page 19
TypethenameofTB
Figure 3.1.1 Wiring schematic diagram for the China 1000 kV AC transmission test
and pilot project (single circuit) system
Figure 3.1.2 Schematic diagram for the China 1000 kV AC South Anhui-Shanghai
double-circuit tower arrangement transmission system
3.1.2 POWER FREQUENCY TEMPORARY OVERVOLTAGE (TOV) AND THE PARAMETER
SELECTION FOR METAL OXIDE ARRESTERS (MOA)
3.1.2.1 A MPLITUDE
AND DURATION OF
TOV
As for calculating the TOV, the following fault types shall be mainly taken into account.
Page 20
TypethenameofTB
As for the single-circuit transmission lines, two kinds of failures shall be usually taken into account, namely, the
load rejection under the normal operation and the load rejection in the case of the line single-phase grounding
failure.
As for the double-circuit tower arrangement transmission line, the double-circuit operation or one circuit out service
/the other circuit operation shall be considered and the failures causing double-circuit 6-phase load rejection shall
be taken into account.
Most of China 1000 kV lines are relatively long and the high voltage shunt reactors are generally installed in the
lines. The largest TOV may generally occur in the single-phase grounding fault case, in which the fault line circuit
breaker at one line side shall be three-phase tripped and the circuit breaker at the other line side shall not be
tripped. The above failure case may occur under the following two situations:
(1) During the normal operation process, the line single-phase is grounded and the single phase reclosing is not
successful the three-phase circuit breaker is tripped by relay protection.
(2) During the line live working process, the single-phase reclosing shall be required to withdrawal; at this time, the
single-phase grounding fault occurs and the line three-phase circuit breaker are also tripped.
The main measure to limit the power frequency overvoltage is to install the line high voltage shunt reactor.
The maximum TOV shall be allowed no more than 1.4 p.u. at the line side and the maximum TOV shall be allowed
no more than 1.3 p.u. at the bus side in China.
The TOV duration may play an important role in the choice of the arrester rated voltage and the equipment
insulation level. The relay protection mode is adopted in which the UHV line both side circuit breakers are tripped
synchronously, so as to shorten the duration of TOV and lower the energy absorbed by MOA. The maximum trip
delay for the both side circuit breakers shall generally be controlled within 0.2 seconds; the TOV duration shall be
no more than 0.5 seconds even if it is considered that the one side circuit breaker is failure to trip and the tripping
shall be carried out by the back standby circuit breaker.
3.1.2.2 MOA
PARAMETER SELECTION
In the past, the traditional MOA rated voltage selection principle was UnTOV; whereas, the traditional MOA rated
voltage selection principle has been broken through in China UHV project, namely, that Un is allowed to be less
than TOV. The MOA rated voltage Un of the UHV system was selected as 828 kV, which is equivalent to 1.3 p.u.
and is less than the maximum TOV (1.4 p.u.) of the UHV lines. Because the MOA is with the good short time power
frequency overvoltage withstand capacity, the Un being lower than TOV for short time shall be permitted. According
to the test data from the China MOA manufacturers, the TOV duration for MOA withstanding 1.4 p.u. is 10 seconds.
The main electrical parameters for MOA (Un=828 kV) are listed in Table 3.1.1.
The energy absorption allowable value is 40 MJ. The calculation results may show that the MOA actual maximum
absorption energy shall be less than 10 MJ while the maximum TOV duration is 0.5 seconds and under the
switching overvoltage cased by two times closing operation. Therefore, the Un is selected as 828 kV and there shall
be relatively great margin for MOA absorption energy.
The lower MOA rated voltage causes the MOA residual voltage to be lowered so as to lower the substation
overvoltage amplitude and the requirement for the equipment insulation level; as plays some certain role for
lowering the line overvoltage.
Page 21
TypethenameofTB
System
nominal
voltage
Installation
location
Rated
voltage
(RMS)
Continuous
operation
voltage
(RMS)
Switching impulse
residual voltage under
30/60 s and 2 kA
Lightning impulse
residual voltage under
8/20 s and 20 kA
1000
828
638
1460
1620
3.1.3.1 C LOSING
The closing and reclosing line overvoltage may play the control role to the insulation design of China 1000 kV lines.
The main measures to limit the closing and reclosing line overvoltage are that the closing resistor is installed on
line circuit breakers. The closing resistor is taken as 600 and the closing resistor pre-insert time is 9.51.5 ms.
The maximum phase-to-ground statistical switching overvoltage along the line shall be no more than 1.7 p.u. for
China 1000 kV lines; the substation maximum phase-to-ground statistical switching overvoltage shall be no more
than 1.6 p.u. and the maximum phase-to-phase statistical switching overvoltage shall be no more than 2.9 p.u. The
maximum statistical switching overvoltage at the substation bus side shall be no more than 1.55 p.u.
The front time of the 1000 kV line closing and reclosing line overvoltage shall generally be above 1000-3000 s,
which may greatly influence the air clearance selection of transmission line tower.
3.1.3.2 G ROUND
FAULT OVERVOLTAGE
The single-phase grounding fault type shall be taken into account for the calculation to the ground fault overvoltage
of the China UHV systems; the overvoltage amplitude is relatively low; the maximum 2 % overvoltage shall be
lower than 1.51 p.u.. As for the China UHV lines, the ground fault overvoltage shall not be the control factor to
determine the line insulation level.
3.1.3.3 C LEARING
FAULT OVERVOLTAGE
This clearing fault overvoltage means the overvoltage occurred in the adjacent non-fault lines while the short-circuit
fault in the fault line is cleared up.
The fault type may significantly influence the overvoltage amplitude. The clearing single-phase grounding fault
overvoltage amplitude shall be within the allowable range. The clearing 2-phase or 3-phase grounding fault
overvoltage may be relatively high and the overvoltages in some lines may exceed the allowable values. The
opening resistor may be installed in the line circuit breaker so as to lower the clearing fault overvoltage.
The following situations shall be taken into account:
(1) The opening resistor may absorb great energy and the operation may be complicated so that not only the cost
of the circuit breaker may be increased but also the probability of the circuit breaker failure may be increased.
(2) The probability of the 2-phase or 3-phase grounding fault shall be very low.
Page 22
TypethenameofTB
(3) The maximum overvoltage may occur on lines rather than in substations. It may cause the line insulation
flashover; however, the substation equipment may not generally be damaged by line overvoltage.
As for the JindongnanNanyangJingmen UHV transmission line, the opening resistor is not necessary to be
installed.
3.1.3.4 E NERGIZING
As shown in the research and field tests in China, the 500 kV circuit breaker is not equipped with the closing
resistor for the energizing unloaded UHV transformer at the 500 kV side; there shall be no any relatively high
resonance overvoltage. The inrush current and the overvoltage are within the allowable range.
The possibility of the resonance overvoltage and the inrush current from the energizing unloaded UHV transformer
at the 1000 kV side may be greater than that at the 500 kV side. The closing resistor may be adopted so as to be
beneficial to lowering the resonance overvoltage and the inrush current; however, the closing resistor may not be
valid for all system construction and operation modes; moreover, the closing resistor may increase the equipment
cost and cause the switching operation mechanism to be complicated as well as lower the reliability.
Under the normal circumstances, the energizing unloaded UHV transformer at the 500 kV side shall be provided.
Page 23
TypethenameofTB
Figure 3.1.3 Substation connection modes taken into account for the lightning
invading overvoltage calculation
Figure 3.1.4 Ground wires have been adopted in the entrance line section of the
UHV substation
The substation maximum lightning overvoltage may be caused by the lightning shielding failure invading wave in
the entrance line section of substation; two measures, namely, decreasing the maximum lightning shielding failure
current in the entrance line section [the entrance line ground wire protection angle is decreased to -4 and three
ground wires have been adopted in the entrance line section (as shown in Figure 3.1.4). and optimizing the arrester
layout, have been taken in China UHV AC transmission test and pilot project so as to decrease the lightning
shielding failure invading overvoltage. Finally the scheme with the small quantity of arresters has been adopted: 1
group of MOAs is installed in each circuit entrance location; 1 group of MOAs is installed in each bus section; 1
group of MOAs is installed beside the transformer. The overvoltage values may be different for various substations.
The typical values of the maximum lightning overvoltage of the equipment may be listed as follows: 2040 kV for
Page 24
TypethenameofTB
GIS, 1854 kV for the shunt reactance and 1796 kV for the transformers. The lightning impulse withstand rated
voltage of the transformer and the shunt reactance is 2250 kV and lightning impulse withstand rated voltage of
other devices is 2400 kV in China. The allowable values of the equipment lightning impulse insulation levels shall
be more than the maximum lightning invading overvoltage, which shall be meet the requirements of the internal
insulation margin (15 %) and the external insulation margin (5 %). As for the single-line mode, the internal
insulation margin may be lowered to 10 % because its occurrence probability is very small.
As for the calculation of the substation lightning invading overvoltage, the interval statistics method may be used
besides the deterministic method; moreover, the substation lightning MTBF (mean time between failures) shall be
required to be more than 1500 years.
Page 25
TypethenameofTB
3.1.7.1 P RINCIPLES
OF INSULATION COORDINATION
By combining the research results on the China UHV overvoltage and internal and external insulation
characteristics and according to the safety and economy principals, the insulation level may be determined.
As for the China insulation coordination principle, some certain insulation safety margin has been required to
reserve; and the internal and external insulation safety margins are 15 % and 5 %, respectively.
As for the air clearance of UHV AC line and substation, discharge voltage test at power frequency voltage, lightning
and switching impulse voltage have been relatively systematically and comprehensively carried out with the 1:1
tested objects in China and a series of discharge voltage curves have been obtained; as may provide the
reasonable basis for the insulation coordination.
3.1.7.2 S UBSTATION
The insulation levels of the main equipments in China UHV substation are shown in Table 3.1.2.
On the whole, the insulation levels of the main equipments in China UHV substation may be lower than those of the
Russia UHV equipment, but they may be higher than those of the Japan UHV equipment. It is determined by
combining the China UHV overvoltage level and the corresponding equipment manufacturing experiences; and it
shall be suitable for the Chinese situation.
(Unit: kV)
Power-frequency shortduration withstand voltage
Lightning impulse
withstand voltage
2250
(chopped impulse:
2400)
Switching impulse
withstand voltage
1800
1100 (5 min)
2400
1800
1100 (1 min)
2550
1800
1100 (1 min)
2400
1800
1300 (5 min)
2400
(chopped impulse:
2760)
1950
1200 (5 min)
Bushing (GIS)
2400
1800
1100 (1 min)
2400+900
1675+900
1100+635 (1 min)
Equipment
Transformer and reactor
GIS (circuit breaker and
disconnecting switch)
Post insulator and disconnector
(open type)
Table 3.1.2 The rated insulation withstand voltage for 1000 kV equipment
Page 26
TypethenameofTB
3.1.7.2.1 T RANSFORMER
INSULATION LEVEL
The insulation level of China UHV transformers may be shown in Table 3.1.3.
(Unit: kV)
Country
China
Lightning impulse
withstand voltage
2250
Switching impulse
withstand voltage
1800
Power-frequency
withstand voltage
1100 (5 min)
3.1.7.2.2 L ONGITUDINAL
DISCONNECTOR
The longitudinal insulation lightning impulse testing voltage of China UHV circuit breaker and disconnector should
be 2250 + 900 (kV), in which 900 kV is the peak value of opposite polarity working voltage,1100 2 / 3 =Un-m=
900 kV.
The amplitude of the opposite polarity power frequency component recommended in IEC60071-1 is 0.7Um 2 /
3 namely the amplitude of the working voltage should be multiplied with the coefficient 0.7.
In China national standard Insulation Coordination for High-voltage Transmission and Distribution Equipment
(GB311.1), the coefficient is stipulated among 0.7-1.0, which should also be determined together with the electric
power companies and the manufacturers.
If this coefficient is 0.7, in 1 cycle of the working voltage (the 360 phase), there is about 1/4 cycles not being
included. Therefore, the guarantee probability is 0.75.
As a result of the importance of the UHV equipments, China advocate that the amplitude of the opposite polarity
power frequency component is 11100 2 / 3 =900 kV while the longitudinal insulation lightning impulse tests for
the UHV circuit breaker and disconnector are carried out.
3.1.7.3 S UBSTATION
The A values of the substation minimum air clearance shall be divided into three categories: A1' value, A1" value
and A2 values.
Page 27
TypethenameofTB
The A1' value is the minimum electrical distance of the substation conductor to the frame; the A1" value is the
minimum electrical distance of the substation equipment to the frame; and A2 value is the minimum electrical
distance between the phases in the substation.
As for the areas whose altitude being no more than 1000 m, the minimum air clearance for the 1000 kV substation
are listed in Table 3.1.4.
(Unit: kV)
The action voltage type
A1 value
A1
Power frequency
Switching impulse
Lightning impulse
A2 value
A1
4.2
6.8
7.5
5.0
6.8
10.1 (ring - ring)
9.2 (four sub-conductors - four subconductors)
11.3 (tubular bus - tubular bus)
5.5
Table 3.1.4 The minimum air clearance for the 1000 kV substation
3.1.7.4 L INE
INSULATION LEVEL
The contaminated insulator withstand voltage method or the specific creepage distance method may be used to
determine the insulator number.
With comprehensively considering the insulation margin and the simplified design and other factors, the insulator
configuration for the common lines in China UHV AC transmission test and pilot project may be listed as follows:
(1) he suspension string: 54 double-shed disk-type insulators (for 300 kN) are used in the II grade polluted area;
the composite insulator whose structural height is 9750 mm and creepage distance is 30300 mm are used in the III
and IV grade polluted areas.
(2) he tension string: the 44, 54 and 60 disk-type insulators (550 kN and creepage distance is 700 mm) are used in
II, III and IV grade polluted areas, respectively.
The minimum air clearance values for China 1000 kV lines are listed in Table 3.1.5.
As for the cup type-towers or cat-head type towers in the single-circuit line, the middle phase minimum air
clearance is controlled by the switching overvoltage; on the other hand, the out phase minimum air clearance is
controlled by the power frequency voltage under the weather condition of gale rather than the lightning overvoltage.
Therefore, there shall be no any regulation on the minimum air clearance under the lightning impulse.
However, the towers of the UHV double-circuit tower arrangement transmission line may be very high so as to
increase the attracting lightning area and weaken the conductor shielding effect from the earth; thus, the line may
easily stricken by the lightning. Because too large lightning current is not able to cause the conductor shielding
failure, so that the line maximum lightning shielding failure current amplitude is limited, the insulation level of the
UHV double-circuit tower arrangement transmission line may be enhanced appropriately so as to effectively lower
the lightning shielding failures. The main measure to enhance the line insulation level is to increase the distance
from the conductor to its lower cross arm or the tower body; especially the former may significantly enhance the
line insulation level. Therefore, the minimum clearance distance from the line lightning impulse voltage to the tower
has been regulated for 1000 kV double-circuit tower arrangement transmission line.
While the discharge voltage within the air clearance under the switching impulse voltage is determined, 1000 kV
line real type towers are tested in order to take fully account of the influence to the gap discharge voltage from the
tower width; The wave front time of the switching impulse test voltage is 1000 s.
Page 28
TypethenameofTB
The influence to the discharge voltage from the parallel connected number of gaps shall be taken into account in
the insulation coordination and the entire line insulation flashover rate under the switching overvoltage has been
calculated; the variation of the calculated switching overvoltage values at various locations along the line and the
actual distribution shall be taken into account. The entire line insulation flashover rate under the switching
overvoltage shall be no larger than 0.01 times/year.
(Unit: m)
The action voltage type
Page 29
TypethenameofTB
Figure 3.2.1 Power Map of Western Region showing proposed 1200kV corridor in
India
The connectivity diagram for the 1200kV corridor with 765kV system is shown in Figure 3.2.2.
Page 30
TypethenameofTB
1.7 p.u.
line energization/reenergization/
de-energization, line fault clearing, Reactor
Switching, Transformer Energizing and tripping
on external fault as well as on no load
Yes
Reclosing scheme
Single-phase reclosing
1
2
3
4
5
0.5
1.0
2.0
10.0
20.0
1380
1440
1500
1600
1700
Page 31
TypethenameofTB
850 kV
Switching impulse
1500 kV at 2 kA
Lightning impulse
1700 kV at 20 kA
3.2.4.1 D ETERMINATION
PROCESS FOR
SIWV
AND
LIWV
The location of Surge Arresters is very critical in order to contain voltage rise due to Switching Surges as well as
Lightning Surges. For adequate protection of Equipments against Lightening surges, it was decided to place Surge
Arresters at Line Entrance and near Transformers/ Reactors. In addition, Surge Arresters in the Bus may also be
considered necessary.
The values of SIWV and LIWV are obtained considering adequate margin over the values of SIPL and LIPL as
obtained from V-I characteristics of the Surge Arrestor. The SIWV for 1200kV equipments has been considered as
1800kV with about 20% protective margin. The LIWV has about 35% protective margin for 1200kV equipments and
about 25% protective margin for 1200kV transformer and reactors. The values of LIWV for 1200kV equipments and
transformers/reactors have been considered as 2400kV and 2250kV respectively.
The following tables show us the LIWV and SIWV for 1200 kV equipments.
Substation type and MOSA layout
Substation Type (GIS, MTS, AIS)
MOSA layout
(unit)
Line entrance
Yes
Busbar
Transformer
Yes
Lightning overvoltage
1700 kV
Calculation method
Safety factor
Transformer
2250 kV
Other equipment
2400 kV
LIWV (kV)
Page 32
TypethenameofTB
Switching
overvoltage
Representative Value
(Max and/or 2 % value)
(kV, p.u.)
Transformer
Other equipment
Calculation method
Safety factor
Atmospheric correction factor, altitude (m)
SIWV (kV)
(phase-to-ground)
SIWV (kV)
(phase-to-phase)
Transformer
1800 kV
Other equipment
1800 kV
Transformer
2970 kV
Other equipment
2970 kV
12.3 m (rod-cond.)
For determination of temporary over voltages, preliminary studies have been carried out with single-line-to-ground
fault followed by three phase interruption at receiving end. The studies have been carried out with different source
strengths and reactive compensation as mentioned above. The study result showed a value of TOV as 1.33 p.u.
with 10000 MVA short circuit level at the sending end. For the purpose of insulation coordination study the value of
TOV is taken as 1.4 p.u..
To determine the discharge capability of the surge arrester, the most severe condition of operation has been
considered. The series of events start with single line to ground fault followed by opening of local and remote end
CB of the faulted phase. The single phase auto recloser becomes effective after a dead-time of 1000 ms. However
the reclosing becomes unsuccessful and the breaker could not be opened due to struck breaker condition. The
sequence of events is shown at Figure 3.2.3.
Steady
T=0ms
T=100ms T=120ms
Dead Time
1000 ms
SLG fault
Opening of
local end
CB of
faulted
Opening of
remote end
CB of
faulted
Stuck
breaker
Closing
of local
end CB
through
SPAR
Closing
of
remote
end CB
through
Opening of
CB due to
permanent
fault
LBB
Operates
and clears
the fault
Page 33
TypethenameofTB
The discharge capability required for surge Arresters for above conditions is given below:
Total energy = 2Long discharge (IEC Class-5) + TOV + Margins
= 25 MJ + 35 MJ + 10 MJ = 55 MJ
The information for TOV and Energy absorption by surge arrester is tabulated below:
Overvoltage on healthy phase in case of ground fault
1.4
Calculation method
EMTP/ATP
Transformer
1200 kV
Other equipment
1200 kV
Analysis program
EMTP/ATP
Page 34
TypethenameofTB
400kV line
400kV line
Satna line
400kV Bina Bus
1200kV line
1200/400kV Transformer
1200/400kV Transformer
To 400kV Satna line
Figure 3.2.4 Power flow from Satna to Bina will be diverted via 1200kV test station
1200kV Equipment has been indigenously developed by Indian Manufacturers. 1200kV equipment will be field
tested under full operating voltage for fine tuning and optimization of system parameters. Further, 1200kV Test
station will help to learn safety, Quality & operational requirement for UHV AC Technology.
Page 35
TypethenameofTB
Sea of Japan
Tokyo
o
Pacific Ocean
Voltage
550kVMOSA
LIWV (Substation)
Switching Overvoltage
(Transmission line)
TOV
GIS:
2250kV
Transformer: 1950kV
1.61.7pu
1.5pu
1100kVMOSA
1620kV
(1.8pu)
Higher Voltage
Stress
mA 10kA20kA
Current
(1 p.u. =1100 kV 2 / 3 )
Page 36
TypethenameofTB
3.3.3.1 S WITCHING
The predominant factors that determine the size of towers are the switching overvoltages generated in the system
that include closing, opening, and ground fault overvoltages. The level of these switching overvoltages must be
mitigated, limiting them to a value as low as possible. Among these overvoltages, ground fault overvoltages are
generated independently of circuit breaker switching, and generally, there are no effective means of controlling
them, except those near substations that can be reduced by surge arresters (Figure 3.3.4). In contrast, closing and
opening overvoltages can be controlled effectively by the insertion of closing and opening resistors. TEPCO
decided to control these overvoltages below the ground fault overvoltage level by employing the closing/opening
resistor (700 ).
Overvoltage [ pu]
1.8
1.5
UN
Gunma
49km
Higashi
Yamanashi
Nishi
Higashi
Gunma Gunma
138km
44km
Minami
Iwaki
194km
[3], [5]
The evaluation of switching overvoltages is important for the insulated design of a transmission line. In an analysis
of ground fault overvoltages, the maximum overvoltage levels were obtained, using locations and phases of faults,
voltage phases at the time of faults, ground fault resistance, etc., as parameters. In an analysis of closing
overvoltages, considering that the overvoltage level depends on the closing timing lags of the main and resistance
contacts of 3-phase circuit breakers, maximum values and overvoltage distributions were obtained from a Monte
Carlo simulation using closing timings of 6 contacts as parameters. System conditions chosen are mainly the 1circuit trial closing from the 2-citcuit unenergized state of the transmission line, which is expected to cause high
overvoltages, or the reclosing of the transmission line. Opening overvoltage analysis mainly focused on
interruptions of ground fault currents, which are expected to cause high overvoltages. The values of ground fault
current were taken as currents at the maximum level in each phase of system growth, ultimately ranging up to 50
kA, which is the interrupting current of the circuit breakers.
From the results of overvoltage analysis, the insulation design levels for switching overvoltage (basic for air
clearance design) were determined as 1.6 p.u. (2.6 p.u. between phases) for the shorter north-south route of the
UHV system (about 140 km) and 1.7 p.u. (2.8 p.u. between phases) for the longer east-west route (about 240 km).
Page 37
TypethenameofTB
Expecting
Transmission
the application
line of 550kV
constructed
technology
this time
Transmission line
constructed this time
Reclosing scheme
826 kV
Switching impulse
Lightning impulse
1620 kV at 20 kA
Page 38
TypethenameofTB
3.3.5.1 D ETERMINATION
3.3.5.1.1 L IGHTNING
PROCESS FOR
LIWV
OVERVOLTAGE STUDY
To analyze overvoltages accurately generated in the system, it is essential to determine a suitable system model.
Adding new knowledge to already established techniques, the study here prepared models that are as accurate as
possible. The main ones are as follows:
Lightning strike current:
200kA, 1/70 s ramp waves for tower back flashover
30kA, 1/70 s ramp waves for direct lightning to power line
Both lightings occur at the towers adjacent to substations.
Transmission line:
Eight (8)-phase frequency dependent distribution constant (Semlyen) model.
Tower:
Multistory transmission tower model shown in Figure 3.3.7. Constants, for example, Zt1=120 , were determined
by carrying out lightning impulse response characteristics tests on actual UHV towers and reflecting them [9].
Back flashovers between horns:
Leader model, time dependent inductance as shown in Figure 3.3.8, with leader strokes taken into account [10]
Substation circuits:
All buses are simulated by 3-phase separated GIS single-phase distribution constant circuits
Page 39
TypethenameofTB
To analyze the estimated maximum overvoltage level, the following three types of severe operating conditions
involving high overvoltages generated at the substation equipment were determined, as shown in Figure 3.3.9,
Circuit condition I: High overvoltages are expected at the transmission line entrance.
Circuit condition II: High overvoltages are expected at buses.
Circuit condition III: High overvoltages are expected at the terminal of the transformer.
The validity of the insulation level determined from this maximum was proved by overvoltage distributions analyzed
as parameters in the unenergized equipment and lightning current during normal operation.
Figure 3.3.9 Service operating conditions of substation circuit for lightning surge
analysis
Page 40
TypethenameofTB
3.3.5.1.2 L IGHTNING
OVERVOLTAGE STUDY
On the basis of analyzed lightning overvoltage under severe circuit conditions, appropriate LIWV levels were
evaluated for different installation schemes of surge arresters. Table 3.3.3 shows that case 6 is economically most
favorable, which has two surge arresters per circuit at the line entrance, two per quarter bus, and one per bank of
transformers installed and employs the following LIWV levels.
- 1950 kV for transformers
- 2250 kV for GIS
LIWV(kV)
1950
Required withstand
2898
voltage (kV)
LIWV(kV)
Cost (100 for case 6)
2900
102
1
1
1
1
2
1
2
0
1
2
1
1
2
2
1
1943
1895
1943
1938
1896
1950
1950
1950
1950
1950
2854
2703
2628
2506
2208
2900
105
2900
109
2700
103
2550
103
2250
100
Page 41
TypethenameofTB
PROCESS FOR
SIWV
OVERVOLTAGE STUDY
In an analysis of ground fault overvoltages, the maximum overvoltage levels were obtained, using locations and
phases and phases of faults, voltage phases at the time of faults, ground fault resistance, etc., as parameters.
In an analysis of closing overvoltages, considering that the overvoltage level depends on the closing timing lags of
the main and resistance contacts of 3-phase circuit breakers, maximum values and overvoltage distributions were
obtained from a Monte Carlo simulation using closing timings of 6 contacts as parameters. System conditions
chosen are mainly the 1-circuit trial closing from the 2-circuit unenergized state of the transmission line, which is
expected to cause high overvoltages, or the reclosing of the transmission line.
Opening overvoltage analysis mainly focused on interruptions of ground fault currents, which are expected to
cause high overvoltages. The values of ground fault current were taken at the maximum level in each phase of
system growth, ultimately ranging up to 50 kA, which is the interrupting current of the circuit breakers. The
closing/opening resistors of 700 were employed to suppress (limit to a maximum value) switching overvoltage as
mentioned in section 3.4.3.
Page 42
TypethenameofTB
The switching overvoltages applied to the substation equipment, obtained under these conditions are shown in
Table 3.3.5. Here, the maximum overvoltage were generated under 1-circuit closing during 2-circuit unenergized
condition. For frequent overvoltages, the maximum value of ground fault and opening overvoltages at the end of
the substation was employed. Closing overvoltages in closing other than trial charges are lower.
Maximum overvoltage
1309 kV (1.46 p.u.)
1400 kV (1.56 p.u.)
Transformer
GIS
Frequent overvoltage
1250 kV (1.39 p.u.)
1250 kV (1.39 p.u.)
OF
SIWV
From the analysis results and equipment insulation characteristics, SIWV were determined as follows.
- 1425 kV for transformers
- 1550 kV for GIS
It is estimated that these withstand voltage levels are also economically acceptable, because insulation component
is basically determined by the withstand voltage values for lightning impulse or power frequency test voltages
described later.
Representative Value
(Max. and / or 2 %
Value)
(kV, p.u.)
Switching
overvoltage
Transformer
Other
equipment
EMTP analysis
Safety factor
1.06, H=1000 m
1425 kV
1550 kV
10.5 - 11.5 m
The GIS disconnector, when switching a charging current, repeats restriking and generates a very fast transient
overvoltage (VFTO). At a UHV substation, lightning overvoltages are effectively suppressed by higher performance
surge arresters. Disconnector overvoltages are likely to exceed the lightning overvoltage if no measures are taken
to control them. Therefore, the resistor insertion (500 ) is employed to suppress the VFTO.
Maximum
VFTO (p.u.)
Without
resistor
With resistor
()
Transformer
Other equipment
Transformer
Other equipment
Page 43
TypethenameofTB
3.3.5.4 TOV
[6]
Since a UHV system is looped with the existing EHV grid, load rejection of the UHV is exceedingly rare. However,
because it will generate high TOV with high energy, a scenario is taken into account, in which a UHV system with
several generators is isolated from the EHV grid due to double-circuit faults and successive malfunction of the
special protection scheme for tripping generators. A TOV analysis at load rejections, using the EMTP, as with
switching overvoltages analysis, simulated systems above 550 kV of all TEPCO systems, including UHV systems,
and also simulated in detail the AVR and governor characteristics of generators, which are expected to have a
large effect on voltage rises. The maximum value of TOV at load rejections in this case is about 1.5 p.u. and its
waveform is trapezoid as shown in Figure 3.3.10. The energy absorption by a surge arrester is so high that the
energy rating of an arrester is specified as 55 MJ and above.
The overvoltage protection relay system to open a no-load UHV transmission line concerned will be applied to
protect surge arresters if necessary, so that their energy absorption does not exceed the energy limit.
Overvoltage
The maximum TOV values for ground faults are about 1.15 p.u. on transmission lines and about 1.1 p.u. at
substations.
1.15 p.u.
Calculation method
EMTP analysis
3.3.5.5 P OWER
UHV power frequency test voltages were decided based on partial discharge tests (Long-duration tests) as shown
in Figure 3.3.11. This test is not only aimed to confirm dielectric strengths, but also to verify strengths by inspecting
whether or not partial discharges are generated as precursors of dielectric breakdowns throughout the test. The
test voltages are the combination of a short-duration section for verifying dielectric strengths to temporary
overvoltages at one-line ground faults and load rejections, and a long-duration section for assuring long-term
strengths to operating voltages.
Page 44
TypethenameofTB
Power frequency test voltages for UHV equipment were determined as follows.
[Transformer]
1.5 p.u. (1 h) +
[GIS]
1.5 p.u. (30 min) + 3 p.u. (1 min) + 1.5 p.u. (30 min)
V2
V1
t1
Longduration
section
V3
t2
t3
LongShortduration duration
section section
Transformer
Assumed overvoltage
condition for power
frequency test
Other
equipment
Transformer
Pattern
Other
equipment
Method of determination
* Effect of trapezoid ACTOV waveforms was reflected in duration from insulation characteristics.
3.3.5.6 C OMPARISON
TEPCO calculates various overvoltages by EMTP. The comparison between switching overvoltage analysis by
EMTP and actual measurement is shown in Figure 3.3.12. Computed values agree with measured ones in
TEPCOs UHV project very well. Therefore, the sophisticated design can be achieved by the precise overvoltage
analysis such as EMTP.
Page 45
TypethenameofTB
Calculation waveforms
Measurement waveforms
Page 46
TypethenameofTB
This has been established as a consequence of an international treaty celebrated between both countries.
However, Brazil has agreed, by contract, to buy the whole of 50 Hz generated energy, except that consumed by
Paraguay, which means that 70 % to 80 % of the 50 Hz energy, in average, was consumed by the Brazilians in the
last years.
FURNAS Centrais Eltricas was responsible for developing the transmission solution in order to integrate Itaipu
powerplant generation into the Brazilian national transmission network, performing the planning studies, equipment
specification, factory and laboratory tests, erection on site, commissioning tests (factory and on-site acceptance
tests), operation and maintenance of the assets, amongst other tasks. Thus, FURNAS started the operation of the
800 kV AC and 600 kV DC systems in 1982 and 1984, respectively, integrating the Itaipu powerplant generation
into the Brazilian National Transmission Network, being, then, the detainer utility of the Brazilian know-how on
transmission issues of such voltage levels, completing now 28 years of successful operating experience.
The building up of the Brazilian knowledge and experience on dealing with 800 kV transmission issues have
started with these investigation studies, in order to define such hybrid transmission system of 900 km of length and
consisting of 3 800 kV AC parallel lines and 2 bi-pole 600 kV HVDC link. This hybrid AC/DC transmission
system planned, shown in Figure 3.4.2, is one of the most important in the occidental world, due to its nominal
voltage levels, rated power capacity and importance for the Brazilian electric industry. Its importance is not only
due to the usage of some of the highest voltage levels in commercial operation worldwide, but also because of the
high capacity of the transmitted power (rated to 12600 MW) for a long distance (about 900 km).
It is important to point out that there were no existing IEC Standards at that time for such a huge hybrid
transmission system, especially for the HVDC link. In this sense, some aspects and faced problems, from planning
and operating experience points of view, concerning the hybrid HVAC and HVDC features and equipment, had to
be deeply investigated as special requirements in the equipment specifications since the beginning of planning
works.
HVDC +/-600kV
switchyard
800kVac
switchyard
Figure 3.4.1 Foz do Iguau substation, belonging to the Itaipu transmission system,
contains both the HVAC switchyard of the three 800 kV circuits and the HVDC
converter substation equipment/switchyard (HVDC link rectifier side)
Page 47
TypethenameofTB
The three circuits of 800 kV AC, each one of about 895 km long, are able to transmit to the Brazilian interconnected
network the rated generated power from the generating units in Itaipu powerplant operating with the nominal
frequency of 60 Hz. The planning criterion of such transmission system considered the possibility of dispatching
such amount of power even in the n-1 elements outage configuration. Also, this 800 kV AC system is responsible
for the energy interchange between the southern and southeastern geographic regions of the brazilian
interconnected national grid as well (refer to Figure 3.4.2).
Figure 3.4.2 More details about FURNAS transmission system for the integration of
Itaipu power plant within the Brazilian electrical network
The electrical requirements for the 800 kV AC equipment has been defined in the late 70s and early 80s. At that
time, the Brazilian network consisted of several sub-networks weakly connected to each other. Although each subnetwork could be considered as solidly grounded, the network as a whole could not. Besides, the 800 kV Brazilian
network would not be a meshed network but a radial one consisting of 3 parallel circuits. Thus, the planning
engineers at that time decided to adopt conservative insulation coordination requirements for the equipment of
such a new and unknown high voltage level in Brazil, derived from system simulations under severe operation and
emergency conditions.
The shunt reactors (fixed at line-ends and switchable at busbars and transformer tertiaries), series capacitors and
synchronous condensers provide, respectively, transient switching overvoltages mitigation (fixed SR at line-ends),
voltage regulation during steady state operating conditions (switchable SR at busbars and transformer tertiaries)
and system dynamic stability (series capacitors and synchronous condensers), therefore, contributing to the
insulation coordination matter from the switching transient overvoltage issue point of view.
Regarding the reliability of the overall transmission system, in terms of potential damages due to strong winds and
heavy thunderstorms, in order to keep the criterion of n-1 element outage as much as possible in real life, the rightof-way of the set of 5 transmission lines (3 AC lines and 2 HVDC bipoles) were defined with a security distance
among them as shown in the Figure 3.4.3 below.
Page 48
TypethenameofTB
Figure 3.4.3 Geographic arrangement of the set of 5 lines that belong to the
transmission system for the integration of Itaipu power plant within the Brazilian
electrical network (the distances are in kilometres)
3.4.4 800 KV SHUNT REACTOR APPLICATION PHILOSOPHY
In FURNAS, as well as in other utilities, the SR are used in power system transmission OH lines for 2 main
purposes: (a) switchable SR at busbars or tertiary-windings of power transformers for steady-state voltage control
issues; (b) fixed SR at the OH line ends to deal with switching transient overvoltage requirements, related to the
line closing and/or opening operation (including fast auto-reclosing issues). Therefore, the SR installed at this 800
kV transmission system follows the same philosophy, giving 70 % of shunt compensation for line charging. In the
next schematic diagram one can see the existing SR application philosophy installed in such transmission system.
For transient switching overvoltage mitigation purposes, the fixed SR have the aid of the arresters installed at the
line ends, having such arresters, along the last 30 years, two very distinct functions on such duty, depending on the
evolution of technology and applied materials by the arresters manufacturers: (a) arresters with active gaps and
non-linear resistances made of silicon carbonate (SiC); and (b) gapless metal oxide arresters (MOSA).
When it was used only arresters with active gaps and SiC non-linear resistances, the very first arrester that sparked
its active gaps was the only responsible for draining all the surge energy, as well as for controlling the switching
overvoltage in conjunction with the fixed SR at the line end.
After the introduction of devices such as gapless metal oxide arresters, the surge energy draining could be divided
amongst all the MOSA, depending on the VxI curve of the used non-linear material (Zinc Oxide - ZnO).
Page 49
TypethenameofTB
Equivalent Source 1
Equivalent Source 2
Zeq 2
Zeq 1
Line-length = 300 km
Figure 3.4.4 Fixed Shunt Reactor application philosophy for transient switching
overvoltages mitigation in the OH line
Regarding the gapless MOSA generation, those installed to protect the SR against lightning impulses could also be
used in conjugation with the MOSA installed at the OH line ends. Thus, using such set of surge arresters (i.e., not
only the ones installed at the OH line ends, but also those provided to protect the fixed SR against lightning surges),
the control of the switching transient overvoltages and the draining of switching surge energy as well, could be
done in a much more efficient/reliable way, helping with the mitigation/control of switching transient overvoltages as
mentioned, in comparison with the old SiC SA used before.
Page 50
TypethenameofTB
technology showed to have a very good performance for transient switching overvoltages mitigation, avoiding
completely the possibilities of resignations during the switching off procedures, since the convenient minimum
arcing time could be properly controlled for that purpose. The extensive field tests/measurements performed by the
company proved such expected performance during several switching off operations of the existing 800 kV busbar
SR. On the other hand, the larger arcing times, provided by the CB equipped with controlled switching device,
would lead to a higher and more frequent switching phenomenon defined as current chopping overvoltages.
The related current chopping switching transient can lead to a transient recovery voltage (TRV) across the circuitbreaker chambers that should be investigated, in principle, as a big concern, in order to avoid undesirable
damages to the SR, or even to the CB. This transient phenomenon is related to the fact that, during the SR
switching off, the nominal current is forced to zero suddenly and prior to the natural power-frequency-zero-crossing,
within the CB chambers, leading to such a phenomenon known as current chopping. Thus, the current chopping,
meaning a sudden change to zero in 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 capacitances, causing,
thus, sudden changes in the load-side transient voltage. This energy-transfer-phenomenon is known as chopping
overvoltage. The lower the stray capacitances are the higher the chopping overvoltages will be during the
occurrence of current chopping.
However, the field experience in FURNAS showed it was much more reliable and secure to have a complete
control over the possibility of no-reignition during the SR switching off, by means of using CB equipped with
controlled switching, compared to the increase in current chopping overvoltages. In the case of 800 kV switchable
SR, the increase of chopping overvoltages, derived from the increase of minimum arcing time (given by the CB with
controlled switching), is very small, due to the fact that the stray capacitances of the SR windings are quite
significant.
Page 51
TypethenameofTB
Page 52
TypethenameofTB
Equivalent Source 1
Equivalent Source 2
Line flow
Zeq 2
Zeq 1
800 kV rms
Line-length
Page 53
TypethenameofTB
SEGMENT 1
SEGMENT 2
Page 54
TypethenameofTB
Platform failure
Current Unbalance (only Alarm)
Current Unbalance (activating SC by-pass)
Optics fiber failure
By-pass breaker failure
SC bank overload
Air-Compressed Central failure (the gap operates with air-compressed-blast, provided by a dedicated central to
the SC)
Regarding the protection coordination of parallel lines with series-capacitor compensation, this issue is also of big
concern and has to do with the distribution of short-circuit currents amongst the parallel circuits, which may cause
undesirable protection operation of such parallel (and not faulted) lines. This may happen due to overvoltage
protection or SC by-pass misoperation.
There is still the issue, also of big concern, related to the sub-synchronous resonance, that may happen when
there is thermal turbines sited electrically near the series-compensated lines. Depending on the compensation
degree provided by the SC, they, in conjunction with the line inductances, may cause a natural oscillation mode
having typical values of resonance frequency below the rated power frequency of the system. Normally the thermal
turbines also have mechanical natural oscillation modes in that range of frequencies, leading, then, to the
possibility of under or sub-synchronous resonance occurrence.
C
1/6
1/3
6/6
5/6
LENGTH
Page 55
TypethenameofTB
Page 56
TypethenameofTB
conditions) is 1.10 p.u. of the rated or nominal voltage value (138 kV/152 kV; 230 kV/253 kV; 345 kV/380 kV; 440
kV/484 kV; 500 kV/550 kV).
Figure 3.4.11 FURNAS 600 kV HVDC bipole tower structures (near Itaber
substation)
The 600 kVDC transmission system main characteristics are as follows:
Two substations (Foz do Iguau the rectifier side and Ibina, the Inverter side);
Nominal Power: 6300 MW;
Nominal Voltage: 600 kV(DC);
Nominal Current: 2625 A(DC);
Two Bipoles ( 600 kV);
Each Bipole composed by two poles;
Each Pole composed by two 300 kV series-connected convertors.
Both AC/DC rectifying and DC/AC inverting processes require great amounts of reactive power and produce
harmonic currents that cannot be injected into the related AC network. At Foz do Iguau substation AC filters
totalizing 1540 Mvar have been installed. Because of its vicinities to Itaipu power plant, the needed reactive power
is mostly supplied by the power plant itself, performing the filters the role of not allowing the harmonic currents flow
into the 50 Hz AC network. At Ibina substation, besides AC filters (totalizing 2480 Mvar) have been installed, also
Page 57
TypethenameofTB
shunt capacitors, totalizing 590 Mvar, and 4 synchronous compensators of 300 Mvar each were defined to supply
the needed reactive power and proper short-circuit power level (for proper inverter thyristor valves operation).
One big concern is related to the application of Single-Phase Auto-Reclosing schemes (SPAR), in the AC system
close to inverter stations of bulk HVDC transmission systems, since this auto-reclosing mode (single phase
operation) may cause several commutation failures (CF) in sequence in the inverter station thyristor valves after the
OH line short circuit application and the opening of only its faulted phase.
This CF process is caused by the big asymmetry among the 3 phase-to-phase AC voltages applied to the thyristor
valves of the inverter station during the dead time of the auto-reclosing scheme (normally from 0.8 to 1.5 seconds).
The operation of the OH line with only two sound phases and one open phase during the dead time of the autoreclosing scheme, at the same time the transient recovery process of the inverter station occurs, establishes an
uneven condition for the HVDC system operation, i.e., the three phase voltages at the AC side in the inverter
station are strongly unbalanced, causing the asymmetry of voltages applied to the thyristor valves, leading to
potential commutation failures in series. This possible behaviour is of big concern since its consequences could be
catastrophic to the Power System stability and reliability (loss of stability and/or voltage collapse could occur).
Figure 3.4.12 Geographical distribution of load centers in Brazil (left) and average
distance between the main load centers and future Madeira river complex (right).
Bulb-type turbines
Page 58
TypethenameofTB
For new long distance transmission systems, some important features and concerns related to the planning/design
stage, as well as to equipment withstand requirements, should also be deeply investigated, such as:
a- comparison between equivalent solutions of 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
impacts caused to the 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 give protection 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 one under unrecoverable short-circuit conditions in the first line;
c- HVDC reliability features specification (ability of the system to transmit the rated power under contingency
conditions and outages), such as: operation modes (bipolar/monopolar) and related engineering/environmental
issues; temporary overloading/overcurrent capacity; AC System Faults/configurations that may create commutation
failures mainly in 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 the implemented mitigation solutions;
f- issues related to filtering requirements (short circuit level criteria, filter overloading, active filters usage 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 equipment withstanding requirements definition in the planning phase, taking into account the
possible/foreseen evolution/expansion of the AC grid/network (up-grading in short-circuit level requirements);
i- enterprise staging definition, i.e., intermediate stages of transmitting power capacity, considering cost constraints,
needs 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 of HVDC converter transformers;
k- possibilities of controlled switching usage.
Also, the following topics must be analyzed 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 performance of the line;
Level of power to be transmitted;
Weight and size of the equipment for shipping and transport (mainly the transformers for AC and DC
alternatives);
Power losses;
Spare parts;
Series connection of converter groups per pole (DC alternative);
Page 59
TypethenameofTB
Page 60
TypethenameofTB
System
nominal
voltage
750
750
Installation location
The bus side and
the transformer side
Line side
(Unit: kV)
The lightning
impulse residual
voltage under 8/20
s and 20 kA
Rated
voltage
(RMS)
Continuous
operating
voltage (RMS)
The switching
impulse residual
voltage under 30/60
s and 2 kA
600
462
1142
1380
648
498
1234
1491
Page 61
TypethenameofTB
3.5.7.1 L INE
INSULATION COORDINATION
The required values for the line tower minimum air gap are shown in Table 3.5.2.
The switching impulse air gap distance shall be selected according to the maximum switching statistical
overvoltage along the line, being 1.8 p.u..
The lightning impulse minimum air gap distance shall be selected according to 0.8 times of the lightning discharge
voltage of the insulator string.
The minimum air gap distance required for the safety distance of live working controls the line tower sizes, in the
Table 3.5.2 0.5m shall be considered for the human body action scope.
Page 62
TypethenameofTB
(Unit: m)
Power frequency
voltage
Switching impulse
Lightning impulse
Safety distance
requirement for live
line working
The required values of the equipment insulation level are shown in Table 3.5.3.
Equipment
Rated lightning impulse withstand
voltage
Rated switching impulse withstand
voltage
Power frequency withstand voltage
for 1minute
(Unit: kV)
Circuit breaker
longitudinal insulation
Transformer
1950
2100
2100+(650)
1550
1550
1300+(650)
900
Reactor 900
Other equipment 960
1270
Table 3.5.3 The rated insulation withstand voltage for 1000 kV equipment
No matter what the stress is the power frequency voltage, the switching impulse voltage or the lightning impulse
voltage, the insulation margin of the equipment internal insulation shall be more than 15 % and the insulation
margin of the equipment external insulation shall be more than 5 % for the insulation coordination.
The power frequency 1min withstand voltage selection shall be taking both TOV and the required power frequency
withstand voltage conversed from the lightning impulse withstand voltage into account.
The insulation coordination for the switching impulse withstand voltage shall be carried out based on the substation
switching overvoltage level (1.8 p.u.) and the arrester switching impulse protection level.
The insulation coordination for the lightning impulse withstand voltage shall be carried out based on the calculation
value of the substation lightning intruding overvoltage.
3.5.7.3 S UBSTATION
The A values of the substation minimum air gap distance shall be divided into three categories: A1' value, A1" value
and A2 values.
The A1' value is the minimum electrical distance of the substation wire to the frame; the A1" value is the minimum
electrical distance of the substation equipment to the frame; and A2 value is the minimum electrical distance
between the phases in the substation.
The recommended values of the minimum electrical distance for the substation are listed in Table 3.5.4.
Page 63
TypethenameofTB
(Unit: kV)
Phase to phase
A2
3.75
6.5
4.8
Phase to earth
A1
2.3
5.1
4.3
Power frequency
Switching impulse
Lightning impulse
A1
2.3
5.5
4.53
Table 3.5.4 The recommended minimum electrical distance value for the substation
(altitude: 1000 m)
3.6 India 800 kV project
3.6.1 FEATURES OF POWERGRIDS 800 KV TRANSMISSION SYSTEM
The 800kV level in the EHVAC transmission system of Indian Power System has been planned as a next higher
level to 400kV with an aim to transfer bulk quantum of power from power surplus region/area to power deficit
region/area. A number of 800kV transmission lines are being planned for transfer of power from Eastern Region,
already a power surplus region, and Southern Region, a future power surplus region, to power deficit regions like
Western and Northern Region. In this process large 800kV transmission network is being established not only for
transfer of power across the various regions but also for dispersal of the same within the regions specially from
Eastern or generation dominated part to Western i.e. demand dominated part of Northern and Western Region.
Initially, some of the 800kV lines were planned to operate at 400kV level, however with requirement of enhanced
power flow the same are to be operated at its rated voltage. The first line came into operation at 800kV voltage
level is Sipat-Seoni S/c line in Nov. 2007.
The study results shown here is for the Sipat-Seoni 800kV line (350 km), the Sipat end is a generating station while
the Seoni end is connected to the grid. The installed capacity of generating station at Sipat is 2980 MW (2 units of
500MW and 3 units of 660 MW) and the same is evacuated through 2 nos. of 800kV line connected with Seoni.
The schematic diagram is shown below:
Bina
400 kV
3x1500 MVA
350 km
800 kV
Khandwa
Seoni
Sipat
(3x660+2x500 MW)
400kV
Satpura
Bhilai
800kV
Wardha
Page 64
TypethenameofTB
1.9
Reclosing scheme
624
1180 kV peak at 1 kA
Switching impulse
1220 kV peak at 2 kA
Lightning impulse
1480 kV peak at 20 kA
Page 65
TypethenameofTB
3.6.4.1 D ETERMINATION
PROCESS FOR
LIWV
The severe lightning overvoltages are caused by the lightning within the substation. The lightning overvoltages
taking place on the overhead line several kilometers away from the substation are less harmful to the substation
equipment due to loss of a large portion of the lightning energy along the overhead line.
Studies have been carried out for both types of lightning surges i.e. back flashover and direct lightning stroke. In
case of back flashover, the lightning stroke on the overhead shieldwire or tower may generate overvoltages
resulting sparkover one insulator string. The arc across the insulator string causes transfer of part of the surge
current to the substation. The tower footing grounding also has a major impact on the overvoltages within the
substation as the major part of the stroke current flows into the ground. In case of direct lightning stroke the
penetration into the substation takes place via the phase conductor.
The details of the study result are given at Table 3.6.3.
Substation type and MOSA layout
Substation Type (GIS, MTS, AIS)
MOSA layout
(unit)
AIS
Line entrance
1 unit
Busbar
Transformer
1 unit
Representative
Value (kV)
Transformers : 1300
Calculation method
EMTP Analysis
Other : 1837
Safety factor
LIWV (kV)
Transformer
1950
Other equipment
2100
Back-flashover
Direct lightning
Tower 1
Substation circuit
Tower 1
Substation circuit
Page 66
TypethenameofTB
3.6.4.2 D ETERMINATION
3.6.4.2.1 S WITCHING
PROCESS FOR
SIWV
OVERVOLTAGE STUDY
The switching overvoltage studies are carried out both for line energization/ deenergization and for fault clearing
cases. of the 800kV Sipat Seoni line. The random nature of high switching overvoltages at Seoni substation due
to energization/re-energization of the 800kV Seoni-Sipat line is studied with the help of statistical tests consisting
100 energization per test.
In order to simulate the worst case, a trapped charge of 1.0 p.u. in the line is considered under line energization
condition. High switching overvoltages are also observed during line fault clearing. The magnitudes of overvoltage
vary depending upon source strength, line length, fault types and fault locations. Studies have been carried out for
phase to ground and phase to phase faults at various locations on Sipat Seoni 800kV line.
3.6.4.2.2 D ETERMINATION
OF
SIWV
The representative phase to ground and phase to phase overvoltages having 2% probability being exceeded are
observed as 1.48pu (968 kV peak) and 2.36pu (1543 kV peak) respectively. The same for fault clearing case has
been observed as 1.67pu (1093kV peak) and 2.67pu(1742 kV peak) respectively. Considering safety factor for the
severe condition, the SIWV has been considered as 1550kV peak for phase to ground and 2480kV peak for phase
to phase.
The details of the results are given in Table 3.6.4.
Representative Value
Switching
overvoltage
Transformer
Other equipment
Calculation method
EMTP Analysis
Safety factor
SIWV (kV)
Transformer
1550
(phase-to-ground)
Other equipment
1550
SIWV (kV)
Transformer
2480
(phase-to-phase)
Other equipment
2480
7.6m(cond-cond)
9.4m (rod-cond)
AND
E NERGY
The power frequency overvoltage on the sound phases due to 3-phase load rejection at the receiving end in 1phase to ground fault condition was studied. It has been observed that in some operating condition the temporary
overvoltage can reach as high as 1.4pu. This happened particularly during energizing of Sipat-Seoni line section at
Sipat on a permanent 1-phase to ground fault at the Seoni open line end. Further, studies was carried out for line
charging overvoltage following a load shedding (at Seoni end) or generation rejection (at Sipat end) in the SeoniSipat 800kV transmission system. In certain conditions the overvoltage may go close to 1.4 p.u. The energy
stresses in the surge arrestor may exceed its limiting value i.e. 8MJ, if the overvoltage persists for more than 1 sec.
Page 67
TypethenameofTB
The power frequency test voltage for the substation equipments are 870kV (RMS) for breaker and 975kV (RMS) for
CVT/CT, both for 1 min under dry and wet condition. The data for TOV and energy absorption by surge arrestor are
tabulated below.
Overvoltage on healthy phase in case of ground fault
8.1
Calculation method
EMTP Analysis
Transformer
Other equipment
Analysis program
EMTP-RV
3.7.1.1 T ARGET
OVERVOLTAGE
The switching overvoltage studies are carried out both for line energization/ deenergization and the kind of
overvoltage is classified as external and internal overvoltage. For the external overvoltage, direct stroke and back
flashover from transmission line is considered for the lightning overvoltage calculation. For the internal overvoltage,
the temporary overvoltage caused by the system operation or by fault is calculated. And the switching surge
overvoltage due to CB operation and the ground fault surge overvoltage of the transmission lines is also calculated.
As for the power frequency temporary overvoltage, overvoltage of sound phase due to phase-to-ground fault and
load rejection overvoltage has been analyzed. As for switching overvoltage, various kinds of fault condition have
been analyzed on the premise of application of the circuit breaker equipped with the closing resistors, and the
internal overvoltage for design of insulation level is shown in Table 3.7.1.
Page 68
TypethenameofTB
Classification
Temporary
overvoltage
Switching
overvoltage
Overvoltage p.u.
Phase to ground
Basis of p.u.
1 p.u. = 800 /
3 kV
1 p.u. = 800x 2 / 3 kV
Phase to phase
Note: The phase to ground switching overvoltage is 1.8 p.u. for the substation, and 1.9 p.u. for the transmission line design.
Transformer
Switchgear/Bus
HSGS
Calculated value
1710 kV peak
2015 kV peak
2025 kV peak
OF SYSTEM GROUNDING
The system is effectively grounded. The ratio of zero sequence impedance vs. positive sequence impedance shall
be maintained within the following range at any point under the condition of single phase to ground fault.
R 0 / X 1 1, X 0 / X 1 3
Neutral grounding shall be done solidly, not through any impedance. Grounding the whole system shall be done in
consideration of the following conditions in order to maintain the effective ground range. As for neutral grounding, a
bank of maximum capacity at every substation in order to the whole system to be within the effective ground range
(below 75 % of effective coefficient) shall be grounded.
If there is a limit to LIWV or the power station / substation are not sufficient to the fault current enough to the relay
operation, all the necessary transformers may be grounded. But, the capacity of the grounded transformers shall
be more than of the power capacity. The single phase to ground fault current shall not be less than 60 % of the
three phase short-circuit current at any point. The voltage rise of the neutral point that is not grounded shall be
maintained within 138 kV. The voltage rise due to single phase to ground fault are as follows;
Vb E a
( a 2 1)Z 0 ( a 2 a )Z 2
Z 0 Z1 Z 2
Vc E a
(a 1)Z 0 (a a 2 )Z 2
Z 0 Z1 Z 2
The voltage rise of sound phase due to single phase to ground fault shall not exceed 120 % of the maximum
system voltage (phase voltage). That is, Vb = Vc = 800 / 3 kV 1.2 = 555 kV. The standard lightning arrester can
be selected considering some safety factor to the above calculated voltage.
Page 69
TypethenameofTB
3.7.1.3 P ROTECTION
The key protection characteristics of the lightning arrester in Table 3.7.3 are as.
Classification
20
more than 5
Note: The rated voltage has been determined as 580 kV class so that both 576 kV of IEC and 588 kV of ANSI can be adopted.
Page 70
TypethenameofTB
3.7.1.4 R ATED
The key protection characteristics of the lightning arrester in Table 3.7.3 are as.
(170 kA, 170 s)
Rated insulation strength
of equipment
Transformer
(kV peak)
Switchgear/Bus
(kV peak)
HSGS
(kV peak)
Lightning impulse
insulation
2250
2250
Switching impulse
insulation
1500
1425
1425
830
830
Specification applied
ANSI
C57.12.00-1987
IEC 60694-1980
Substation Arrangement
- Scale of Substation
- Equip. Arrangement
- Surge Arrester Location
- Insulation type
Switching Overvoltage
- Overvoltage factor
- Arrester Energy
Surge Arrester
- Ratings
- Energy
Lightning Overvoltage
- Overvoltage of Equip.
- Arrester Energy
T.O.V
- 1L-G Fault
- Load Rejection
Lightning Design
- Shielding of Substation
- Selection of BIL
HSGS Design
- TRV Rating
- Voltage, Current Rating
Page 71
TypethenameofTB
The length of the line shall not be longer than 160 km in order to restrict the line end voltage rise within 800 kV
(1.05 p.u.) due to charging capacitance during the normal system operation.
3.7.2.2 A NALYSIS
The negative sequence current per power station due to un-transposed line of the 765 kV system during the heavy
load has shown 5.0 % in unbalance rate. This is because the voltage is stepped up from 22 kV to 765 kV on the
generator terminal.
3.7.2.3 A NALYSIS
OF LIGHTNING OVERVOLTAGE
For analysis of lightning surge, EMTP, commonly being used for analysis of the transient phenomena, has been
used and it has been assumed that the back flashover has penetrated into the substation by lightning stroke on to
the first steel tower located nearest to the substation. The insulation design of the substation against the lightning
stroke propagation into the substation is affected by the parameter such as magnitude of lightning, polarity, wave
front, transmission line configuration, place of lightning stroke, tower-footing resistance, etc., and especially by the
configuration of the substation and the operation condition. Table 3.7.5 shows the calculated results and margin to
LIWV.
Page 72
TypethenameofTB
Location of
overvoltage
Overvoltage
occurred (kV)
LIWV (kV)
Margin (%)
Remark
Incoming part of
substation
1937
2147
2250
4.8
10.8 % increased
Bus part of
substation
2120
2099
2250
7.2
1.0 % decreased
Transformer
1715
2050
19.5
3.7.2.4 A NALYSIS
OF SWITCHING OVERVOLTAGE
Switching overvoltage to be considered at insulation design for the transmission system can be divided into three
different types according to the cause of occurrence, the first one is, overvoltage due to phase to ground fault, and
the second one is overvoltage due to fault clearing and the last one is overvoltage due to line energizing. In case of
overvoltage due to line energizing, reclosing overvoltage may be divided separately considering residual electric
charge remains.
As for the 345 kV system and lower, switching overvoltage is not critical to consider since the magnitude of
overvoltage due to phase to ground fault and overvoltage due to fault clearing are relatively lower than the
overvoltage due to line energizing. However, as for the 765 kV system and above, the overvoltage due to phase to
ground fault and overvoltage due to fault clearing may show higher than the overvoltage due to line energizing
since the target value of the overvoltage due to line energizing is suppressed by pre-insertion resister.
The EMTP has been used for this analysis. The systems in 2005 and 2010 have been selected as the object and
the overvoltage of each case have been reviewed respectively. The calculation results of the switching overvoltage
are as follows:
Page 73
TypethenameofTB
(Unit: p.u.)
Classification
kind of surge
Phase to ground
Phase to phase
Remark
T/L
1.9
3.5
Closing resistor
1000
Ground fault
1.6
2.3
Fault clearing
1.7
2.9
Energizing/
Re-energizing
1.8
3.5
Summary
1.8
3.5
0.3
S/S
Neutral point
of Tr.
Supposing
failure of resistor
closing
3.7.3.1 F LOW
OF INSULATION COORDINATION
Insulation coordination in the power system shall mean a series of work in order to select each kind of equipment
comprising the system and insulation strength. The flow of insulation coordination according to the condition of IEC
60071-1 has been shown in Figure 3.7.3.
3.7.3.2 S ELECTION
OF REPRESENTATIVE OVERVOLTAGE
The representative overvoltages within the selective range are shown for the insulation design of the object
substation based on the analyzed results of lightning overvoltage and switching overvoltage shown in Table 3.7.7.
Table 3.7.8 shows the conditions and assumption of calculation conditions in representative overvoltage.
Page 74
TypethenameofTB
Determination of required
withstand voltage
- Test conditions
- Test conversion factor
- Standard withstand voltage
Figure 3.7.3 Flow chart for the determination of rated or standard insulation level
Kind
Value (kV)
Remark
Representative overvoltage of
power frequency
Representative overvoltage of
slow front
Representative overvoltage of
fast front
Representative overvoltage
of slow front(phase to phase)
1430-2020
Switching overvoltage
phase to ground
peak value
Lightning overvoltage for individual
equipment
peak value
Switching overvoltage
(phase to phase)
peak value
Page 75
TypethenameofTB
Representative overvoltage
Switching overvoltage
Lightning
overvoltage
Lightning current
15
Configuration of substation
Page 76
TypethenameofTB
Classification
Switching impulse
Transformer
Standard withstand
voltage (kV)
1425 I, 1500 A
Maximum
overvoltage (kV)
1176
1.32
1950 I
1.14
2050 A
1715
1.20
1.23
2100 I
1.04
2015
1.12
2250 I
2100 I
1.11
1890
Sh.R/HSGS
ACWV
1.21
1550 I
2100 I
Lightning
impulse
Safety factor
2250 I
1.19
Transformer
800 A
1.44
830 I
Sh.R/HSGS
830 I
555
1.50
1.50
Table 3.7.9 Withstand voltage available for selection of rated insulation level
3.7.3.3 N EUTRAL
INSULATION OF THE
765 K V
TRANSFORMER
Insulation strength of the transformers which have the maximum voltage of 300 kV and above have been designed
as non-uniform insulation transformer. The neutral point of the non-uniform insulation transformer have been
directly / indirectly connected with the ground and in case of indirect connection with the ground, there shall be
installed proper protection device between the neutral point and the ground in order to restrict the overvoltage
(switching and lightning overvoltage, and etc.).
Table 3.7.10 below described the rated insulation strength of the transformer neutral point for the 765 kV system.
The rated insulation strength has been selected from the standard lightning impulse withstand voltage in
consideration of the safety factor of more than 1.25 at the protective level of lightning impulse of the lightning
arrester. The power frequency temporary overvoltage has been selected from the rated insulation strength referring
to the relation between the power frequency temporary overvoltage and the rated insulation strength which is
recommended by IEC.
Page 77
TypethenameofTB
Item
Value
Remark
144 kV
550 kV
230 kV
3.7.4.1 D EFINITION
OF
A IR
CLEARANCE
Page 78
TypethenameofTB
3.7.4.2 A IR
CLEARANCE
The air clearance between phase to ground and phase to phase of the 765 kV bus shall be in accordance with
Table 3.7.11.
Nominal
voltage (kV)
765
Standard
Minimum
Outdoor
Phase to phase
Phase to ground
11000
8500
7000
5000
3.7.5.1 B ACKGROUND
It has been analyzed if installation of lightning arrester is necessary for the primary bus of the 765 kV GIS and for
the secondary main bus, incoming part and the shunt reactor terminal of the 345 kV GIS and if the overvoltage
exceeds the LIWV at the initial operation of the 765 kV substation.
The initial scheme of the 765 kV gas insulated substation is 4 circuits of transmission line and 2 banks of
transformer, half of the final size, and in this case the necessity of installation of the lightning arrester for the both
ends of the main bus has been analyzed using EMTP (Electro-Magnetic transient Program).
3.7.5.2 C ONCLUSION
OF ANALYSIS
- 765 kV side
The lightning arrester shall be installed at the both ends of the main bus even at the initial operation with 4 circuits
of transmission line and two banks of transformer and positioning lightning arrester shall be within six to eight
meters from the bus connection, so that the overvoltage can be reduced. The maximum overvoltage due to
lightning stroke in case of installation of lightning arrester at the both ends of the bus is 1765 kV and it has about
27 % margin for LIWV 2250 kV.
- 345 kV side
The maximum overvoltage will not exceed the standard LIWV allowing that the lightning arrester is installed at the
incoming part of the transmission line only and not installed for the bus and the shunt reactor additionally. The
maximum overvoltage at the incoming part and bus is 830 kV and has about 42 % margin for LIWV 1175 kV. The
maximum overvoltage of the shunt-reactor and the main transformers is 827 kV and has about 27 % margin for
LIWV 1050 kV.
3.7.6.1 R ATED
The rated voltage of lightning arrester is the maximum permissible power frequency r.m.s value on both ends of
lightning arrester for operating duty test and it is normally decided based on power frequency temporary
overvoltage, that the voltage rise of sound phase during single line to ground faults.
The multiple factor of power frequency temporary overvoltage applied to the lightning arrester for the line is 1.2 p.u.
which was selected according to the guidelines for insulation design of the transmission line and its analysis.
Based on the above, the rated voltage of the lightning arrester for 765 kV system is shown in the Table 3.7.12.
Page 79
TypethenameofTB
765
Power frequency
temporary overvoltage
(p.u.)
IEC code
(kV)
ANSI code
(kV)
1.2
(For phase)
2424 = 576
2425 = 600
588
612
0.3
(For neutral )
1212 = 144
1213 = 156
144
800
The discharge characteristics are represented as the line discharge test of the lightning arrester and the energy
absorbed by the lightning arrester has been reviewed. Based on the above, the nominal discharge current and the
line discharge class to be applied to the 765 kV systems have been shown in Table 3.7.13.
Use
Phase
20 (kA)
Neutral
10 (kA)
Table 3.7.13 Nominal discharge current and discharge class of lightning arrester
for 765 kV system
3.7.6.3 P ROTECTIVE
Characteristics
580 kV class, 20 A
( phase )
144 kV, 10 kA
( neutral )
Discharge voltage of
fast-front wave
Protective level of
lightning impulse
Protective level of
switching impulse
Notes: Discharge voltage of fast-front wave = Protective level of lightning impulse 1.1
Protective level of switching impulse = Protective level of lightning impulse (0.85-0.9)
3.7.7.1 A NALYSIS
Most of overvoltage in the substation is caused by lightning surge due to back flashover from the transmission line
and the direct lightning stroke. The required insulation strength of the substation equipment is determined mainly
by lightning surge due to back flashover. In case of a direct lightning stroke on to the equipment in the substation,
there may be a outage of power supply. In order to prevent outage it is a trend to protect the facilities from the
direct lightning stroke with installation of the overhead shielding wire. As for the 765 kV system facilities of the 765
kV substation, live parts are exposed up to the outgoing steel structure but the other facilities are enclosed in GIS.
Therefore, the height and distance and etc. of the overhead shielding wire to be installed on the dead-end steel
structure have been reviewed. At the same time, the voltage rise between the enclosure struck by lightning and the
relevant phase conductor, that is caused by lightning stroke and substation ground resistance, has been reviewed
by EMTP and the necessity of the shielding wire over the 765 kV switchyard has been reviewed.
Page 80
TypethenameofTB
3.7.7.2 E XAMPLE
OF ANALYSIS
The proper height and distance of the overhead ground wire in the 765 kV substation are in Table 3.7.15 below. It
is assumed that the 765 kV section of 765 kV substation is full GIS type, so the live part is the outgoing bushing
terminal of the transmission line and this is located under the dead-end steel structure. According to this calculation,
it is necessary to construct the gantry tower of 60 meter height for the overhead grounding wire every 2 Bay
(distance: about 130 m) in case that the height between the live part of the in/outgoing bushing of GIS and the
ground is 15 meters.
height of
shield wire
30m
35m
40m
45m
50m
55m
60m
10m
99.2
108.6
116.4
123.0
128.4
132.8
136.4
15m
87.6
99.2
108.6
116.4
123.0
128.4
132.8
20m
72.9
87.6
99.2
108.6
116.4
123.0
128.4
height
of live part
OF PROTECTION OF
GIS
- Purpose of review
In general, the outdoor steel structure type substation has the overhead ground wire over the switchyard in order to
protect the equipment from lightning. However, in case of GIS substation, since the GIS enclosure is grounded, it
seems not to flashover on the conductor even if lightning strikes the enclosure. The in/outgoing bushing to be
installed near the gantry structure can be protected by the overhead grounding wire to be installed over the gantry
structure. Therefore, voltage rise and possibility of lightning stroke have been reviewed by using EMTP and the
possibility of lightning stroke in order to judge the necessity of overhead grounding wire over the switchyard.
- Result of analysis
Lightning overvoltage between the bus conductor and the enclosure caused by the direct lighting stroke on to the
enclosure is proved to be about 1311 kV in the worst case, that is, the case of the substation ground resistance of
2 and further lightning stroke of 200 kA. In case the LIWV of GIS is designed 2250 kV, this value proved to be
only about 58 %. It has been proved that the overhead grounding wire over the GIS switchyard would not be
necessary in consideration that the substation ground resistance could be kept below 1 and the lightning stroke
of 200 kA class might occur one time after 300 years.
Page 81
TypethenameofTB
References
Japan
[1] A. Nakamura, et al., 1100 kV AC Transmission Project in Japan, International Symposium on International
Standards for Ultra High Voltage, 2007
[2] H. Okamoto, Progress of, and prospects for, commercial applications of UHV 1100 kV Transmission System in
TEPCO; Issues un upgrading existing 1100 kV designed lines from 550 kV to 1100 kV, Second International
Symposium on Standards for Ultra High Voltage Transmission, 2009
[3] E. Zaima, et al., Insulation Coordination for UHV AC Systems based on Surge Arrester Application (CIGRE
C4.306), Second International Symposium on Standards for Ultra High Voltage Transmission, 2009
[4] T. Watanabe, et al., Insulation Coordination for UHV System, CIGRE paper 33-101, 1998
[5] T. Takebe et al., Insulation Design for UHV Transmission Line in TEPCO, ISH 2005
[6] Y. Ishizaki, et al., High Performance Surge Arresters for UHV Systems, Second International Symposium on
Standards for Ultra High Voltage Transmission, 2009
[7] T. Kobayashi, et al., Compactness and High-reliability Technology for UHV Circuit-breakers, disconnectors and
earthing switches, Second International Symposium on Standards for Ultra High Voltage Transmission, 2009
[8] Y. Yamagata, et al., Utilities experience on design and testing for UHV equipment in Japan, Second
International Symposium on Standards for Ultra High Voltage Transmission, 2009
[9] Y. Yamada, et al., Experimental Evaluation of a UHV Tower Model for Lightning Surge Analysis, IEEE Trans.
on Power Delivery, vol.10, no.1, pp.393-402, 1995
[10] T. Shindo, et al., A New Calculation Method of Breakdown Voltage-Time Characteristics of Long Air Gaps,
IEEE Trans. on Power Apparatus and Systems, vol.104, no.6, pp.1556-1563, 1985
Brazil
[1] Itaipu Transmission System Consolidation of FURNAS Experience - Book published only in portuguese
under the coordination of S. S. G. MOTTA and H. A. M. FADINI ISBN 85-85996-01-3 - Rio de Janeiro 1995
[2] Standards Library (International Electrotechnical Commission IEC)
[3] America: Achieving its Potential in Electrical, Electronic and Multimedia Technologies IEC Latin America
Regional Centre Seminar 2007 (IEC-LARC) - May 2nd and 3rd, So Paulo - Brazil
[4] Regulatory Acts Library of the Brazilian Regulatory Agency for electric energy (ANEEL)
[5] HVDC Converter transformer performance on Itaipu. system - Gilson Bastos, FURNAS (Brazilian Cigr
National Committee Seminar; published only in Portuguese)
[6] Real-Time Simulations for Long-Term Stability Analysis of FURNAS 750 kV HVAC Transmission System, S.
Esprito Santo et allis, XIII SNPTEE-1995, Cigr-Brazil, Florianpolis, Brazil, October 1995 (Brazilian Cigr National
Committee Seminar; published only in Portuguese)
[7] Technical Requirements for UHV Switching Equipment Hiroki Ito et allis (CIGR Working Group A3-22);
IEC/CIGRE UHV Symposium Beijing, July 2007.
Page 82
TypethenameofTB
Page 83
TypethenameofTB
Page 84
TypethenameofTB
Page 85
TypethenameofTB
Country/Utility
China/SGCC
1100
1.4
0.5
India/Power
Grid
1200
1.4
1
Single phase
ground fault,
Load rejection
Ground fault
followed by
load rejection
N/A
40
55
EMTP analysis
EMTP analysis
Energy absorption of
surge arrester (MJ)
Calculation method
Italy/CESI
Japan/TEPCO
Russia/
1050
N/A
N/A
1100
1.5
0.17
Load rejection
followed by
ground fault
with selfexcitation of
generators
1200
1.3
3
N/A
55
40 *1
N/A
EMTP analysis
N/A
N/A
*1 Estimated from a specified current impulse withstand capability of 2.8 kA with a shape of 3/8 ms and used MOSAs.
Brazil/Furnas
800
1.25
1 cycle of 60 Hz
load rejection, line
energization, transformer
no-load energization, or
auto reclosing schemes
operation
China/SGCC
800
1.4
N/A
India/Power Grid
800
1.4
1
Energy absorption of
surge arrester (MJ)
N/A
15.27kJ/kV
8.1
Calculation method
EMTP/TNA analysis
EMTP analysis
EMTP analysis
Country/Utility
South Korea/KEPCO
800
1.2
Duration (sec.)
10
800
1.09
Relays trip for 1.05 p.u. / 5
min.
Transformer Energization
caused sympathetic
saturation of other
transformers at that bus
Energy absorption of
surge arrester (MJ)
2.6 kJ/kV
7.7
Calculation method
EMTP analysis
EMTP/PSCAD analysis
ATP/EMTP analysis
Venezuela
/EDELCA
800
1.4
1
Page 86
TypethenameofTB
Page 87
TypethenameofTB
UN
UN
U2
U1
U2
U3
U3
UN
U2
U1
U5
U3
U1
U2 U5 U4
U3
p.u
1.2
p.u
1.2
Voltage on sound phase(pu)
UN
LG
LG
1.1
1.1
1.0
1.0
U2
UN U3
U2
U5
U4
U2
U1
UN U3
U2
U5
U4
1 pu
U1
1100
kV
3
Overvoltage
The load rejection with the fault at long line that is connected through a transformer to a generator causes TOV
amplitude by Ferranti effect under self-excitation of generator. Figure 4.2.3 shows example of analyzed TOV in
1100kV system of Japan [3, 4, 6]. Such a case is quite rare in the system because the 1100kV system is looped
with the existing EHV grid. The simulation shows typical maximum TOV waveform whose amplitude and duration
are 1.5 p.u. and 0.17 second. Since the ratio of protection level and operating voltage is lower in the 1100kV class
than in the lower voltage classes, the energy absorption by a surge arrester is higher and the energy rating of an
arrester is specified as 55MJ and above in 1100kV system of Japan. To avoid an overload of the surge arrester,
the overvoltage protective relay system to open no-load 1100kV transmission line concerned will be applied.
Page 88
TypethenameofTB
The shunt reactors compensate degree to the line reactive power is usually in the range of 60%90%, depending
on allowed TOV, system can absorb reactive power, and load transmission capacity. The higher compensate
degree is, the lower TOV is. In order to avoid power frequency resonant overvoltage in non-three-phase operation
near 100% compensate degree is not allowed.
IEC60071-2 points out that usually the selection of the rated voltage of the surge arrester is based upon the
envelope of the TOV expected, taking into account the energy dissipation capability of the surge arrester. The
energy capability of the MOA under TOV stress is expressed as an amplitude/duration characteristic furnished by
the manufacturer. Therefore the rated voltage of the arrester usually is chosen same as the amplitude of the TOV
where the arrester is installed. Although the TOV on line side(e.g. 1.4p.u) is higher than bus side(e.g. 1.3p.u) The
arresters at line side select with same rated voltage of the bus side, taking advantage of the excellent withstand
TOV amplitude/duration characteristic of the UHV higher performance arresters. The lower TOV is, the lower rated
voltage of the arrester is selected.
Page 89
TypethenameofTB
In the 1100kV system of China, since the transmission lines is long, the high voltage shunt reactors are installed.
The shunt reactors has the following function:
(1) To limit the TOV no more than 1.4 p.u. at the line side and no more than 1.3 p.u. at the bus side.
(2) To compensate the reactive power produced by line capacitance, smooth the voltage distribution along line,
prevent self-excitation of generators and favor synchronism parallel operation of the line CB.
(3) To suppress the second arc current and transient recover voltage during single phase ground fault and favor
the fast single phase reclosing, coordinating with neutral reactor of the shunt reactor.
(4) To help reduction of the rated voltage of arrestor and switching overvoltage.
The maximum TOV occur in the single-phase ground fault case, in which CBs at the fault line side operate with
three-phase tripped and the CB at the other line dont operate. This case occurs under follow situations.
(1) In case the single phase reclosing operation after the single line fault is not successful, the CBs at three phases
are tripped by relay protection.
(2) In case of Live-Line Working, the single phase reclosing operation is forbidden and the CBs at three phases are
also tripped when a single phase ground fault occurs.
The relationship between TOV and compensation degree of the shunt reactor is shown in Figure 4.2.4 for the
China UHV pilot project, which dose not consider the transformer saturation and the TOV limitation of the arresters
in the TOV calculation. The TOV can reach more than 2.0p.u at zero compensation degree. Since the
compensation degree of shunt reactor is 87.8% for the China UHV pilot project, the maximum TOV occur less than
1.4 p.u. at the line side and 1.3 p.u. at the bus side. The relay protection to trip three phase CBs of both end of a
line synchronous is adopted by communication to shorten the duration time of TOV for reason of suppression of the
arrester absorbed energy. The duration time of TOV is 0.2 second as maximum delay time of trip system and less
than 0.5 second in which the CB is failure and backup CB operate. The TOV duration for surge arrester
withstanding 1.4 p.u. is 10 seconds according to the test data. The energy absorption allowable value 40 MJ is
decided.
Applying controllable shunt reactor is the best solution to coordinate the TOV limitation and reactive balance of the
UHV line. The controllable shunt reactor can automatically adjust the compensation degree according to the load
change, e.g. controllable shunt reactor operates at lower compensation degree when heavy load is transmitted and
quickly adjusts to maximum compensation degree to limit the TOV when the heavy load is rejected.
There are two types of controllable shunt reactors in service in China. One is the magnetically controlled type. The
output inductive capacity of the reactor can be smoothly and continually be adjusted from 10% to 100% by ion-core
saturation, which is controlled by injecting dc current to the magnetizing winding of the reactor. Another is the
reactance stepping controlled type. The configuration is shown in Figure4.2.5. The output inductive capacity of the
reactor can be stepping adjusted at 25%, 50%, 75% and 100%. The thyristor is connected in parallel with circuit
breakers to provide a fast switching operation and reduce switching transient. Both types of controllable shunt
reactors have been operated for more than 8 yeas in 500kV system. The 750kV stepping controlled type shunt
reactors is put into operation in 2011year and it is going to be used for the UHV ac double circuit transmission line
in China.
Page 90
TOV
TypethenameofTB
Compensation degree
Figure 4.2.4 The relationship between TOV and compensation degree of the shunt
reactor for the China UHV pilot project (U0 is caused by load rejection, U1 is
caused by load rejection in combination with single phase ground)
Figure 4.2.5 the single phase configuration of the stepping controlled shunt
reactors
4.2.2.3 INDIAN EXPERIENCE
In the 1200 kV system of India, preliminary studies of TOV have been carried out with ground fault of single line
followed by three phase interruption at receiving end. The result showed a value of TOV as 1.33 p.u. with 10000
MVA short circuit level at the sending end and the representative value of TOV is decided as 1.4pu. The required
energy capability of the surge arrestor has been considered. Figure 4.2.6 shows sequence of events for calculation
of surge arrestor energy accumulation. The series of events start with single line-to-ground fault followed by
opening of local and remote end CB of the faulted phase. The single phase auto recloser becomes effective after a
dead-time of 1000 ms. However the reclosing becomes unsuccessful and the CB could not be opened due to stuck
CB condition. The capability of energy assumption for the surge arrester could be calculated as follows.
Page 91
TypethenameofTB
Page 92
TypethenameofTB
4.3 Switching overvoltages caused by closing and opening with ground fault
overvoltage
4.3.1 GENERAL
UHV system demands more insulation clearance than current EHV systems, especially in case of clearance of
switching impulse voltage in air under rain condition. Since equipments sizes may be larger and total costs higher
according to the increasing insulation level, accurate estimation of overvoltage is needed. SIWVs are evaluated
based on amplitude of switching overvoltages (SOV) derived by calculation analysis, which are influenced on
decision of basic insulation design, fundamental allowance levels and selection of system components such as
overvoltage suppression equipments and protective devices. Electro-Magnetic Transient Program (EMTP) analysis
is generally used to evaluate amplitudes of overvoltages and shape of waveforms occurring in actual and planed
transmission circuits. (Table 4.3.1 and Table 4.3.2)
SOVs arise from closing, reclosing and opening operation in transmission line and at ground fault. Their circuits are
considered as representative circuits for transition analysis. The amplitudes of the overvoltages tend to depend on
the length of energized line, number of connections and length of other lines which are energized/reenergized, type
of circuit breaker (closing/opening resistor or not), arrangement and specification of arrestors and so on [1].
The occurrence probability of SOV is one of the factors to evaluate representative value. It is considered that the
probability can be approximated by Gaussian distribution. IEC 60071-2 express the representative value of
overvoltage that is characterized by its 2 % value of statistics distribution using lots of calculation results. Some
users evaluate maximum value as truncation voltage.
Representative switching overvoltage of UHV systems and 800 kV systems in each country would be suppressed
to become practical levels by application of CB with closing and /or opening resistor and effective specifications
and optimum placement of MOSAs. [4-5] Finally, SIWVs are decided from representative overvoltage and some
users added a safety margin.
Page 93
TypethenameofTB
Page 94
TypethenameofTB
Highest Voltage
Country/Utility
800 kV
765 kV
800 kV
Brazil / Furnas
- Opening resistor
1000-4000
- Closing resistor
400-600
- MOSA
- CB with controlled
switching device
Installed
800 kV
The United States /
AEP
Shunt reactor
Not installed
N/A
Reclosing scheme
Multi-phase reclosing
Fast singlephase
reclosing
TPAR scheme
Switching overvoltage(kV)
and occurrence case
(Transmission line)
N/A
1306 (2 p.u.)
(Maximum)
EMTP analysis
Simplified method
based on protection
level V2kA of surge
arrester
Gilman/Whitehead
model and EMTP
analysis
EMTP or PSCAD
Analysis
Tr.
1500
1520
1550
1700
Other
1425
1425
1550
1425
Tr.
N/A
N/A
N/A
N/A
Other
N/A
N/A
N/A
N/A
800 kV
The United States /
AEP
800 kV
800 kV
Venezuela/
EDELCA
800 kV
Calculation method
SIWV (kV)
(phase-to
ground)
SIWV (kV)
(phase-to
phase)
Highest Voltage
Country/Utility
India/Power Grid
South Africa
- Closing resistor
Closing resistor 450
- 37070 Ohm inserted Closing resistor 450
for 9 ms and line surge
for 8.3 ms and line
and line surge arrester
arrester
surge arrester
Shunt reactor
Reclosing scheme
Singlephase or three
phase reclosing in 0.5
sec.
Switching overvoltage(kV)
and occurrence case
(Transmission line)
1306 (2 p.u.)
(Maximum)
Calculation method
EMTP or PSCAD
Analysis
EMTP Analysis
ATP/EMTP analysis
EMTDC/PSCAD
Tr.
1700
1550
1550
1425
Other
1425
1550
1425+438
(Open CB)
1550/1425
Tr.
N/A
2480
N/A
N/A
Other
N/A
2480
N/A
N/A
SIWV (kV)
(phase-to
ground)
SIWV (kV)
(phase-to
phase)
Installed
(depends on line
Installed
length)
Not in service but has
singlephase or three
facilities to
Singlephase or threephase auto reclosing to single or three phase phase autoreclosing
auto reclosing
1090 (1.67 p.u.)
1306 (2 p.u.)
1241 (1.9 p.u., 2%
(2 % Value)
(2 % value)
value)
Ground fault clearing
Installed
Page 95
TypethenameofTB
Highest Voltage
1200 kV
Country/Utility
Russia/
Suppression measures
for switching
overvoltage
Closing resistor
and line surge
arrester
Closing resistor
600 and line
surge arrester
Closing resistor
600 and line
surge arrester
Shunt reactor
Installed
Installed
Reclosing scheme
Fast single
phase
reclosing
Fast single
phase
reclosing
Installed
Fast single
phase
reclosing within
1 sec.
Closing/Openin
g resistor 700
and line
surge arrester
Not installed
Fast single
phase
reclosing within
1 sec.
1570(1.6 p.u.)
(Maximum)
1473 (1.64
p.u.)
(2 % value)
Closing
1437-1527
(1.6-1.7 p.u.)
(Maximum)
Ground fault
1715
(ph.- ground
2 % value)
2743(ph.- ph.
2 % value)
Special
programme
Simplified
method based
on protection
level V2kA of
surge arrester
EMTP analysis
TNA (Transient
Network
Analyzer)
Simulation
Tr.
1800
1800
EMTP or
Simplified
method based
on protection
level V2kA of
surge arrester
1800
1425
1800
Other
1800
1800
1800
1550
1675
Tr.
N/A
2970
N/A
N/A
N/A
Other
N/A
2970
N/A
N/A
N/A
Switching
overvoltage(kV) and
occurrence case
(Transmission line)
Calculation method
SIWV (kV)
(phase-to
ground)
SIWV (kV)
(phase-to
phase)
1200 kV
India/Power
Grid
1100 kV
1100 kV
1050 kV
China/SGCC
Japan/TEPCO
Italy/CESI
Closing resistor
500 and gap
surge arrester
Installed
Fast single
phase
reclosing *1
Table 4.3.2 Condition of switching overvoltage studies for 1100 kV and above
systems
Canada 800kV
3
2.5
2.5
2.2
1.8
1.6
1.8
2.04
1.9
1.4
1.5
1
0.5
0
No mitigation
With closing
resister
With MOSA
With MOSA
at the line ends at the middle of
the line
Page 96
TypethenameofTB
Figure 4.3.4 Typical closing overvoltage with different closing resistances in 1100
kV system of China
with reactors
without reactors
2% Overvoltage(p.u)
2.2
2
1.8
1.6
1.4
1.2
1
0
Jindongnan
20
40
60
% of line length
80
100
Nanyan
Page 97
TypethenameofTB
Page 98
TypethenameofTB
4.3.2.3 G ROUND
FAULT OVERVOLTAGES
When a ground fault occurs, the phase voltage changes to zero. This phenomenon seems equivalent to inject the
surge waveform with reverse polarity voltage at the phase. The injected surge waveform induced to other healthy
phases of OH lines and multiple reflections generated in the lines are superimposed on AC waveform. Figure 4.3.9
shows analyzed waveform of maximum SFO at ground fault and Figure 4.3.10 shows ground fault overvoltage
distribution in a 1100 kV system of Japan [2, 4, 14]. Ground fault occurred at phase c and the maximum SOV of
1.62 p.u. has been induced in phase b in the middle of OH line. The overvoltage near substations can be reduced
by MOSA. In this case, the ground fault SOV gives the lower limit to determine the suppression level in the resistor
closing / opening of the circuit breaker.
Figure 4.3.11 shows example of analysis results of closing overvoltage in the 1100 kV system of China [3]
Jindongnen-Nanyang-Jingmen 1100 kV transmission line clearing ground faults leads to SOV. The maximum 2 %
voltage is 1.79 p.u. without resistor and 1.54 p.u. with 700 opening resistor for three-phase line-to-ground faults
and three-phase line breaker opening.
Page 99
TypethenameofTB
Jindongnan
Nanyang
Jingmen
(a) The circuit for overvoltage analysis when clearing line fault
Page 100
TypethenameofTB
Protection level of
Surge arrestor
Maximum SOV at OHlines
Maximum SOV at three
substations under the
above condition
Without R
500
1280 kV
(1.42 p.u.)
1280 kV
(1.42 p.u.)
1210 kV
(1.35 p.u.)
1250 kV
(1.39 p.u.)
1640 kV
(1.83 p.u.)
1770 kV
(1.97 p.u.)
1280 kV
(1.42 p.u.)
1390 kV
(1.55 p.u.)
V20kA =1620 kV
V20kA =1800 kV
V20kA =1620 kV
V20kA =1800 kV
With Resistor
750
1360 kV
(1.52 p.u.)
1380 kV
(1.54 p.u.)
1230 kV
(1.37 p.u.)
1290 kV
(1.44 p.u.)
1000
1440 kV
(1.60 p.u.)
1470 kV
(1.64 p.u.)
1250 kV
(1.39 p.u.)
1310 kV
(1.46 p.u.)
140km
223mH
U9
230km
140mH
U7
127mH
U6
Figure 4.3.12 Analyzed system model for the opening overvoltage for the
Jindongnan-Nanyang-Jingmen 1100 kV transmission line in China
Page 101
TypethenameofTB
Highest Voltage
Country/Utility
Representative
Value(kV)
Tr. /
reactor
765 kV
Canada/
Hydro
Quebec
800 kV
Brazil /
Furnas
800 kV
The United
States /
AEP
1306
(2.0 p.u.)
(Maximum)
1306
(2.0 p.u.)
(Maximum)
800 kV
800 kV
India/
Power Grid
Venezuela/
EDELCA
1093
(1.67 p.u.)
(2 % Value)
1093
(1.67 p.u.)
(2 % Value)
1318
(2.02 p.u.)
(Maximum)
1318+438
(Maximum)
(Open CB)
1140
(1.75 p.u.)
N/A
1140
(1.75p.u.)
N/A
1200
1140
1321
(at 3kA)
1200
1220
1197
Tr.
0.98
1.00
N/A
1.09
0.90
1.10
Other
0.98
1.00
N/A
1.09
0.90
1.10
Other
SIPL (kV)
at 2 kA
Representative
overvoltage /
SIPL
800 kV
South
Korea /
KEPCO
1176
(1.8 p.u.)
(Maximum)
1176
(1.8 p.u.)
(Maximum)
1200 kV
Country/Utility
Russia/
1100 kV
1100 kV
1050 kV
China/SGCC
Japan/TEPCO
Italy/CESI
1450
(1.69p.u.) *2
(Maximum)
1450
(1.69 p.u.) *2
(Maximum)
1450 (at 3kA,
30/60us)
1570
(1.6 p.u.)
(Maximum)
1570
(1.6 p.u.)
(Maximum)
1570
(at 2.8kA)
1500
(1.53 p.u.)
(2 % Value)
1500
(1.53 p.u.)
(2 % Value)
1392
(1.55 p.u.)
(2 % value)
1437
(1.60p.u.)
(2 % value)
1309
(1.46 p.u.)
(Maximum)
1400 (1.56
p.u.)
(Maximum)
1500
1460
(1440) *1
Tr.
1.00
1.00
1.00
0.91
1.00
Other
1.00
1.00
1.00
0.97
1.00
Tr.
Representative
Value(kV)
Other
SIPL (kV)
at 2 kA
Representative
overvoltage /
SIPL
1200 kV
India/Power
Grid
Page 102
TypethenameofTB
IEC 60071-2 proposes that the assumed maximum value of the representative switching overvoltage is equal to
the truncation value of the overvoltages or equal to the switching impulse protective level of the surge arrester,
whichever is the lower value. The truncation voltage at fault and fault clearing conditions can be estimated by
conservative method but the value of representative voltage is higher in comparison with maximum voltage that
occurred in actual power systems. In the meantime, since overvoltages calculated by computer simulations has
been recognized as accurate and credible, value, the calculated value is generally applied to representative value
without margin of error for simulation results. There are many results of simulations of actual and planned
transmission networks of above 800 kV systems, which are reported by in CIGRE A3.22WG Technical Brochure
362 and Technical Brochure 456 [4-5].
Some utilities have evaluated maximum value/2 % value calculated by simulations as representative overvoltage
and determined on SIPL which is in accordance with their policies. In India and some countries, the protection
levels of surge arrestors are selected on the basis of MCOV and energy capability due to TOV and SOV. As a
result, the representative value is closed to SIPL, which is consistent with the proposal of IEC 60071-2.
It may seem that some margins are not applied to the representative overvoltage value. However, in some cases,
the representative overvoltage is evaluated as the value with predominately low risk to failure. In case of South
Korea, since the most overvoltages will be much reduced with pre-insertion resistor of the 800 kV GCB and
maximum overvoltage is assumed for closing resistor insertion failure, which is quite rare, it is considered that a
higher margin is not necessary to the representative value.
As for other factors that include influence of small variations of the insulation strength and arrestor protective level,
some margin is considered as co-ordination factor to multiply the values of the representative overvoltages in some
systems.
Page 103
TypethenameofTB
(1)
When b is equal to 50a, the ratio of the wave tail to the wave front is 10, and this wave form is often used for
switching impulse flashover test. If it is assumed that b= 50a, then
(2)
0.85Vmax V (T0.85 ) k (e aT
0.85
e 50aT
0.85
(3)
dV (Tf ) / dt 0
(4)
From the equations given above, then the time-to-peak is determined by the following equation [7].
Tf 1.678(Tf T0.85 )
(5)
Page 104
TypethenameofTB
Page 105
TypethenameofTB
In China, some methods to define the front time Tf of slow front waveform are under consideration. The following 6
definitions of the front time are studied and tried to adopt analysis results of slow-front waveform that is occurring in
the Chinese UHV systems. Figure 4.3.15 shows the front time definitions.
Definition 6; Based on the actual waveform, the front time is determined by curve fitting with digital calculation. One
method is to adopt the double exponential fitting [9]. There are other methods such as modeling with
Steiglitz-Mcbride coefficients and low pass filtering [11].
Page 106
TypethenameofTB
1.5
1.6
*106
1.2
*106
Tf
1.0
Definition 1
Definition 2
0.8
0.5
0.4
T30%
T90%
0.0
0.0
-0.4
-0.5
-0.8
-1.0
-1.2
Tf
-1.6
-1.5
5
10
15
20
25
30
1.6
*106
1.2
35
[ms]
40
10
15
20
25
30
35
[ms]
40
[ms]
40
1.6
*106
1.2
Definition 3
Definition 4
0.8
0.8
T85%
0.4
Tp
0.0
0.0
-0.4
-0.4
-0.8
-1.2
-1.6
-1.6
10
15
20
25
30
Tp
-0.8
T f 1.678(Tp T85% )
-1.2
5
T70%
0.4
35
[ms]
40
T f 1.45(Tp T70% )
5
10
15
20
25
30
35
x 105
1.6
*106
1.2
U
vs. t
fit 1
14
Definition 5
Tangent
12
Definition 6
0.8
10
0.4
T85%
8
U(v) 6
0.0
-0.4
T85
-0.8
T f 1.18(T 85)
-1.2
-2
-4
-6
-1.6
5
10
15
20
25
30
35
[ms]
40
t(s)
8
x 10-3
281.3km
Jinmen substation
Jindongnan substation
960MVAR
720MVAR
720MVAR
600MVAR
Page 107
TypethenameofTB
Operation
Line name
JingdongnanNanyang
Closing point
Jingdongnan
1.42
50%
Nanyang
1.58
40%
Nanyang
1.60
60%
Jingmen
1.42
90%
at Nanyang side of
Nanyang-Jingmen
line
1.65
60%
at Nanyang side of
JingdongnanNanyang line
1.55
40%
Line energizing
_
NanyangJingmen
Single-phase
ground fault and
clearing
Single phase
ground Fault point
Place where
maximum
2% SOV
overvoltage
along line
occurs
*1
(p.u.)
(% of total line
length)
JingdongnanNanyang
_
NanyangJingmen
Definition
front time /
time-to-half
ms
1
2
4
1
2
1
2
3
4
4*2
5
1
2
1
2
3
4
4*2
5
1
2.9 / 5.6
1.3 / 4.0
1.95 / 4.2
4.4 / 6.4
1.5 / 6.73
3.83 / 6.73
4.23 / 7.13
3.07 / 5.97
2.95 / 5.85
1.76 / 5.3
6.14/ / 9.04
3.56 / 5.76
1.11 / 3.31
2.46 / 5.46
2.35 / 5.35
1.0 / 4.0
1.46 / 4.46
1.14 / 4.5
3.7 / 6.7
2.88 / 5.18
2.51 / 4.81
Table 4.3.6 Front-times, time-to-half and amplitudes of SOVs of UHV singlecircuit transmission lines caused by line energizing or Single-ground fault and
clearing
Page 108
TypethenameofTB
1 .5
[M V ]
1 .0
0 .5
0 .0
- 0 .5
- 1 .0
- 1 .5
0 .0 0
0 .0 2
0 .0 4
0 .0 6
0 .0 8
[s ]
0 .1 0
0 .6 0
*10 6
0 .1 4
T85
- 0 .3 2
- 0 .7 8
Tangent
- 1 .2 4
- 1 .7 0
2 1 .0
2 2 .5
2 4 .0
2 5 .5
2 7 .0
2 8 .5
[m s ]
3 0 .0
Figure 4.3.17 Calculation result of switching overvoltage on the 1100kV UHV GIS of
China
Figure 4.3.18 The UHV double circuits transmission lines in Huainan Shanghai
system
Page 109
TypethenameofTB
Operation
Line
energizing
Single-phase
ground fault
and clearing
Line name
Huainan
Wannan
(with outage
of one circuit)
Closing point
Single phase
ground Fault
point
Huainan
Place where
maximum
2% SOV
front time /
overvoltage
Definition time-to-half
along line
occurs
(p.u.)*1
(ms)
(% of total
line length)
1.43
100%
1.26
100%
1.62
42% of
Huainan
Wannan line
_
Wannan
Zhebei
(with outage
of one circuit)
Huainan
Wannan
Wannan
at Wannan side
of Wannan
Zhebei line (with
outage of one
circuit)
1
2
3
4
4*2
1
2
3
4
4*2
1
2
3
4
4*2
4.59 / 5.79
3.33 / 5.79
1.84 / 5.79
2.44 / 5.79
2.03 / 5.79
4.01 / 7.25
3.61 / 7.25
0.81 / 4.25
0.76 / 7.25
0.90 / 7.25
4.08 / 5.93
3.68 / 5.93
1.71 / 5.93
2.35 / 5.93
1.90 / 5.93
Table 4.3.7 Front-times, time-to-half and amplitudes of SOVs of UHV doublecircuit transmission lines caused by line energizing or Single-ground fault and
clearing
The maximum level of SOV is 1.65 p.u., which is less than the SOV insulation design level of 1.7 p.u. The results
correspond with those reported in CIGRE A3.22 for technical specifications of UHV systems.
The results of Table 4.3.6 and 4.3.7 suggest that the front times of switching overvoltages range from 0.76ms to
6.14ms. The front time of SOV on the 1100 kV systems of Japan and China tends to be longer than that of EHV
systems, because of long length of transmission lines. The definition of SOV front time with non-standard shape is
under consideration. It is necessary to clear the relationship between flashover voltage and front shape of nonstandard waveform regarding SOV. Further information on the definition of SOV front time will be provided by
CIGRE WG D1.
Page 110
TypethenameofTB
Page 111
TypethenameofTB
2
1.8
1.6
1.76
1.69
1.83
1.62
1.42
1.4
1.22 1.22
1.2
1.66 1.69
Maximum value
The location of
maximum value generated
Without resistor
1.36
1.33
1.23 1.22
1.20
1.25
3LG-3LO
1LG-3LO
Figure 4.3.19 Relationship between the locations of the fault and maximum SOV
Figure 4.3.20 Maximum SOV for different fault types and opening practice
Page 112
TypethenameofTB
1100 kV
Designed with
550 kV technology
Constructed tower
550 kV
S hin-tama line
Page 113
TypethenameofTB
[12] H.Ito, A.L.J.Janssen, D.Dufournet, Y.Yamagata, U.Riechert, P. Fernandez, M. Kosakada and D.Peelo,
Background information and Study Results for the Specification of UHV Substation Equipment, SC A3 colloquium,
Vienna, 2011
[13] H.Ito, A.L.J.Janssen, D.Dufournet, Y.Yamagata, U.Riechert, P. Fernandez, M. Kosakada and D.Peelo,
Considerations and Recommendations for the Specification of UHV Substation Equipment, CIGRE symposium,
Bologna, 2011
[14] Eiichi Zaima and Hiroshi Okamoto, System Design and Insulation Coordination for UHV AC Transmission
System in TEPCO, International conference of UHV Power Transmission Technology, Beijing, 2006
Page 114
TypethenameofTB
Transmission lines
Transmission tower
Page 115
TypethenameofTB
Highest Voltage
Country/Utility
Calculation method
Lightning current
Amplitude (kA)
Lightning current
waveshape
Backflashover
Lightning
stroke point
Substation
circuit
Lightning current
amplitude (kA)
Lightning current
waveshape
Direct
lightning
Lightning
stroke point
Substation
circuit
Highest Voltage
Country/Utility
Calculation method
Lightning current
Amplitude (kA)
Lightning current
waveshape
Backflashover
Lightning
stroke point
Direct
lightning
Substation
circuit
Lightning current
amplitude (kA)
Lightning current
waveshape
Lightning
stroke point
Substation
circuit
800 kV
Brazil/Furnas
Gilman/Whitehead model and EMTP
800 kV
India/Power Grid
EMTP
No data available
200
No data available
No data available
10
No data available
Ramp shape
(1.0 s/5 s)
1st tower
1st tower
1 & CB, 3 dia, 3 transformers, 2
lines with line reactors, 1 bus reactor
800 kV
China/
SGCC
EMTP
800 kV
Korea/
KEPCO
EMTP
800 kV
ATP/EMTP
225
170
20 to 300
Ramp shape
(1 s/70 s)
1st tower
Ramp shape
(0.5 to 3 s/50 s)
One tower of first five
line spans; usually the
second tower from S/S
line entrance
One and half CB or
Double CB
16
Not considered
N/A
1st span
N/A
N/A
Venezuela
1 to 30
Ramp shape
(0.5 to 1 s/50 s)
One phase of 1st tower
from S/S line entrance
One and half CB or
Doble CB
Page 116
TypethenameofTB
Highest Voltage
Country/Utility
Calculation method
Backflashover
Lightning current
Amplitude (kA)
Under evaluation
Lightning current
waveshape
Under evaluation
Lightning
stroke point
Under evaluation
Substation
circuit
Direct
lightning
1200 kV
India/Power Grid
Simplified method
based on
protection level
V20kA of surge
arrester
Lightning current
amplitude (kA)
Lightning current
waveshape
Lightning
stroke point
Substation
circuit
1200 kV
Russia/
1100 kV
China/SGCC
1100 kV
Japan/TEPCO
Special
programme
EMTP
EMTP
200-300
200
Ramp shape
(1s/70s)
Every tower in 2
km near S/S
-1 line and CB
open
-1 line and 2
busbar and 1
transformer
-1 line and CB
open
-1 line and 1
busbar
-1 line and 1
busbar and 1
transformer
Statistic method
taking into
account
Statistic method
taking into
account
Dependence
waveshape from
value of lightning
surge
Under evaluation
N/A
Under evaluation
N/A
Under evaluation
N/A
Under evaluation
N/A
Under evaluation
N/A
Ramp shape
(1s/70s)
-1 line and CB
open
-1 line and 2
busbar and 1
transformer
-1 line and CB
open
-1 line and 1
busbar
-1 line and 1
busbar and 1
transformer
30
1st tower
Table 4.4.2 Condition of lightning overvoltage studies for 1100 kV and above
system
Page 117
TypethenameofTB
I
If
If
tf
tf
t [s]
Figure 4.4.2 Ramp shape of current waveform: tf is the front time, tt is the time to
half value and If is the crest value of current [1]
According to the observational result of lightning stroke current to the transmission tower, the front duration tends
to be longer with larger amplitude of lightning stroke current, as shown in Figure 4.4.3. From the viewpoint of the
characteristics of the front duration in Figure 4.4.3, 1.0 s of the front duration in Korea and Japan seems to be
very short for higher lightning currents over 100 kA.
Page 118
TypethenameofTB
[kA]
0
PEAK : -120.0kA
Tf : 6.4s
-50
-100
0
20
40
60
80
100
Maximum rate-of-rise(kA/s)
102
10 1
10 1
CIGRE (y=3.9x0.55)[5]
y=1.27 x0.81(r=0.846)
95% confidence
interval of data
10 -1
101
10 2
101
10 2
Figure 4.4.3 Relationship between lightning stroke current amplitude and rate-ofrise [2, 5]
The lightning stroke positions are commonly the 1st or the 2nd tower near the substation. Meanwhile, China
considers every tower within 2km from the substation as the lightning location for the 1100 kV system.
With regard to the substation circuit, when calculating overvoltages, the normal operating configuration of one and
half CB system is assumed in Korea, India/800 kV and Venezuela. On the other hand, Japan has applied a circuit
configuration where the maximum overvoltage can occur at a line entrance, a busbar, or a transformer
independently. China provides the circuit conditions similar to Japan regarding to line entrance and transformer for
1100 kV system. As for the 800 kV system in China, the circuit of 1 line and 1 transformer is used.
The statistic method has been used in Russia. In USA/AEP, Brazil and Italy detailed data is not available. The
evaluation is proceeding in India/1200 kV.
Page 119
TypethenameofTB
Cumulative probability(%)
99.9
99
(1)
95
90
80
70
60
50
40
30
20
10
5
1
0.1
10
100
Highest Voltage
800 kV
800 kV
800 kV
800 kV
Country/Utility
Brazil/
Furnas
China/
SGCC
India/
Power Grid
Korea/
KEPCO
1591
1300
1710
1716
1837
2025
1573
1380
(busbar or Tr.)
1491
(line entrance)
1480
1310
1420
1418
Tr.
N/A
1.15
0.88
1.31
N/A
1.03
Other
N/A
1.15
1.24
1.55
N/A
N/A
Representative
Value(kV)
Tr. / reactor
Other
LIPL (kV)
at 20 kA
Representative
overvoltage /
LIPL
No data
available
No data
available
800 kV
The United
States/
AEP
No data
available
No data
available
800 kV
Venezuela
1463 kV
No data
available
Page 120
TypethenameofTB
Highest Voltage
1200 kV
1200 kV
1100 kV
1100 kV
1050 kV
Country/Utility
India/Power
Grid
Russia/
China/SGCC
Japan/
TEPCO
Italy/CESI
N/A
1796
1896
N/A
2040
2208
1760 (at
15kA)
1620
1620
1800
N/A
1.11
1.17
N/A
N/A
1.26
1.36
N/A
Representative
Value(kV)
Tr. / reactor
1700
Other
LIPL (kV)
at 20 kA
Representative
overvoltage /
LIPL
1700
Tr.
No data
available
No data
available
1.0
Other
Page 121
TypethenameofTB
REFERENCES
[1] IEC TR 60071-4, Insulation co-ordination - Part 4: Computational guide to insulation co-ordination and
modelling of electrical networks, First edition, 2004.
[2] J. Takami and S. Okabe, Observational Results of Lightning Current on Transmission Towers, IEEE Trans. on
Power Delivery, vol. 22, pp. 547-556, Jan. 2007.
[3] J. Takami and S. Okabe, Characteristics of Direct Lightning Strokes to Phase Conductors of UHV Transmission
Lines, IEEE Transactions on Power Delivery, Vol.202 No.1, pp.537-546, 2007.
[4] T. Watanabe, Y. Yamagata and E. Zaima, Insulation Coordination for UHV System, CIGRE Paper, No.33-101,
1998 Session.
[5] CIGR Technical Brochure 362 WG A3.22, Technical Requirements for Substation Equipment Exceeding 800
kV - Field experience and technical specifications of Substation equipment up to 1200 kV, December 2008, ISBN :
978-2 85873-049-0
Page 122
TypethenameofTB
Parameters
Z=92.68, v=273m/s
Ra=2, La=0.5H
C=126pF
C=5000pF
C=5000pF
C=5000pF
Z=310~363, v=300m/s
0.001s
Page 123
TypethenameofTB
The initial voltage between DS contactors before restrike depends on the trapped charge voltage of load side and
the instantaneous voltage of source side. Considering the extreme unfavourable condition, it is supposed that
restrike occurs when the trapped charge voltage of load side is -1.0p.u. and the instantaneous voltage of source
side is 1.0p.u. (1.0p.u. = 1100kV 2 / 3 = 898kV, it's the peak value of maximum continuous operation voltage
of China 1100kV UHV AC system).
As there is no standard VFTO waveshape and withstand voltage of GIS, the required VFTO withstand voltage of
GIS is specified by the lightning impulse withstand voltage (LIWV) divided by safety factor. The safety factor for
China UHV AC Demonstration project was specified as 1.15, which takes into account that the VFTO withstand
voltage is higher than LIWV for sound GIS insulation. For the other types of equipment, the safety factor is the
same as GIS, excluding overhead busbar.
A deterministic approach is applied for VFTO insulation coordination. The calculated maximum value of VFTO at
each equipment is assumed as their representative VFTO and is compared with the required VFTO withstand
voltage shown in Table 4.5.2 [5]. The calculated maximum value of VFTO should be lower than the required VFTO
withstand voltage. Arrester is not able to protect equipment efficiently from VFTO due to the very fast front of VFTO,
the inductance and distance effect of arrester.
Transformer
2250
1957
Shunt reactor
2250
1957
GIS
2400
2087
CVT
2400
2087
CT
2400
2087
Table 4.5.2 Standard LIWV and required VFTO withstand voltage of China 1100kV
equipment
4.5.2.1.2 SIMULATION RESULTS
1) VFTO in Changzhi GIS system
In Changzhi GIS substation, the double busbar and single busbar system are considered in design. Figure 4.5.2 is
the electric equipment arrangement 1 of single busbar system. It can be easily extended to one and a half busbar
system in the future. Figure 4.5.3 is the electric equipment arrangement 2 of double busbar system [6].
Page 124
TypethenameofTB
Arrangement
Arrangement 1
Arrangement 1
Arrangement 2
Damping resistor
Without
With
Without
GIS
2249
1250
2260
GIS bushing
2295
1141
1722
Transformer
954
905
942
Shunt reactor
1252
982
1012
CVT
1233
928
1015
Page 125
TypethenameofTB
Page 126
TypethenameofTB
Substation
Nanyang
Jingmen
Damping resistor
Without
With
Without
With
Shunt reactor
1052
989
1018
943
CVT
1006
966
1041
993
Overhead busbar
2742
1306
2421
1314
Table 4.5.4 Maximum values of VFTO in Nanyang and Jingmen Hybrid-IS systems
(kV)
3) Conclusions
(1) VFTO in Changzhi GIS system is severe. The maximum values of VFTO inside GIS in both arrangements reach
to 2249kV and 2260kV, exceed the required VFTO withstand voltage of GIS (2087kV), and it is necessary to use
damping resistor in DS.
(2) Damping resistor of 500 is used in Changzhi GIS system. The maximum value of VFTO can be reduced from
2249kV to 1250kV in arrangement 1, the amplitude and gradient of VFTO are decreased effectively and high
frequency components of VFTO are reduced obviously.
(3) VFTO values at all equipment in Nanyang and Jingmen Hybrid-IS systems dont exceed the required VFTO
withstand voltage, and it is unnecessary to install damping resistor in DS.
4.5.2.2 EXPERIMENT STUDY ON VFTO TEST CIRCUITS OF FULL SCALE UHV GIS
4.5.2.2.1 INTRODUCTION
In 2009, two full scale VFTO test circuits with practical 1100kV GIS were established by SGCC at UHV AC test
base in Wuhan, more than two thousand tests were carried out, and VFTO measurement systems were carefully
developed to capture the whole and accurate waveforms.
Page 127
TypethenameofTB
AS
BS1
BS2
BG2
BG1
DS2
DS1
DS
:GIS bushing
:Switching DS
:Auxiliary DS
:DC power source
AC Tran.
SHIKY
DC gen.
PINGGAO
PINGGAO DS
SHIKY DS
Page 128
TypethenameofTB
high frequency performance of measurement system were calibrated under voltage sources with different
waveshapes using a specially developed calibration setup in laboratory. The impedance transformation circuit was
designed to amplify the output impedance of capacitive divider and achieve the impedance matching with the
following circuit, thus the low cutoff frequency of the sensor can be extended greatly. The calibration results
showed that the bandwidth of measurement systems are about 0.01Hz-200MHz, there is no parasitic oscillation,
and the voltage ratio is stable, the performance is adequate for VFTO measurement.
Precharging
1.0p.u DC
voltage
Yes
No
Damping
resistor
PINGGAO
With
Without
SHIKY
Without
PINGGAO
Without
SHIKY
Without
120
450
450
450
450
Test GIS
9
3
0
9
Page 129
TypethenameofTB
Figure 4.5.11 Whole VFTO waveform during opening at the load side
The restrike numbers are statistically analyzed under condition that voltage across the DS contactors before
restrike is higher than 0.3p.u., as shown in Figure 4.5.12. The average restrike numbers are about 5 during
opening and 6 during closing of PINGGAO DS, which are about half of those of SHIKY DS.
Page 130
TypethenameofTB
SHIKY GIS
PINGGAO GIS
SHIKY GIS
PINGGAO GIS
30
Occurence probability / %
Occurence probability / %
35
30
25
20
15
10
5
0
25
20
15
10
5
0
10
12
14
16
10
12
14
16
18
Occurence Probability / %
14
SHIKY GIS
PINGGAO GIS
12
10
8
6
4
2
0
-1.0 -0.8 -0.6 -0.4 -0.2
0.0
0.2
0.4
0.6
0.8
1.0
Page 131
TypethenameofTB
Amplitude / kV
500
A1
-500
A2
-1000
-1500
-300
-150
150
300
450
600
750
Time / ns
PINGGAO
GIS
SHIKY GIS
Place
Magnitude
(p.u.)
Rise
time(ns)
A1
A2
A1
A2
0.77
2.38
0.99
2.64
8.8
366
6.4
298
Maximum
gradient
(kV/ns)
Strike
duration(s)
78.6
9.7
138.9
5.6
Page 132
TypethenameofTB
Measurement location
PINGGAO GIS
SHIKY GIS
1
2.01
2.20
2
1.98
2.17
3
2.27
1.86
4
1.47
1.30
30
25
20
15
10
5
0
1.0
1.2
1.4
1.6
1.8
Closing operation
Opening operation
35
Occurrence Probability / %
Occurrence Probability / %
35
2.0
2.2
2.4
30
25
20
15
10
5
0
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
Maximum VFTO/p.u.
Maximum VFTO/p.u.
Measurement location
PINGGAO GIS
SHIKY GIS
1
2.01
2.03
2
2.00
1.92
3
2.23
1.86
4
1.33
1.27
Page 133
TypethenameofTB
35
30
Occurrence Probability / %
Occurrence Probability / %
35
25
20
15
10
5
0
30
25
20
15
10
5
0
1.0
1.0
1.2
1.4
1.6
1.8
2.0
2.2
1.2
2.4
1.4
1.6
1.8
2.0
2.2
2.4
M aximum VFTO/p.u.
Maximum VFTO/p.u.
1
1.79
1.90
2.01
2
1.65
2.00
1.88
2.23
4
1.53
1.33
1
2.01
1.33
2
1.98
1.31
3
2.27
1.25
4
1.47
1.22
Page 134
TypethenameofTB
Figure 4.5.17 Waveforms of VFTO generated from PINGGAO GIS during closing
4.5.2.2.4 CONCLUSIONS
(1) The average restrike numbers are about 5 during opening and 6 during closing of PINGGAO GIS, which are
about half of those of SHIKY GIS, under condition that re-striking voltage across DS contactors is higher than 0.3
p.u.
(2) The probability of trapped charge voltage with negative polarity is higher than that with positive polarity when
moving contactor of DS is connected to source side. The probability of trapped charge voltage with high magnitude
of SHIKY GIS is obviously lower than that of PINGGAO GIS.
(3) The maximum VFTO is about 2.27p.u., which are close during opening and closing operation with or without
precharging 1.0 p.u. DC voltage.
(4) Damping resistor fixed in DS can mitigate the amplitude and gradient of VFTO significantly.
Page 135
TypethenameofTB
Chang-Nan
Line I
CVT
Disconnector
Br eak
Circuit Breaker
M1
VFTO-2
Bushing
T0111
VFTO-X
T011
T0112
VFTO-3
T0121
T0131
VFTO-1
CVT
VFTO-4
500kV
system
T012
1Main power
tr ansfor mer
T0122
Page 136
TypethenameofTB
Switching DS
T0122
T0131
T0111
T0121
T0112
1.11
1.29
1.00
1.30
1.07
1.25
1.06
1.00
1.00
1.06
1.00
1.00
1.02
Page 137
TypethenameofTB
Substation and
switching DS
T0222
T0211
Nanyang
T0221
T0212
T0312
T0321
Jingmen
T0322
T0311
1.00
1.05
1.02
1.22
1.13
1.12
1.11
1.05
1.07
Page 138
TypethenameofTB
1500
2000
1300
1500
Voltage [kV]
Voltage [kV]
1100
900
700
500
1000
500
0
-500
300
Measurement
Calculation
100
Measurement
Calculation
-1000
-1500
-100
1.8
1.9
2.1
2.2
2.3
2.4
1.8
2.5
1.9
2.1
2.2
2.3
2.4
2.5
Time [us]
Time [us]
Figure 4.5.21 VFTO calculation and measurement when switching busbars with a
GIS DS as per IEC 62271-102, without pre-charging (left), with pre-charging (right)
As stated above trapped charge remaining on the load side of a DS must be taken into consideration [13]. Totally
400, 520 records of measured trapped charge voltages during testing of DS have been evaluated for 550 kV, 1100
kV respectively. For the 550 kV DS, the maximum trapped charge reaches 0.6 p.u. during the tests with a source
voltage of 1.1 p.u. resulting in a most unfavorable voltage collapse of 1.7 p.u. In case of 1100 kV DS 90 % of the
trapped charge voltages were limited to 0.45 p.u. (see Figure 4.5.22).
8%
F(x) %
1100 kV
99,9
550 kV
99
6%
4%
Probability
95
90
70
50
30
10
5
1100 kV
2%
550 kV
0,10
0%
-1
-0.75
-0.5
-0.25
0.25
0.5
0.75
0.0
0.2
0.4
0.6
0.8
[pu]
1.0
Page 139
TypethenameofTB
roughness and the gas pressure were taken into account. The second parameter is the contact speed. The
simulation model simulates the test circuit for the relevant test duty according IEC62271-102 Annex F. The
capacitance values of the source and load side have been determined according the length and typical value of
GIS bus duct capacitance. The disconnector is modeled as an ideal switch that closes if the command is >0 and
opens at the next current zero if the command is equal to 0. The command is set to 1 if the voltage across the
disconnector exceeds the breakdown voltage characteristic of the disconnector. The starting point of the contact
separation has been varied in 1 steps all around 360. It can be seen that the measured trapped charge
distributions can be reproduced with good agreement for the 1100 kV disconnector (see Figure 4.5.23 right).
20%
100%
Simulated
Measured
80%
40%
TD1
TD2
TD3
10%
Probability
all TD
Cumulative probability
60%
15%
5%
20%
0.2
0.4
0.6
0.8
0%
1.0
0.0
0.2
0.4
0.6
0.8
1.0
40%
0.1 m/s
0.3 m/s
1.5 m/s
3 m/s
20%
Cumulative probability
60%
Trappedchargevoltage(abs.) [pu]
80%
calculated
90% probability
95% probability
99% probability
measured
0.8
0.6
0.4
0.2
0
0
0.5
1.5
2.5
3.5
0%
0.0
0.2
0.4
0.6
0.8
1.0
Page 140
TypethenameofTB
for VFTO calculations. The evaluation of type test results for the 1100 kV disconnector have revealed that the 90 %
voltage associated with trapped charges where 0.4 p.u. at a source voltage of 1.0 p.u. (see Figure 4.2.25). That
means for this 1100 kV disconnector having such a trapped charge behavior, a safety factor of around 1.43 (= 2
p.u./1.4 p.u.) is included when 1 p.u. trapped charge voltage is assumed for the VFTO calculation, assuming a
linear relationship between the voltage drop before closing and VFTO peak value. Moreover, it must be considered
that the VFTO peak value also depends on the location and the polarity of the trapped charge voltages. In case of
very low trapped charge voltages after opening, the VFTO peak values during opening could be higher compared
to a closing operation. A detailed study or the entire opening and closing operation is required in such cases.
Page 141
TypethenameofTB
Disconnector
under test
BDV characteristic:
- linear
- non-linear
Resistance at:
- Open state 1e16
- Closed state: 1e-16
Busbar model
Arc Model
Arc resistance = 2
Auxilay
disconenctor
HV Bushing
Surge impedance:
- AIS side: 347
- GIS side: 81.73
Capacitance to ground: 50 pF
Page 142
TypethenameofTB
i
r = r(t,uS,uL,i)
uS
uL
Figure 4.5.26 EMTP-ATP model of the disconnector [6]; r TACS controlled time
dependent resistance, uS disconnector source side voltage, uL disconnector
load side voltage
For the purpose of the model development the MODELS tool of the EMTP-ATP program has been used [16]. By
means of the control procedure, a decision is being made whether the spark is to be ignited or extinguished.
The decision is based on the values of potentials on the source and on the load side of the disconnector, which are
compared with the withstand voltage of the contact gap. The withstand voltage is calculated in each simulation step
on the basis of moving contact velocity obtained from mechanical characteristics of a real disconnector and the
withstand voltage characteristics of the disconnector contacts system. Trapped charge voltage is modeled with the
use of an additional capacitance, which is disconnected by the ideal switch after the ignition of the first spark in the
closing operation process.
Page 143
TypethenameofTB
Design parameters of the disconnector, such as contact speed, SF6 gas pressure and dielectric design of the
disconnector contact system, influence the withstand characteristics of the disconnector and hence influence the
Trapped Charge Voltage behavior of the disconnector.
The breakdown voltage (BDV) characteristic of the disconnector during the operation is one very important
parameter as input for simulation of the trapped charge behavior of the disconnector. Two types of breakdown
voltage characteristics were used for the simulations:
- Linear BDV (1)
- Nonlinear BDV (2)
The linear BDV characteristic was evaluated using the measurement of each breakdown voltage during one
switching operation. The obtained values are correlated to the position of the moving contact. Because the BDV
depends on the polarity of each contact a large scattering of measured values could be observed as shown in
Figure 4.5.27. The approximation used for the simulations is shown in Figure 4.5.27 (right). In general the linear
BDV does not include worst case of breakdown between contacts, thus electrical withstand rises faster, causing
higher TCV and lower sparking time.
Figure 4.5.27 Linear BDV (1) used as input, together with the approximation used
for simulations
The Nonlinear BDV characteristic was evaluated by lightning impulse and AC withstand voltage tests for different
contact distances and comparison with dielectric field calculations and is shown in Figure 4.5.28. For easier
simulation the BDV characteristic as shown in Figure 4.5.28 (right) was used. Generally the nonlinear BDV includes
worst case of the breakdown between contacts thus whole process of opening lasts longer and there is lower TCV
and longer sparking time.
Page 144
TypethenameofTB
Figure 4.5.28 Nonlinear BDV (2) used as input, together with the approximation
used for simulations
Trapped Charge Simulation for the SHIKY set-up
The simulation results for the slow acting disconnector in the SHIKY test set-up with use of linear BDV (1) are
presented in Figure 4.5.29. Two independent research institutes have performed the simulations of trapped charge
voltage using the same BDV characteristic. Both TCV distribution functions using the linear BDV characteristic are
in good agreement, but there are characterized by unsymmetrical distribution. The probability function (shape) is
the same for both simulations. Nevertheless it was proven that the differences between the simulations or model
parameter are very small.
Page 145
TypethenameofTB
Figure 4.5.30 Trapped Charge Voltage distribution, simulation results with use of
nonlinear BDV characteristics; opening operation
The simulation results have same probability function as for measurements. Trapped charge voltage obtained for
the model with nonlinear BDV has more realistic probability shape. It is close to the Gaussian function. But
simulation results have a lower TCV value than for measurements. There could be the following possible
explanations of the disagreement between simulation results and measurements: the nonlinear BDV characteristics
is likely to be more accurate characteristics and linear BDV is too much simplified.
To prove the accuracy of simulation results obtained with the use of nonlinear BDV, the following comparisons
were made: Trapped charge voltage distribution with use of nonlinear BDV vs. measurements during type tests of
1100 kV disconnector and sparking time simulations with use of nonlinear BDV vs. measurements from SHIKY test
set-up.
In Figure 4.5.31 the measurement results are presented which were used as a reference for comparison with
simulation results, see Figure 4.5.32. The simulation results were obtained with use of nonlinear BDV
characteristics. During the tests also the trapped charge voltages were measured using sensors on the load side.
Figure 4.5.31 Measured Trapped Charge Voltage Distribution (left); based on the
type Test measurements in Test Duty 1 set-up according to IEC (right)
Page 146
TypethenameofTB
Both simulations and measurements were made for the following data: 1100 kV design, opening operation at 0.4
m/s contact speed and test duty 1 set-up according to IEC. More details are given in [14, 15].
Page 147
TypethenameofTB
0.1
0.24
0.3
0.36
1.0
0.55
1.5
0.68
3.0
1.00
Page 148
TypethenameofTB
Page 149
TypethenameofTB
Figure 4.5.36 TCV and contact speed impact on sparking time and number of
sparks
4.5.3.5 COMPARISON OF MEASURED AND SIMULATED VFTO WAVEFORM
The waveform was measured on the load side of the operated disconnector, thus the VFTO has a classical stepwise shape, see Figure 4.5.37 and Figure 4.5.38. The trapped charge voltage levels are indicated.
1.5
TCV = 1
1*724.5 kV = 0.81 p.u.
0.5
-0.5
TCV = 1.1
-1
1.1*724.5 kV = 0.89 p.u.
-1.5
-0.02
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
Figure 4.5.37 VFTO waveform measured by SGCC in SHIKY layout sensor 2; entire
closing operation
Page 150
TypethenameofTB
2.5
2
1.5
1.5
1
0.5
1
0
-0.5
0.5
-1
0
-1.5
0.02
0.03
0.04
0.0307 0.0307 0.0307 0.0307 0.0307 0.0307 0.0307 0.0307 0.0307 0.0307
Figure 4.5.38 VFTO waveforms measured by SGCC in SHIKY layout sensor 2; closing
operation, max VFTO (second spark, TCV = -0.89)
The waveform was measured on the source side of the operated disconnector, thus the VFTO is superimposed on
the 50 Hz sine wave, see Figure 4.5.39.
Max VFTO = 2.1984 * 691254 =
= 1.5 MV = 1.69 p.u.
2
1.5
1
0.5
0
-0.5
-1
-1.5
-2
-0.02
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
Figure 4.5.39 VFTO waveforms measured by SGCC in SHIKY layout sensor 3; closing
operation, max VFTO (second spark, TCV = -0.89)
Comparison between simulation and measurement VFTO values for SHIKY set-up are given in Table 4.5.16. There
is a full agreement between measurements and simulations in sensor M2. Also simulation of VFTO value at
position M3 is in good agreement. This proves that the set-up was modeled correctly, in terms of surge impedance,
as based on the set-up elements dimensions, speed of electromagnetic wave, as based on the state-of-the-art
value and layout arrangement, as based on the layout drawing.
Page 151
TypethenameofTB
VFTO measured
VFTO calculated
with use of nonlinear BDV
M2
M3
Measuring point
Table 4.5.16 VFTO measured and calculated with use of nonlinear BDV
For verification the maximum VFTO peak values were calculated for the different measuring points of the SHIKY
test set-up. For the simulation the different BDV characteristics as well as the different trapped charge values are
used, summarized in Table 4.5.17.
nonlinear BDV
linear BDV
TCV distribution %
50%
98%
Worst case
50%
98%
Worst case
TCV p.u.
0.23
0.42
-1.0
0.50
0.92
-1.0
VFTO p.u. M1
1.82
1.94
2.34
2.00
2.28
2.34
VFTO p.u. M2
1.53
1.61
1.86
1.65
1.84
1.86
VFTO p.u. M3
1.83
1.95
2.35
2.02
2.29
2.35
VFTO p.u. M4
1.40
1.47
1.66
1.49
1.64
1.66
nonlinear BDV
linear BDV
TCV distribution %
50%
98%
Worst case
50%
98%
Worst case
TCV p.u.
-0.23
-0.42
-1.0
-0.49
-0.92
-1.0
VFTO p.u. M1
1.87
2.00
2.42
2.06
2.37
2.42
VFTO p.u. M2
1.45
1.52
1.73
1.55
1.70
1.73
VFTO p.u. M3
1.94
2.08
2.53
2.14
2.47
2.53
VFTO p.u. M4
1.52
1.60
1.85
1.64
1.82
1.85
Page 152
TypethenameofTB
The Layout of test set-up has negligible influence on TCV distributions. The highest impact parameter is the
disconnector design. This is by means of the disconnector Breakdown Voltage Characteristics: contact speed,
contact system design and SF6 pressure.
4.5.3.6 CONCLUSIONS
Normally the VFTO during closing is higher compared to the VFTO generated during opening operation. Therefore
the closing operation is more important, especially because of the trapped charge voltages as a result of the last
spark from the last opening operation.
The nonlinear BDV approach leads to the Trapped Charge Voltage (TCV) distributions with significantly lower
mean values than in the case of when linear BDV is applied. This in consequence leads to significantly lower VFTO.
By comparison of TCV simulations with TCV measurements, these results are in agreement in terms of the
distribution shape: both simulations and measurements have same Gaussian shape of probability function. The
prove of approach for BDV modeling is given by the comparison of sparking time distribution calculated with the
use of nonlinear BDV and the sparking time values measured. Here, good agreement was achieved for the
modified
The cumulative Trapped Charge Voltage distributions lead to the conclusion, that the value of -1 p.u. TCV is a
conservative assumption. The assumption of a more realistic value of the TCV, can lead to more realistic insulation
co-ordination calculations and in consequence allow for reducing the total cost of UHV substations. This can be
achieved by the reduction of the clearances and hence the size of the substations when the lower VFTO levels are
considered.
The application of slow operating disconnector doesnt produce high VFTO. It has to be mentioned however, that
such a speed reduction would lead to the increase of both sparking time and the number of sparks during the
disconnector operation process. The optimum design of the disconnector should be considered to achieve
significant VFTO reduction with acceptable sparking time and number of sparks.
Page 153
TypethenameofTB
Figure 4.5.40 1100kV GIS circuitry used for the surge calculation
GIS
Transformer
ZnO surge arrester
Overhead transmission line
Bushing
Opened circuit breaker
DS resistance
Calculation step time
Z = 95 , = 270 m/s
4600 pF
V10kA = 1550kV, V1mA = 1080kV
Z = 230 , = 300 m/s
500 pF
400 pF
50 - 1 k
1ns
Page 154
TypethenameofTB
The results of surge calculation for DSs at various points in Figure 4.5.40 are shown in Figure 4.5.41 (a), (b) and (c).
They show the followings:
With no resistor in the DSs, the crest value of overvoltages is 2.8p.u. (2510kV) and exceeds LIWV
2250kV (2.5p.u.).
f the resistance of the DS resistors is 200 or above, overvoltages can be suppressed below about
1.5p.u. When the resistance exceeds 500, they saturate approximately 1.2p.u.
The larger the resistance, the higher the voltage applied to the resistor. Dielectric capability of resistance
is required to be 1700kV when the resistance is 500.
Page 155
TypethenameofTB
The larger the resistance, the larger the energy consumed by the resistor, in the resistance range below
200, beyond which the energy eases off. Again, the larger the capacitance at the load side of the DS,
the larger the energy. The resistor energy shown in Figure 4.5.19 (c) was max when the longest bus
(35m) was used at the load side of the DS.
DS with resistor was adopted to reduce the disconnecting surge below LIWV level. To suppress the overvoltage as
low as possible and to reduce the dielectric and thermal capability of resistor, the value of resistor was determined
to be 500.
Page 156
TypethenameofTB
Page 157
TypethenameofTB
Without
resistor
1.0
0
-1.0
With
resistor
(Measured)
(Analyzed)
Page 158
TypethenameofTB
Without bushing
With busing
High-voltage line
terminal ground
Without bushing
With bushing
Lightning impulse
voltage(2)
(1.2/50 s)
23.1%
15.3%
2.3%
37.3%
23.7%
4.5%
16.8%
13.4%
4.5%
Table 4.5.21 Voltage generated inside winding (near high-voltage line terminal)
Table 4.5.22 gives the calculated voltages based on Table 4.5.21 when a DS surge of (2/3)*1,100kV restrike and
lightning impulse voltage of 1,950kV are applied. These are rough estimates by conversion to the standard wave.
The values indicate that a voltage equivalent to or higher than LIWV may frequently be generated locally. Thus, it
would be preferable to adopt a DS with resistor for securing insulation reliability.
An intruding DS surge may also cause local increase of the voltage distribution in winding-type equipment such as
a voltage transformer (VT) or a bushing with internal shield for field control. Therefore, a DS without resistor would
be better to be used with careful attention given to DS surges, as with a transformer.
Page 159
TypethenameofTB
Measurement
point
Between turns of
first Section
Between first and
second sections
Between second
and third sections
Lightning impulse
voltage
(1.2/50 s)
2/ 3*1100kV*0.153/1.4 = 137kV
34kV
1950*0.023 = 45kV
2/ 3*1100kV*0.237/2.3 = 92kV
23kV
1950*0.045 = 88kV
2/ 3*1100kV*0.134/2.3 = 52kV
13kV
1950*0.045 = 88kV
Note: (1) Breakdown voltage of a steep wave having a pulse width of 70 ns was 1.4 times higher than that of the standard
wave between turns and 2.3 times higher at an oil gap [1] and their multiples were used for conversion to the
standard wave.
(2) DS surge with resistor was set to 1/4 of that without resistor.
Page 160
TypethenameofTB
[5] Du Shuchun, Zhang Cuixia, Ge Dong. Lightning protection for 1000kV UHV AC substation and switching station
[Technical Report]. China Electric Power Research Institute, Beijing. 2006.
[6] Ban Liangeng, Lin Jiming, Xiang Zutao, et.al. Very fast transient overvoltage analysis for GIS and HGIS of
China 1000kV UHV Pilot Project [Technical Report]. China Electric Power Research Institute, Beijing. 2006.
[7] Y. Yamagata, et al. Suppression of VFT in 1100kV GIS by adopting resistor fitted DS IEEE Transactions on
Power Delivery, Vol. 11, No. 2. April 1995
[8] YAMAGATA, Y.; OKABE, S.: Utility's experience on design and testing for UHV equipment in Japan, The
second IEC CIGR International Symposium on International Standards for UHV Transmission, 29-30 January
2009, New Delhi, India, proceedings pp. 142-154
[9] CIGR Working Group A3.22: Background of Technical Specifications for Substation Equipment Exceeding
800 kV AC, Brochure , to be published in 2010
[10] CIGR Working Group 15.03: GIS Insulation Properties in Case of VFT and DC Stress, Report 15-201, 36th
CIGR Session, Aug. 25 to 31, 1996, Paris, France
[11] CIGR Working Group 33/13-09: Monograph on GIS Very Fast Transients, Brochure 35, July 1989
[12] Riechert, U.; Holaus, W.; Krsi, U.; Sologuren, D. Gas-Insulated Switchgear for 1100 kV - Challenges in
Development and Testing, CIGR 6th Southern Africa Regional Conference, Colloquium of CIGR SC A2/A3/B3,
2009, Somerset West, 17 - 21 August 2009, South Africa
[13] Riechert, U.; Krsi, U.; Sologuren-Sanchez, D.: Very Fast Transient Overvoltages during Switching of BusCharging Currents by 1100 kV Disconnector, CIGR Report A3-107, 43rd CIGR Session, August 22-27, 2010,
Palais des Congrs, Paris, France
[14] Riechert, U.; Holaus, W.: Ultra High Voltage Gas-Insulated Switchgear A Technology Milestone, Euro.
Trans. Electr. Power (2011), (wileyonlinelibrary.com). DOI: 10.1002/etep.582
[15] Szewczyk, M.; Stosur, M.; Piasecki, W.; Florkowski, M.; Fulczyk, M.; Steiger, M.; Kostovic, J.: New
disconnector model for Very Fast Transient studies in High Voltage Gas Insulated Substations, in Proc. 2010
EEUG Conf., pp.154-163
[16] Szewczyk, M.; Piasecki, W.; Stosur, M.; Riechert, U.; Kostovic, J.: Impact of Disconnector Design on Very
Fast Transient Overvoltages in Gas-Insulated UHV Switchgear, Proceedings of 17th International Symposium on
High Voltage Engineering (ISH), August 22nd 26th, Hannover, Germany
[17] Riechert, U.; Neumann, C.; Hama, H.; Okabe, S.; Schichler, U., on behalf of CIGR WG D1.36 and AG D1.03:
Basic Information and Possible Counter Measures Concerning Very Fast Transients in Gas-Insulated UHV
Substations as Basis for the Insulation Co-ordination, CIGR SC A2 & D1 Joint Colloquium 2011, Kyoto, Japan,
PS3-O-5
[18] M. Szewczyk, M. Stosur, W. Piasecki, M. Kuniewski, P. Balcerek, M. Florkowski, Uwe Riechert:
"Measurements and Simulations of Very Fast Transients during Disconnector Type Testing in UHV Gas-Insulated
Switchgear, Electrotechnical Review" (accepted for publication)
Page 161
TypethenameofTB
Figure 4.6.1 500 kV circuit breaker TRVs and 396 kV rated surge arrester
Page 162
TypethenameofTB
Figure 4.6.2 800 kV circuit breaker TRVs and 588 kV rated surge arrester
Figure 4.6.3 1100 kV circuit breaker TRVs and 828 kV rated surge arrester
Page 163
TypethenameofTB
Figure 4.6.4 1200 kV circuit breaker TRV and 850 kV rated surge arrester
Review of Figures 4.6.1 to 4.6.4 show that the MOSAs will limit the peak values for T60 and below at 800 kV and
for T100 and below at 1100 kV and 1200 kV. The results of simulation studies for the 1100 kV case are shown in
Figures 4.6.5 and 4.6.6[3-5]. The limiting effect of the MOSA is evident.
Figure 5.6.5 Simulation of MOSA influence on T100 and T60 TRV values [4]
Page 164
TypethenameofTB
2000
TLF
L L F (T 3 0 )
T60
T100
1800
1600
1478kV
1411kV
1400
1200
1000
800
600
400
200
0
0
200
400
600
800
T R V p e a k in 1 1 0 0 k V s y s te m w ith o u t M O S A
Source: Cigre WG A3.22 Second Technical Brochure
Figure 4.6.6 Simulation of MOSA influence on T100, T60, T30 (LLF - longline fault)
and TLF (transformer limited fault)
Note: The TRV value is the voltage difference between the circuit breaker terminals in each phase after the current interruption,
while the MOSA SIPL is the arrester residual voltage phase to ground. MOSAs are applied in their usual role in power systems,
i.e., to ensure lightning and switching overvoltage protection of the apparatus in their vicinity. Therefore the MOSA are
considered as devices connected from each phase to ground, on one or both sides of circuit breaker.
Page 165
TypethenameofTB
across the circuit breaker rather than to ground. Note that the use of this possibility may be limited by system outof-phase switching considerations.
The MOSA applications for two examples have completely different purposes so it is recommend that they should
deal with different technologies as shown in Table 4.6.1.
Turkey 420 kV
series compensated line
All ratings
TRV limitation
Applications
Voltage
Purpose of MOSA
Installation of MOSA
Influence on TRV
Compensation scheme
Series capacitor
2000
Similar TRV for 40km line fault to LLF: Breaking current 9.6 kA (T30)
Standard TRV-Uc:1660kV
TRV across CB
Voltage(kV)
1500
1000
500
Time(ms)
Figure 4.6.7 Influence of MOSA on TRV in the UHV system, where MOSA is applied
to suppress the lightning and switching overvoltages
Page 166
TypethenameofTB
4.6.4 CONCLUSIONS
The value of MOSAs in limiting TRVs, particularly for UHV systems, is evident and the following can be concluded:
The influence of MOSAs on mitigating or limiting circuit breaker TRV peak values has to date been viewed as
latent hidden value.
Consideration of this influence has considerable merit but generalization is difficult because MOSA ratings, and
hence protective levels, are system-specific being dependent on expected TOV magnitudes and durations. Also,
TOVs characteristics may vary over different parts of the system.
Users should recognize the technical advantage of TRV peak limitation and the associated economic value that
the approach offers by allowing the use of circuit breakers with fewer interrupters in series.
If the MOSA TRV limiting approach is pursued, then the user is advised to ensure that the actual application
TRVs are covered by available circuit breaker type test reports. The most recent amendment to the IEC circuit
breaker standard addresses the TRV requirements for 1100 and 1200 kV circuit and the calculation of the test
TRVs as influenced by opening resistors [6]. The additional influence of MOSAs can be calculated in a similar
manner.
REFERENCES
[1] B.L. Avent, D.F. Peelo, J.H. Sawada, Circuit Breaker TRV Requirements for a Series Compensated 500 kV
Line with MOV Protected Series Capacitors. Colloquium of Cigre Study Committee 13, Florianopolos, Brazil,
September 1995.
[2] F.M. Gatta, F. Illiceto, S. Lauria, B. Dilli, TRVs Across Circuit Breakers of Series Compensated Lines. Analysis,
Design and Operational Experience in the 420 kV Turkish Grid. Cigre Paper No. 13 109, Session 2002.
[3] Cigre Technical Brochure 362 Technical Requirements for Substation Equipment Exceeding 800 kV.
December 2008.
[4] Cigre Technical Brochure 456 WG A3.22, Background of Technical Specifications for Substation Equipment
Exceeding 800 kV AC, April 2011, ISBN: 978-85873-145-9.
[5] A.L.J. Janssen, Y. Yamagata, D. Dufournet, H. Ito, M.C. Bhatnagar, M. Kosakada, Y. Filion, J. Amon and P.
Fernandez, UHV Equipment Requirements: State of the Art and Prospects for Equipment, Paper No. 4-1, New
Delhi 2009.
[6] IEC 62271-100A1: Amendment 1 High voltage switchgear and controlgear Part 100: Alternating current
circuit-breakers (Addition of requirements and tests for 1100 and 1200 kV)
Page 167
TypethenameofTB
4.7 Conclusions
UHV clearance in the air may require longer distance than those applied to EHV system, because the dielectric
strength does not increase linearly with an increase of the air gap. Accordingly, optimal insulation co-ordination
becomes extremely important for UHV systems to achieve a compact, economic and environmental-friendly UHV
transmission. The precise calculation methods, advanced simulation and mitigation techniques for the overvoltages
in the system can achieve the reduction of the dielectric requirements. The analytical tools like EMTP are very
useful for the precise evaluation of the overvoltages. They can optimize the insulation co-ordination design for the
power network systems and rationalize the test voltage for the power equipment.
The amplitude and duration of temporary overvoltage occurring in the UHV systems are calculated. The amplitude
of temporary overvoltage levels and required energy absorption of surge arrestor in UHV systems are relatively
higher than those in EHV systems. The waveforms of switching overvoltage occurring in UHV transmission lines
are estimated and their front times are more than several hundred microseconds that is bigger than those in EHV
systems. As for lightning current, its waveform is investigated in actual UHV transmission lines and the ramped
shape waveform is used for the overvoltage analysis. The lightning currents more than 200 kA for back flashover
calculation are specified in the over 800 kV systems and the lightning overvoltages are estimated and lightning
currents for shield failure calculation are determined by EGM and other methods according to tower configuration.
In GIS or Hybrid-IS system, the switching of disconnecting switch will produce very fast transient overvoltage
(VFTO). VFTO has very high gradient and high frequency components. The serious VFTO can damage the
insulation of GIS and the interturn insulation of transformer, and cause transient enclosure voltage (TEV) and
electromagnetic interference (EMI) to low voltage control systems.Here, VFTO was investigated in detail based on
Chinese and Japanese data.The applications of the suppression technologies to reduce these overvoltages are
adopted in over 800kV systems. They contribute the downsizing of their power equipments and improve the
reliability of the system.
For 800 kV and the above systems, the representative values of overvoltages, their waveforms and occurrence
conditions have been practically investigated for the circuits of each country. The overvoltage characteristics
reported by each country are summarized as follows.
1) In UHV systems of various countries, representative levels of temporary overvoltages and switching
overvoltages are in the range from 1.3 to 1.5 p.u. and from 1.6 to 1.7 p.u., respectively. Here, the voltage of 1 p.u.
is defined as the peak value of phase to the earth of operating voltage.
2) The energy absorption of surge arrester for temporary overvoltages of UHV system is specified from 40 to 55 MJ,
which are bigger than those for 800 kV and EHV systems.
3) The methods such as CB with opening and/or closing resistor, shunt reactor and appropriate MOSA installation
are verified to be effective to mitigate the system overvoltages.
4) The front times of the switching overvoltages in the UHV systems, which are calculated by the simulations, tend
to be more than several hundred micro seconds. These are longer than those of the standard SI test waveform
defined in IEC 60060-1.
5) The front time duration of non-standardized waveform strongly depends on how to convert the waveform to the
standard test waveform. Accordingly, the clear understanding of the conversion method and the definition for the
front time is required.
6) The simulation results show that the CB with opening and/or closing resistor can reduce the maximum switching
overvoltage by 10% and more. MOSA installed at the line entrance is effective to reduce the switching overvoltage
levels of the substation and transmission line.
7) In Chinese system, the shunt reactors installed at the line side is special effectively to limit the TOV and
switching overvoltage for long transmission line. In addition, they have other functions such as balancing system
Page 168
TypethenameofTB
real power, suppressing the second arc current, preventing self-excitation of generators and reducing the MOSA
rated voltage (reducing SIPL and LIPL as well).
8) Regarding lightning overvoltage, the simulation method and the circuit conditions are studied by using EMTP
analysis. The results show that the lightning current waveform is much influenced by the lightning overvoltages.
The maximum lightning current more than 200 kA is generally used in the 800 kV and the above systems for back
flashover calculation and lightning currents for shield failure calculation are determined by EGM and other methods
according to tower configuration.
9) The recent investigation regarding lightning current waveform makes clear the relationship between the current
peak value and the front time duration. However, the lightning current waveform and its maximum value should be
defined based on the actual phenomena.
10) The ratios of representative overvoltages to LIPL for transformers in UHV systems are in the range from 1.0 to
1.17 p.u. Those for other equipment are from 1.0 to 1.36 p.u. In 800 kV systems, the ratios for transformers and
other equipment are in the range from 0.88 p.u. to 1.31 p.u. and from 1.15 p.u. to 1.55 p.u., respectively. The ratios
for UHV systems are likely to be smaller than those for 800 kV and the lower voltage systems, because the
overvoltage mitigation techniques including higher performance arrestors can drastically reduce the lightning
overvoltage levels.
11) The cumulative Trapped Charge Voltage distributions lead to the conclusion, that the value of -1 p.u. TCV is a
conservative assumption. The assumption of a more realistic value of the TCV, can lead to more realistic insulation
co-ordination calculations.
12) The application of slow operating disconnector doesnt produce high VFTO, but increase of both sparking time
and the number of sparks during the disconnector operation process. The optimum design of the disconnector
should be considered to achieve significant VFTO reduction with acceptable sparking time and number of sparks.
13) Damping resistor installed in DS can significantly suppress the amplitude and gradient of VFTO caused by
switching of DS. Application of DS with damping resistor or no use of damping resistor should be decided,
according to VFTO simulation considering the substation layout of GIS/Hybrid-IS. Regarding with neighbouring
equipment, the influence should be considered.
14) MOSA is effective to suppress the overvoltage of phase to the earth. In UHV systems, MOSA can also improve
the peak voltages of TRV in case of terminal faults and TLF.
Furthermore, this chapter summarizes the detailed circuit conditions when the maximum overvoltages occur and
shows the statistical evaluation of overvoltage occurrence. The useful information in this chapter contributes to the
precise overvoltage evaluation of UHV systems. Also, these advanced analytical techniques and the knowledge
are recommended to be added to IEC60071 Insulation co-ordination.
Page 169
TypethenameofTB
M inami-Iwaki S/S
s/s
Actual measurement
1.61 [p.u.] 3%
(with measurement error)
EMTP Analysis
1.58 [p.u.]
Page 170
TypethenameofTB
OF
C HINA ) [3]
Comparison between measured waveform and simulated waveform, shown in Figure 5.1.2 (a) and Figure 5.1.2 (b)
respectively, was carried out for switching overvoltage in the UHV transmission line shown in Figure 5.1.2. The line
is approximately 280km long. The simulated overvoltage waveforms are consistent with the measured ones. In the
calculation example, the maximum overvoltage is 959.9 kV, which is consistent with the measured result of
955.172 kV when switching at the same phase angle.
Figure 5.1.2 Measured and simulated waveforms at Nan Yang side during
energizing Nan-Jing TL from Jingmen (Maximum: (a) 955.2kV vs. (b) 959.9kV)
5.1.2.3 AEP (A MERICAN E LECTRIC P OWER ) [4-8]
Comparison between simulated waveform and measured waveform, shown in Figure 5.1.4(a), was carried out for
switching overvoltage in the 765kV transmission line shown in Figure 5.1.3. The line sections are 290km and
243km long. The simulated overvoltage waveforms by TNA and digital computer are 10 -15 % higher than
measured ones as shown Figure 5.1.4.
Page 171
TypethenameofTB
(a) Measured and simulated switching surges at Marysville (Energization from Dumont)
Figure 5.1.4 Measured and simulated switching surges at 765kV line (290km,
closing)
Page 172
TypethenameofTB
Figure 5.1.5 Comparison of field test and computer result [349 km, Reclosing A-B
in (a)]
5.1.2.5 EEA (T HE E GYPTIAN E LECTRICITY A UTHORITY ) [10]
Comparison between measured waveform and simulated waveform, shown in Figure 5.1.6 (a) and Figure 5.1.6 (b)
respectively was carried out for switching overvoltage in the 500 kV and 220 kV transmission line which are 209 km
and 40 km long respectively and shown in Figure 5.1.6(c). The simulated overvoltage waveforms by EMTP are
consistent with the measured ones as shown in Figure 5.1.6 and Table 5.1.1.
Page 173
TypethenameofTB
Table 5.1.1 Measured and calculated overvoltage factors for 220kV line
5.1.3 LIGHTNING OVERVOLTAGE
Page 174
TypethenameofTB
5.1.4 VFTO
Page 175
TypethenameofTB
Page 176
TypethenameofTB
b) Measured results
c) Simulated results
Page 177
TypethenameofTB
Figure 5.1.11 VFTO calculation and measurement when switching busbars with a
GIS DS as per IEC 62271-102, without pre-charging (left), with pre-charging (right)
5.1.5 SUMMARY
(1) The accuracy comparison of switching overvoltage between simulations and measurement, the TNA and digital
simulation cases are 10 - 15 % higher than measured ones in early 1970s. In the case of UHV line in Japan and
China, the accuracy was in the range 1 - 5 %. So, the maximum error between measurements and simulation is
about 5 %, which is acceptable in the design of transmission lines.
(2) There are so many simulation studies, but both simulation and measurement case are rare. In the case of UHV
line in Japan, the measurement and simulation waveform are identical in magnitude but wave front time are slightly
different. The lightning waveform and peak amplitude has statistical distributions, so it is necessary to use the
cumulative distribution data than single waveform.
(3) An accurate modeling of each individual GIS component makes it possible to reproduce VFTO waveforms with
a relatively high precision (differences lower than 5%). Nevertheless, variations of more than 10% depending on
the calculation methods are reported. Therefore, it is important to verify the simulation results by measurements.
Page 178
TypethenameofTB
References
[1] E. Zaima, et al., Insulation Coordination for UHV AC Systems Based on Surge Arrester Application (CIGRE
C4.306), IEC-CIGRE Second International Symposium on Standards for Ultra High Voltage Transmission, New
Delhi, 2009
[2] N. Ryu, et al., Comparison between EMTP Waveform and Field Observed Waveform of Switching Surge in the
UHV Designed 500kV Transmission Line, Proc. of the Eleventh Annual Conference of Power & Energy Society,
IEE of Japan, No.380, Sapporo, Japan, Aug. 2000 (in Japanese)
[3] ZHENG Gin, et al., Comparative Analysis on Simulation and Test of Overvoltages in China UHV AC Pilot
Project, 2009 International Conference on UHV Transmission, Beijing, 2009
[4] Barnes, H. C. "Preliminary Analysis of Extensive Switching Surge Testing of American Electric Power's First 765
kV Line and Stations," IEEE Trans. Power Apparatus And Systems, vol. PAS-90, March 1971
[5] Scherer, H. N., Vassell, G. S., "Experience with the AEP 765 kV System, Part I: Overview, 1972 IEEE Winter
Power Meeting, New York, N.Y., Publication No. 72 CH0645-2 PWR.
[6] Haas, R. J., et al, "Experience with the AEP 765 kV System Part IV, Overvoltage and Staged Fault Tests:
Instrumentation and Control," IEEE Paper Submitted for Presentation at the 1972 Winter Power Meeting.
[7] Garrity T, et al, "Experience with the AEP 765 kV System Part IV, Overvoltage and Staged Fault Tests:
Analysis," IEEE Paper Submitted for Presentation at the 1972 Winter Power Meeting.
[8] Fakheri (Keri), A, J., Haahr, J, "Experience with the AEP 765 kV System-Dumont-Marys.-Kammer Field Tests,"
IEEE Trans. Power Apparatus And Systems, July/Aug, 77.
[9] G. Catenaccl The Calculation Of Switching Surges III: Transmission Line Representation For Energization And
Re-Energization Studies With Complex Feeding Networks, GIGRE Working Group 13.05, Electra, No. 62 (1079)
[10] M.M. Awad, et al The Egyptian Experience With New Technologies For Field Transient Overvoltage
Measurement, GIGRE 33.203, 1996
[11] J. Takami, et al., Characteristics of Direct Lightning Strokes to Phase Conductors of UHV Transmission Lines,
IEEE Trans. Power Del., vol. 22, no. 1, pp. 537-546, Jan. 2007
[12] Y. Yamagata, et al., Very Fast Transients in 1000 kV Gas Insulated Switchgear, IEEE Transmission and
Distribution Conference, pp. 501-508 vol.2, Apr. 1999.
[13] A. Eriksson, et al Studies of Very Fast Transient in a 765 kV substation, CIGRE Working Group 34.12., report
No. 35 Monograph On Gis Very Fast Transients pp. 43 ~ pp.48
[14] S. Ogawa, et al., Estimation Of Restriking Transient Overvoltage On Disconnecting Switch For GIS, IEEE
PES 1985 Summer Meeting, 7 Vancouver, 85 SM 367-8, 1985
[15] U. Riechert, et al., Basic Information and Possible Counter Measures Concerning Very Fast Transients in
Gas-Insulated UHV Substations as Basis for the Insulation Co-ordination, CIGRE SC A2 & D1 Joint Colloquium
2011, KYOTO JAPAN, PS3-O-5, 2011
[16] M. Szewczyk, et al., Impact Of Disconnector Design On Very Fast Transient Overvoltage In Gas-Insulated
UHV Switchgear, XVII International Symposium on High Voltage Engineering, No.B-013, 2011
Page 179
TypethenameofTB
Figure 5.2.1 Concept and specification values of insulation co-ordination for UHV
systems in Japan [1]
These results greatly contributed to optimize the reduction of size of UHV equipments and the economical design
of UHV power transmission insulation (reduction of steel tower size, ground area for substation, etc.).
Figure 5.2.2 shows an example of the insulation design optimization based on MOSAs.
In order to realize low LIWV level of UHV systems, the higher performance MOSAs with low residual voltage are
considered as essential, since their protective levels are remarkably lower compared with the conventional MOSAs
according to existing standard, and which have been developed by applying and comparing the two technologies,
as shown in Figure 5.2.3.
Page 180
TypethenameofTB
Conventional
500kV MOSA
Page 181
TypethenameofTB
Main
performance
Protective performance
(Non-linearity)
Life
performance
Requirements
1.55
90 %
300 J/cc
(55MJ as for MOSA)
Page 182
TypethenameofTB
UHV
China
Japan
(1) Max. system voltage (kVrms)
1100
1100
(2) Peak value of max. line to ground voltage (kVp)
898 (1.00)
898 (1.00)
for transformer
2250 (2.51)
1950 (2.17)
(3) LIWV of system (kVp) (*)
for other equipments
2400 (2.67)
2250 (2.51)
for transformer
1800 (2.00)
1425 (1.59)
(4) SIWV of system (kVp) (*)
for other equipments
1800 (2.00)
1550 (1.73)
GIS/AIS
GIS
(porcelain)
(5) Nominal discharge current (kAp)
20
20
20
(6) Rated voltage of surge arrester (kVrms)
850
828
826
(7) Continuous operating voltage (COV) (kVrms)
693, (MCOV: 723)
638
635
(8) LIPL at 20 kAp
1700 (1.73)
1620 (1.80)
1620 (1.80)
(9) DLR(*) (8)/(6)
2.00
1.96
1.96
(10) SIPL at 2 kAp
1500 (1.53)
1460 (1.64)
(11) TOV (MJ)
55
20
55
(12) 2 shots total energy at operating duty test (MJ)
10+10 (margin)
40
40
(13) Line discharge class as per IEC
5
5
(14) Insulation withstand of LIWV
2400 (2.45)
2400 (2.67)
2250 (2.51)
housing (kVp)
SIWV
1800 (1.84)
1800 (2.00)
1550 (1.73)
(15) Short circuit current rating (kArms)
40
50
50
IEC 60099-4
GB 50260-1996
JEAG 5003-1998
(16) Mechanical requirement
0.3 G
0.3 G resonant 5
0.3 G resonant 3
cycles beat wave
cycles wave
less than 1.05
< 1.10 (GIS type)
< 1.10
(17) Voltage distribution factor
< 1.15 (porcelain
type)
Note: (*) It is reported that LIPL and SIPL values are determined by the simplified method for the Indian UHV System. The
values of SIWV and LIWV are obtained considering adequate margin over the values of SIPL and LIPL as obtained from V-I
characteristics of the higher performance surge arrester.
Metal-Oxide Surge Arrester (MOSA)
System
India
1200
980 (1.00)
2250 (2.30)
2400 (2.45)
1800 (1.84)
1800 (1.84)
AIS (polymer)
800 kV
System
Canada
(1) Max. system voltage (kVrms)
(2) Peak value of max. line to ground voltage (kVp)
for transformer
(3) LIWV of system (kVp)
for other equipments
for transformer
(4) SIWV of system (kVp)
for other equipments
(5) Nominal discharge current (kAp)
(6) Rated voltage of surge arrester (kVrms)
(7) Continuous operating voltage (COV) (kVrms)
(8) LIPL at 20 kAp
(9) DLR(*) (8)/(6)
(10) SIPL at 2 kAp
(11) TOV (MJ)
(12) 2 shots total energy at operating duty test (MJ)
(13) Line discharge class as per IEC
(14) Insulation withstand of LIWV
housing (kVp)
SIWV
(15) Short circuit current rating (kArms)
(16) Mechanical requirement
(17) Voltage distribution factor
(a)
765
625 (1.00)
1950 (3.12)
2100 (3.36)
1425 (2.28)
1425 (2.28)
AIS (porcelain)
20
550
442
1340 (2.14)
2.44
1080 (1.73)
700 kV 1 sec.
8.0
2100 (3.36)
1425 (2.28)
50
sin wave 5
vibration
0.23 G
-
Page 183
(b)
765
625 (1.00)
1950 (3.12)
2100 (3.36)
1425 (2.28)
1425 (2.28)
AIS (porcelain)
20
612
442
1560 (2.50)
2.55
1200 (1.92)
700 kV 1 sec.
8.0
2100 (3.36)
1425 (2.28)
50
sin wave 5
vibration
0.23 G
-
Brazil
800
653 (1.00)
2100 (2.34)
1550 (2.37)
AIS (porcelain)
20
588
470
1440 (2.21)
2.44
1210 (1.85)
2.9 (5 kJ 588)
N.A.
5
2100 (3.22)
1550 (2.37)
50
IEC 6099-4
< 1.10
TypethenameofTB
800 kV
System
Korea
(1) Max. system voltage (kVrms)
(2) Peak value of max. line to ground voltage (kVp)
for transformer
(3) LIWV of system (kVp)
for other equipments
for transformer
(4) SIWV of system (kVp)
for other equipments
(5) Nominal discharge current (kAp)
(6) Rated voltage of surge arrester (kVrms)
(7) Continuous operating voltage (COV) (kVrms)
(8) LIPL at 20 kAp
(9) DLR(*) (8)/(6)
(10) SIPL at 2 kAp
(11) TOV (MJ)
(12) 2 shots total energy at operating duty test (MJ)
(13) Line discharge class as per IEC
(14) Insulation withstand of LIWV
housing (kVp)
SIWV
(15) Short circuit current rating (kArms)
(16) Mechanical requirement
800
653 (1.00)
2250 (3.45)
1500 (2.37)
1425 (2.18)
AIS (porcelain)
20
588
462
1400 (2.14)
2.38
1300 (1.99)
6.6
4
2250 (3.45)
1425 (2.18)
50
JEAG
0.3 G
< 1.18
China
for Bus
800
653 (1.00)
2100 (3.22)
1550 (2.37)
GIS/AIS
20
600
462
1380 (2.11)
2.30
1142 (1.75)
18
5
2100 (3.22)
1550 (2.37)
63
JEAG
0.25 G
< 1.15
for Line
800
653 (1.00)
1950 (2.99)
1550 (2.37)
AIS
20
648
498
1491 (2.28)
2.30
1234 (1.89)
18
5
2100 (3.22)
1550 (2.37)
63
JEAG
0.25 G
< 1.15
Table 5.2.2 Main characteristics of systems and MOSAs for UHV (1200, 1100 kV)
and 800 kV systems [2-12]
Figure 5.2.5 shows the UHV higher performance MOSAs in test stations of India, China and Japan. The long-term
reliability of UHV higher performance MOSAs for GIS has been confirmed since 1996. The long term tests of the
polymer-housed and porcelain type MOSAs have just started and their reliability and stable performances will be
conformed in the near feature.
Page 184
TypethenameofTB
Figure 5.2.4 Representative GIS, Polymer-housed & Porcelain type UHV higher
performance MOSAs [1-3]
Figure 5.2.5 UHV higher performance MOSAs in test station of India, China and
Japan [1, 9]
Page 185
TypethenameofTB
In Japan, 550 kV higher performance MOSA, whose characteristics are shown in Table 5.2.3, has been adopted
since 1988. Figure 5.2.6 shows operating experience of the 550 kV higher performances MOSA in Japan. About
1900 units of the MOSAs have already been operated in the fields, and their reliability has been confirmed to be
very high.
The characteristics of MO element of 550 kV higher performance MOSA are same to those of 1100 kV higher
performance MOSA. The LIWV level of 550 kV GIS is reduced to 1425 kV by adopting the higher performance
MOSA in Japan as shown in Figure 5.2.2.
In addition, many higher performance MOSAs for the systems of 550 kV and above have also been used around
the world.
550 kV system
420 (1.00 p.u.)
870 at 10 kA
(2.07 p.u.)
1 block
1100 kV system
826 (1 p.u.)
1620 at 20 kA
(1.96 p.u.)
4 parallel blocks
2.3 2.8
--
Page 186
TypethenameofTB
5.2.3 SUMMARY
(1) For the insulation co-ordination studies, digital simulation packages, such as EMTP, have generally been used
since 1980s. The accuracy of simulations is mainly dependent on the simulation conditions, and may, in some
instances, not correlate with the measured results. Therefore, the simulation conditions are introduced in a
technical report on guidance for insulation co-ordination studies, IEC 60071-4 in detail [13]. It provides details and
examples of the subtleties associated with the representation of networks and equipment for various types of
analysis; lightning or switching transients and temporary overvoltages.
(2) In addition to the insulation coordination studies, the depth of energy analysis is considered as essential, in
order to achieve the desired accuracy for practical design of surge arrester of the complex and important network
system. The energy duties of switching surges and TOVs of UHV power transmission system should be analyzed
using EMTP program. Therefore, it is desirable that this analyzing measure, using EMTP program, is
recommended in the standard on application guide for insulation co-ordination, IEC 60071-2, as essential to UHV
power transmission system analysis. [14]
(3) The UHV higher performance MOSAs have a higher protective performance due to their lower residual voltage
level, which is a decisive factor for UHV power transmission insulation design. Therefore, the main characteristics
of UHV higher performance MOSAs, conforming to the special requirements for them, are recommended to be
introduced in the standard on application guide for insulation co-ordination, IEC 60071-2.
Page 187
TypethenameofTB
References
[1] T. Nakamura, T. Kobayashi, Y. Ishizaki, S. Watabiki, S. Ishibe and T. Kawamura: Insulation Co-ordination
based on 1100 kV High performance Surge Arresters, CIGRE SC C4 2009 Kushiro Colloquium
[2] T. Chiba, Y. Kayano, K. Hirai, et al.: Consideration Of Surge Arresters Used For The UHV Transmission
Systems In China, International Conference of UHV Power Transmission Technology, Oct., 2006, Beijing China,
pp989-995
[3] Y. Ishizaki, T. Kobayashi, M. Kan: The Consideration and Experiences of UHV Metal Oxide Surge Arresters
(MOSAs), CIGRE International Conference on Development of 1200 kV National Test Station, 29 &30 September
2010, New Delhi
[4] M. Reinhard, V. Hinrichsen, B. Richter and F. Greuter: Energy Handling Capability of High-Voltage Metal-Oxide
Surge Arresters Part 2: Results of Research Test Program CIGRE Aug. 2008, A3-309
[5] V. Hinrichsen, R. Gohler, M. Clemens, T. Steinmets and P. Riffon: External Grading System for UHV MetalOxide Surge Arresters A New Approach to Numerical Simulation and Dielectric Testing, CIGRE Aug. 2008, A3205
[6] IEC 60099-4: Ed.2.2, 2009-05, Surge Arresters Part4 Metal-Oxide Surge Arresters Without Gaps for a.c.
Systems
[7] Ning Qiu, Guangzhou Zhang, et al.: Environmental Considerations for UHV Substation, IEC/CIGRE UHV
Symposium Beijing 2007 Session 2-3-5, 2007
[8] Seo II, Especificao Tpica De Pra-Raios, DAQ.G/DEL.E Outubro 2007 Volume II Seo II 1/15,
FURNAS CENTRAIS ELTRICAS S.A
[9] LI Ming-gang, SONG Ji-jun, HAN Shu-mo, FANG Yu-ying: Development and Application of Tank Type Surge
Arrester for 1100kV System in China, UHV09_CP0410-EN, 2009 International Conference On UHV Transmission
(May 20-22 Beijing, China)
[10] WANG Bao-shan, XIONG Yi, TANG Lin, ZUO Zhong-qiu, WAN Ke, CHEN Li: The Summary And Conclusion
Of Chinese 1000 Kv Ultra High Voltage Porcelain Housed Metal Oxide Surge Arresters Development And Type
Test, UHV09_CP0616-EN, 2009 International Conference On UHV Transmission (May 20-22 BEIJING, CHINA)
[11] S. MaruyamaS. ShirakawaK. NakanoJ. Ozawa: Zinc Oxide Surge Arresters for the 735kV Power
Transmission LineCIGRE Open Conference on EHV Transmission Systems and Meeting Study Committees 22
and 23, 1987
[12] WANG Bao-shan, XIONG Yi, TANG Lin, ZUO Zhong-qui, WAN Ke and CHEN Li: The Summary And
Conclusion Of Chinese 1000 Kv Ultra High Voltage Porcelain Housed Metal Oxide Surge Arresters Development
And Type Test, 2009 International Conference On UHV Transmission (May 20-22 Beijing China)
[13] IEC TR 60071-4, Insulation Co-ordination Part 4: Computational Guide To Insulation Co-ordination And
Modeling Of Electrical Networks, 2004-06.
[14] IEC 60071-2, Insulation Co-ordination Part 2: Application Guide, 1996-12.
Page 188
TypethenameofTB
Page 189
TypethenameofTB
Figure 5.3.2 Evaluation method of lightning surge waveform with V-t curve
Page 190
TypethenameofTB
(Frequency: 0.5 - 5.0 MHz for GIS, 0.4 - 1.0 MHz for transformer)
(e) Waveform E: Double-frequency oscillatory waveform
(Lower frequency: Approx. 1.0 MHz, Upper frequency: Approx. 5.0 MHz for GIS)
Waveform B appears depending upon the positional relationship with protected points when the lightning arrester
operates, and waveforms C and D occur as a result of a negative and/or positive reflection inside and/or outside
the substation. Waveform A corresponds to the one when the decay of the DC component and/or oscillatory
component of waveforms B, C and D is large. Also, waveform E occurs during a re-arcing surge inside the
substation, as caused by a disconnector surge, when there exist two oscillation routes within the substation. Thus,
in general, it can be said that these waveforms are representative waveforms in view of the mechanism under
which they occur.
Page 191
TypethenameofTB
5.3.4.1 I NSULATION
CHARACTERISTICS OF
GIS
Figure 5.3.4 Examples of insulation characteristics with respect to the nonstandard waveform of the gap
Page 192
Voltage [pu]
TypethenameofTB
1.2
1.1
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
td1
Peak level
td2
80% level
0.5
1.5
2.5
Time [ s]
W aveform A
W aveform B
W aveform C
W aveform D
W aveform E
Standard lightning im pulse
(V min -0.8VLI,min)/VLI,min
0.1
0.1
10
100
Duration
[ s]
Duration
timetime
[s]
Figure 5.3.6 (V min -0.8V LI, min )/V LI, min characteristics as a function of duration time
over 80 % voltage level. (V min : minimum breakdown voltage at waveform A, B, C and
D, VLI, min: minimum breakdown voltage at standard lightning impulse waveform).
This characteristic corresponds to equations in the Calculation of conversion
factor K box in Figure 5.3. 7.
Page 193
TypethenameofTB
5.3.4.2 E VALUATION
METHOD FOR
GIS
The waveform was divided into elements, including each oscillatory wave and the flat part. It was then evaluated
based on the characteristics of section 5.3.4.1, and finally the crest value that was converted into an equivalent
standard lightning impulse waveform as the overall waveform was calculated. Figure 5.3.7 shows the waveform
evaluation flow for GIS.
Page 194
TypethenameofTB
5.3.4.3 A PPLICATION
TO A LIGHTNING OVERVOLTAGE AT
GIS
IN THE
UHV
SYSTEM
Figure 5.3.8 shows an example of the lightning overvoltage analysis waveform applied to GIS in a UHV substation.
This waveform has the shape of waveform B whose tail level is approximately 60% of the peak, after the steep
oscillatory surge has decayed.
Table 5.3.1 shows the results of analyzing and evaluating this waveform using the flow of Figure 5.3.5. In the case
of this waveform and = 0.60, the wave crest is severer than the wave tail, and the overall waveform is equivalent
to a 2094 kV standard lightning impulse waveform. In other words, the crest value was converted into a value that
was about 25 % lower.
Waveform
element
80 % duration
time
Peak voltage
(kV)
Conversion ratio
Equivalent standard
LI Vol.(kV)
B (wave crest)
0.16s
2617
1.25 (1.30)
2094 [=0.60]
2016 [=0.45]
B (wave tail)
1517
0.90
1686
Table 5.3.1 Evaluation of the lightning surge waveform in the GIS of a UHV system
To think of the principle of construction, this method can be generally applicable to lower voltage classes as well as
550 kV and UHV. In fact, it has been put into practical use in 66 kV to UHV voltage classes in some electric power
companies.
Page 195
TypethenameofTB
5.3.5.1 I NSULATION
IMPULSES
250
-Oscillation frequency: 1
MHz
-Half-amplitude cycle
count: 1 cycle
400
Oil gap
A-waveform
(0.84/1.56 s)
300
Voltage
kV(kV)
Voltage
(kV)
kV
200
200
150
0.841.56
50
100
2
3
Time
(s)
: Breakdown
:1MHz,
:
Partial
:1
discharge
inception
100
10
Time
(s)
Page 196
TypethenameofTB
Voltage [pu]
td1
Peak level
90% level
td2
80% level
td1
0
0.5
1.5
2.5
Time [ s]
The waveform is divided into elements including each oscillatory wave and the flat part. It is then evaluated based
on the characteristics of Section 5.3.5.1, and finally the crest value that is converted into an equivalent standard
lightning impulse waveform as the overall waveform is calculated. Figure 5.3.12 presents the waveform evaluation
flow for an oil-filled transformer.
Page 197
TypethenameofTB
Overvoltage waveform
Resolving into
element waveforms
Pulse-like
(Waveform A and pulse
part of waveform B)
Oscillatory
Flat part
(Waveforms C and D)
Overall evaluation
V = Max [Vsij] i,j
If the voltage exceeds the crest value as a result of conversion
of the flat part, the original crest value is used.
500 K V
SYSTEM
Figure 5.3.13 is an example of a lightning surge analysis waveform applied to a transformer in a 500 kV substation.
In this waveform, the wave tail maintains a level of about 70 % with respect to the peak after the steep oscillatory
surge has decayed.
Page 198
TypethenameofTB
Table 5.3.2 summarizes the results of analyzing and evaluating this waveform using the flow of Figure 5.3.12. The
severest part of the waveform is the pulse-shaped one No.2 with respect to the turn-to-turn insulation. The overall
waveform is dominated by this part, and is equivalent to a 934 kV standard lightning impulse waveform. The crest
value was converted into a value that was about 24 % lower.
Insulation
element
Waveform
element
Oil gap /
Section-tosection
Oscillatory parts
(1~4): C or D
Flat part (5)
Max pulse (2)
Other pulses (1)
Flat part (5)
Turn-to-turn
Conversion
ratio
Equivalent
standard LI
Vol.(kV)
1158
1.40
827
784
1158
1084
784
0.90
1.24
1.24
0.95
871
934
874
825
Duration
time (s)
Peak
voltage (kV)
0.84
0.35
0.35
-
5.3.6 SUMMARY
Field real overvoltages of non-standard lightning impulse waveform are analyzed and the insulating characteristics
of SF6 gas and oil-filled transformer elements for these actual overvoltages are clarified to convert the waveform
into the standard lightning impulse waveform. It was found that the insulation requirements could not be as severe
as those of the standard lightning impulse waveform, since the decay of the field overvoltage is generally large.
Consequently, it could be possible in some cases to use lower withstand test voltages for GIS using SF6 and oilfilled transformers. The proposed conversion method for non-standard lightning impulse waveform is applicable up
to 1100 kV UHV system.
Page 199
TypethenameofTB
References
[1] IEC 60071-1, Insulation Co-Ordination Part 1: Definitions, Principles And Rules, 2006-01.
[2] IEC 60071-2, Insulation Co-Ordination Part 2: Application Guide, 1996-12.
[3] CIGRE WG.C4.302, Insulation Co-Ordination Related To Internal Insulation Of Gas Insulated System In Sf6
And N2/Sf6 Gas Mixtures Under Ac Conditions, CIGRE Technical Brochure No.360, 2008-10.
[4] S. Okabe, S. Yuasa, M. Koto and E. Zaima, Evaluation Of Lightning Surge Waveform For LIWV Reduction Of
Substation Equipment 13th ISH P.05.66, 2003.
[5] B. Wahlstrm, T. Yamagiwa, E. Colombo, F. Gallon, Possible Applications And Benefits Expected Through
Reduced Insulation Co-Ordination Voltages, CIGRE SC A3&B3 Joint Tutorial in Tokyo, 2005.
5.4 Conclusions
(1) In the analysis of power system, digital simulations such as EMTP have been used in many countries. In
addition, the comparison estimation is also performed the analysis results and the measured value in each country,
then, its validity has been confirmed up to UHV system. For example, in the case of UHV line in Japan and China,
the maximum error between measurements and simulation for switching overvoltages is about 5 %, which is
agreeable in the design of transmission lines.
(2) The main characteristics of MOSAs for the UHV projects in the world have been investigated in the WG. The
UHV higher performance MOSAs have a higher protective performance due to their lower residual voltage level,
which is a decisive factor for UHV power transmission insulation design. To illustrate the characteristics of the
higher performance MOSA is desirable to IEC60071-2.
In the UHV system is required to evaluate the capability (switching overvoltage and TOV) of the arrester by a
detailed analysis using EMTP, in IEC 60071-2, it is desirable to mention the need for tolerance evaluation.
(3) Field real overvoltages of non-standard lightning impulse waveform are analyzed and the insulating
characteristics of SF6 gas and oil-filled transformer elements for these actual surges are clarified to convert the
surge waveform into the standard lightning impulse waveform. Non-standard lightning impulse waveform evaluation
method shown in section 5.3 has already been put into practical use to UHV and could be possible in some cases
to use lower withstand voltages. The evaluation method can be reflected in the IEC60071-2 as specific method.
Page 200
TypethenameofTB
8000
7000
6000
5000
Switching impulse
withstand
4000
3000
2000
1000
0
0
10
15
20
25
Figure 6.1.1 Lightning and switching impulse withstand versus gap distance for
rod-rod gap
The historical background to this subject is interesting. The first EHV system was the 400 kV system developed in
Sweden in the mid 1950s. No account was taken of switching overvoltages because their influence was unknown
and obviously no mitigation was considered. The development of the 400 kV system in Italy in the 1960s was the
first real recognition of switching overvoltages. A very large number of switching impulse tests were performed and
resulted in the now well-known U-curves and the notion of gap factors [1, 2]. The explanation for the U-curve was
provided by research work at EdF [3]. Switching impulse breakdown involves both streamer and leader
development (lightning impulse breakdown involves streamers only) with the latter being the main driver. The
minimum value of the U-curves thus represents optimal leader development and occurs at front-times in the range
of 100 to 400 s leading to a selection of a front-time of 250 s for standardization purposes. Detailed descriptions
of electrical discharges and breakdown in air can be found in [4, 5].
A number of switching overvoltage mitigation measures have been proposed or used starting in the 1960s:
Fast insertion of shunt reactors [6].
Closing resistors [6, 7, 8, 9].
Staggered pole closing [10, 17, 18, 20].
Line terminal arresters [10, 11, 12, 13, 14].
Page 201
TypethenameofTB
UD Z
RZ
2U D Z
UTC
RZ
4Z
1
RZ
U RD R
RZ
where URD is the voltage across the resistor at the instant of bypassing.
At the line end with voltage doubling, the voltage is:
URe L
2U RD R
U pf
RZ
where Upf is the frequency voltage at the instant of travelling wave arrival.
Page 202
TypethenameofTB
An upper limit on the value of R is therefore required. The value of R is usually optimized through computer studies
and is typically in the range 250 to 600 dependent on the application.
Generally, the aim with closing resistors is to limit the voltage at the receiving end of the line to 2 p.u. However,
closing resistors do little for the voltage profile along the line and higher overvoltages than 2 p.u. at points where
incident and reflected waves cumulatively coincide.
Page 203
TypethenameofTB
Page 204
TypethenameofTB
4.5
4
3.5
No overvoltage control
measures
2.5
2
1.5
1
0
20
40
60
80
100
Page 205
TypethenameofTB
2.5
60 Hz
Difference 60 vs 33 Hz
2
Difference 60 vs 45 Hz
1.5
Voltage (pu)
0.5
0
0
20
40
60
80
100
120
140
160
-0.5
-1
-1.5
-2
-2.5
Time (ms)
Page 206
180
TypethenameofTB
It is clear that this mitigation measure requires a dynamic controller which can analyze the differential voltage
across the circuit breaker, locate the minimums, predict the future minimums and close the breaker accordingly
when the close signal is applied, all within 0.5 s or perhaps less. The sophistication of the control device obviously
goes far beyond the zero crossing detectors and sequence timers available when this measure was first proposed.
As with controlled capacitor bank switching, the rate of decay of dielectric strength (RDDS) of the circuit breaker is
a factor in the application of this measure.
This measure combined with line end and mid-line 372 kV rated arresters has been successfully in-service at 500
kV on the BC Hydro system since 1995 [20, 21]. The intent was to limit switching overvoltages to 1.7 p.u.
90.0
Staggered pole closing
80.0
70.0
Closing resistors
60.0
50.0
40.0
30.0
20.0
10.0
0.0
1.0
1.5
2.0
2.5
3.0
Page 207
TypethenameofTB
3.5
3
No control measures
Overvoltgae (pu)
2.5
1.5
1
0
10
11
12
Figure 6.7.2 Overvoltage profile along 330 km long transposed 500 kV line for
various overvoltage limitation options
Circuit breakers applied at 1000 kV and 1200 kV will incorporate opening resistors in order to meet transient
recovery voltage (TRV) requirements during fault clearing. The resistors, though probably rated resistance-wise for
opening, can double as closing resistors. However, as shown in Figure 6.7.1 and Figure 6.7.2, the lower the
resistance value the greater the overvoltage limitation value. The choice between the closing resistor versus
closing controller option is a decision for the user and a comparison is given in Table 6.7.1.
Attribute
Proven technology
Complexity
Location
Maintainability
Closing resistor
Yes, in use since 1960
High; multiple moving mechanical
parts
At line potential
Low; requires circuit breaker outage
Provision of spares
Future improvement
potential
Limited if at all
Other
rather
than
Closing controller
Yes, in use since early 1990s
Low; no moving parts, circuit
boards and associated software
only
At ground level in control room
High; does not require circuit
breaker outage
Yes; complete module or circuit
boards
Yes, component and software
advances
-
Page 208
TypethenameofTB
6.8 Conclusions
The above discussion and study results for a transposed 500 kV transmission line illustrate the combined
contribution of line end arresters and closing resistors or a controller to switching overvoltage limitation. The results
can certainly be viewed as indicative of performance but, to determine absolute design parameters, the user needs
to conduct studies focused on the actual application details and considerations. Accurate representation of arrester
and controller characteristics, resistor insertion and line configuration is essential in order to achieve a valid design
basis for lines meeting dependability requirements.
References
[1] G. Carrara, "Investigation on Impulse Sparkover Characteristics of Long Rod/rod and Rod/plane Air Gaps".
Cigre Report No. 328, 1964.
[2] L. Paris and R. Cortina, "Switching and Lightning Impulse Characteristics of Large Air Gaps and Long Insulator
Strings". IEEE Transactions on Power Apparatus and Systems No. 87, 1968.
[3] J. N. Ross et al, "Positive Discharges in Long Air Gap at Les Renardieres." Electra No. 53. 1977.
[4] N. L. Allen, "Mechanism of Air Breakdown." Chapter 1, Advances in High Voltage Engineering (Book), IEE
Press 2004.
[5] V. Cooray, "Mechanism of Electrical Discharges." Chapter 3, The Lightning Flash (Book), IEE Press 2003.
[6] H. B. Thoren, "Reduction of Switching Overvoltages in EHV and UHV Systems." IEEE Trans. PAS-90, 1971.
[7] C. L. Wagner and J.W. Bankoske, "Evaluation of Surge Suppression Resistors in High-Voltage Circuit
Breakers." IEEE Trans. PAS-86, 1967.
[8] P. A. Baltensperger and P. Djurdjevic, "Damping of Switching Overvoltages in EHV Networks New Economic
Aspects and Solutions." IEEE Trans. PAS-88, 1969.
[9] R. G. Colclasser, C.L. Wagner and E.P. Donohue, "Multistep Resistor Control of Switching Surges." IEEE Trans.
PAS-88, 1969.
[10] A. C. Legate, J.H. Brunke, J.J. Ray and E.J. Yasuda, "Elimination of Closing Resistors on EHV Circuit
Breakers." IEEE Trans. PD, Vol. 3, 1988.
[11] J. R. Ribeiro and M.E. McCallum, "An Application of Metal Oxide Surge Arresters in the Elimination of Need for
Closing Resistors in EHV Circuit Breakers." IEEE Trans. PD, Vol. 4, 1989.
[12] J. K. Blakow and T.L. Weaver, "Switching Surge Control for the 500 kV California-Oregon Transmission
Project." Cigre Report No. 13-304, 1990.
[13] A. Eriksson, J. Grandl and O. Knudsen, "Optimized Line Switching Surge Control Using Circuit-Breakers
without Closing Resistors." Cigre Report No. 13-305.
[14] Y. I. Musa, A. J. F. Keri, J. A. Halladay, A. S. Jagtiani, J. D. Manderville, B. Johnnerfelt, L. Stenstrom, A. H.
Khan and W. B. Freeman, "Application of 800 kV SF6 Dead Tank Circuit Breaker with Transmission Line Surge
Arrester to Control Switching Transient Overvoltages." IEEE Trans. PD, Vol. 17, 2002.
[15] E. Maury, "Synchronous Closing of 500 and 765 kV Circuit Breakers: A Means of Reducing Switching Surges
on Unloaded Lines." Cigre Report No. 143, 1966.
[16] H. E. Konkel, A. C. Legate and H. C. Ramberg, "Limiting Switching Surge Overvoltages with Conventional
Power Circuit Breakers". IEEE Trans. PAS-96, 1977.
Page 209
TypethenameofTB
[17] B. Avent and J. Sawada, "BC Hydro's Experience with Controlled Circuit Breaker Closing on a 500 kV Line."
Canadian Electrical Association, Engineering and Operating Division Meeting, March 1995.
[18] K. Froehlich, C. Hoelzl, A. C. Carvalho and W. Hofbauer, "Transmission Line Controlled Switching." Canadian
Electrical Association, Engineering and Operating Division Meeting, March 1995.
[19] A. H. Khan, D. S. Johnson, J. H. Brunke and D. L. Goldsworthy, "Synchronous Closing Application in Utility
Transmission Systems." Cigre Report No. 13-306, 1996.
[20] K. Froehlich, C. Hoelzl, M. Stanek, A. C. Carvalho, W. Hofbauer, P. Hoegg, B. L. Avent, D. F. Peelo and J. H.
Sawada, "Controlled Closing on Shunt Reactor Compensated Transmission Lines: Part 1 Closing Control Device
Development and Part 2 Application of Closing Control Device for High-Speed Autoreclosing on BC Hydro 500 kV
Transmission Line." IEEE Trans. PD, Vol. 12, 1997.
[21] B. L. Avent, D. F. Peelo and J. Sawada, "Application of 500 kV Circuit Breakers on Transmission Line with
MOV Protected Series Capacitor Bank." Cigre Report No. 13-107, 2002.
Page 210
TypethenameofTB
7.2 Air Gap Clearances chosen for UHV Projects in different countries
The air gap clearance values for phase-to-phase as well as phase-to-structure as reported by various countries for
highest voltage level 1000kV and above are given in [18]. For the sake of comparison the values of IEC 60071-1,
Annex A have also been enclosed.
Page 211
TypethenameofTB
Japan
China
Russia
India
1050
1100
1100
1200
1200
SIWV(Phase-earth) (kV)
1675
1550
1800
1800
1800
9.5
7.5 - 8.5
7.5
7.5 - 9.7
6.3 -8.3
12
10.5-11.5
11.3
11.4 - 12.4
10.3 - 12.3
Minimum clearances for the chosen SIWV according to IEC 60071-1 Amd. 1
SIWV(Phase-earth) (kV)
Minimum phase-to- earth (m)
(Conductor-structure /
Rod-structure )
Minimum phase-to-phase (m)
(Conductor-conductor parallel /
Rod-conductor )
1675
1550
1800
1800
1800
5.6 / 7.4
4.9 / 6.4
6.3 / 8.3
6.3 / 8.3
6.3 / 8.3
9.1 / 10.9
(R=1.65)*
8.4 / 10
(R=1.7)*
9.8 / 11.6
(R=1.6)*
10.3 / 12.3
(R=1.65)*
10.3 / 12.3
(R=1.65)*
Table 7.2.1 Comparison of air gap clearance values for different countries
and IEC 60071-1 Ed. 8.1
The values are as per the enquiry sheets submitted by different countries to WG.A3.22 and given in TB 362, page
89 [18]. It may be observed that the air gap clearance data considered by India is same as that of IEC. The
Chinese and Russian data are also quite close to IEC data. Meanwhile, the Italian and Japanese data for both
phase to structure and phase to phase are much higher than the corresponding IEC data.
Page 212
TypethenameofTB
In IEC 60071-1 Annex A the minimum air gap clearances for rod-structure and conductor-structure are given. The
values of the UHV systems have been extracted and are shown in Table 7.3.1and Table 7.3.2 for phase-to-earth
insulation and phase-to-phase insulation respectively.
Minimum phase-to-earth
mm
Rod-structure
6400
7400*
8300*
9500*
Conductor-structure
4900
5600*
6300*
7200*
* under consideration
Rod-conductor
7600
8400*
9100*
9600*
9800*
10300*
11200*
9400
10000*
10900*
11400*
11600*
12300*
13300*
* under consideration
Page 213
TypethenameofTB
Phase-to-Earth Configuration
Under standard atmospheric condition the 50% positive (most critical) flashover voltage U50 of air gaps without
insulators is given by the following equations:
U50 = K 1080 ln(0.46 d + 1)
(1)
(2)
Page 214
TypethenameofTB
(3)
with the per unit value of the conventional deviation (p.u. of U50).
Unless otherwise specified by the relevant Technical Committee , the per unit value of the conventional deviation,
may be assumed conservatively as 0.06 for switching impulse tests on air insulation, without any other insulation
involved.
Phase-to-Phase Configuration
In this case also, the gap factor approach may be used however the gap factor is influenced not only by the gap
configuration but also by the following ratio which describes the balance between the two voltage values on the
two phases:
+
= U / (U + U )
Where, U+ is the maximum positive voltage applied to one electrode and U- is the corresponding value of negative
voltage applied at the opposite electrode.
7.4.1 DERIVATION OF IEC 60071-1 ANNEX A ED 8.1 UHV MINIMUM AIR GAP
CLEARANCES FOR SUBSTATIONS
To derive the UHV minimum-air-gap clearance-values of the IEC tables the following calculations and comparisons
were done by the IEC Maintenance Team in charge of.
With fixation of K and the gap distances can be calculated for standard switching impulse voltage with a 10%
flashover probability (according standard withstand voltage).
Following equation 2 and 3 will lead to equation 4 and the UHV minimum air clearances are then given by:
d = [SIWV / (500 (1-1.3 ) K)]1/0.6 (m, kV)
d = [(1.0846 SIWV) / (500 K)]1/0.6 1000 (mm, kV)
(4)
with an average gap factor K derived for each IEC structure from lower SIWV for which clearances are already
given in the IEC 60071-1:
- for phase to earth: Rod-structure 1.095; conductor.-structure 1.297
- for phase to phase: Conductor-conductor 1.5918; rod-conductor 1.432
- = 6%
With the Equation (4) and above average values of gap factor (K), the minimum phase-to-earth and phase-tophase air clearances are determined for the UHV switching impulse levels. The values thus obtained are compared
with the values given in the tables of IEC 60071-1 (Table 7.4.1 and Table 7.4.2).
Page 215
TypethenameofTB
Standard rated
switching impulse
withstand voltage
kV
Rod-structure
Conductor-structure
Rod-structure
Conductor-structure
1550
6487
4892
6400
4900
1675
7382
5567
7400
5600
1800
8323
6277
8300
6300
1950
9510
7172
9500
7200
Phaseto-earth
kV
1550
1550
1675
1675
Ratio:
Phase-to-phase /
Phase-to-earth
1.6
1.7
1.65
1.7
Phase-tophase
kV
2480
2635
2764
2848
1800
1.6
2880
9773
11648
9800
11600
1800
1950
1.65
1.6
2970
3120
10287
11168
12261
13310
10300
11200
12300
13300
Page 216
TypethenameofTB
Gap Type
Conductor Crossarm
Conductor Window
Typical Range
1.36 1.58
1.22 1.32
1.18 -1.35
1.28 1.63
1.03 1.66
Reference Value
1.45
1.25
Conductor-Plane : 1.15
Conductor-Rod : 1.47
1.45
1.35
Table 7.4.3 Typical gap factors ( K ) for switching impulse breakdown phase-to-earth
(Summary of Table G.1 of IEC 60071-2)
= 0.5
1.80
1.65
1.62
1.50
1.45
Configuration
Ring-ring or large smooth electrodes
Crossed conductors
Rod rod or conductor-conductor (along the span)
Supported busbars (fittings)
Asymmetrical geometrics
= 0.33
1.70
1.53
1.52
1.40
1.36
1675 kV
1800kV
1950 kV
From To
Eq-1
Eq-2
Eq-1
Eq-2
Eq-1
Eq-2
Eq-1
Eq-2
1.15 - 1.2
461
409
555
466
664
525
817
600
1.2 - 1.25
402
366
482
417
573
470
702
537
1.25 - 1.3
353
329
421
375
499
422
609
483
1.3 - 1.35
312
297
371
338
438
381
532
436
1.35 - 1.4
278
269
329
306
387
345
468
395
1.4 - 1.45
249
245
293
278
344
314
415
359
1.45 - 1.5
224
223
263
254
308
286
369
327
1.5 - 1.55
202
204
237
232
277
262
331
299
1.55 - 1.6
183
187
215
213
250
240
298
275
1.6 - 1.65
167
172
195
196
226
221
269
253
1.65 - 1.7
153
159
178
181
206
204
244
233
1.7 - 1.75
140
147
163
167
188
189
223
215
1.75 - 1.8
129
136
150
155
173
175
204
200
1.8 - 1.85
119
126
138
144
159
162
187
185
1.85 - 1.9
110
118
128
134
147
151
172
173
Page 217
TypethenameofTB
14000
1950 kV - IEC equation 2
1950 kV - IEC equation 1
1800 kV - IEC equation 2
1800 kV - IEC equation 1
1675 kV - IEC equation 2
1675 kV - IEC equation 1
1550 kV - IEC equation 2
1550 kV - IEC equation 1
12000
10000
8000
6000
4000
2000
1,8
1,75
1,7
1,65
1,6
1,55
1,5
1,45
1,4
1,35
1,3
1,25
1,2
1,15
1,1
1,05
gap factor K
Figure 7.4.1 Dependency of gap clearance for gap factor variations in steps of 0.05
Representative for all SIWV values the sensitivity due to gap factor variations (0,05 steps of K) according Table
7.4.5 is plotted for 1800 kV in Figure 7.4.2. Here the sensitivity due to the 0,05 steps are related to the absolute
value of gap clearance shown in Figure 7.4.1. The related sensitivity due to the gap factor steps are in the range of
5% to 8% for the calculations with 1800 kV (Figure 7.4.2). The difference between low and high gap factor values is
only 4% for standard impulses and 6% for critical wave shapes.
12%
error [%] - IEC equation 1
error [%] - IEC equation 2
10%
8%
6%
4%
2%
1,85
1,8
1,75
1,7
1,65
1,6
1,55
1,5
1,45
1,4
1,35
1,3
1,25
1,2
1,15
1,1
1,05
0%
gap factor K
Figure 7.4.2 Sensitivity of gap clearance differences for gap factor variations
(0.05 steps) related to the absolute gap clearance
(Calculated for 1800 kV only)
Page 218
TypethenameofTB
0,06
=
deviation
0,03
= 3% deviation
6% deviation
12000
IEC equation 2
IEC equation 1
10000
8000
5796
6000
6035
5733
5526
5329
5157
4000
2000
1,8
1,7
1,6
1,5
1,4
1,3
1,2
1,1
1,8
1,7
1,6
1,5
1,4
1,3
1,2
1,1
0
gap factor K
Figure 7.4.3 Dependency of gap-distance from per unit values of the conventional
deviation, for calculations with 1800 kV SIWV
Page 219
TypethenameofTB
Figure 7.5.1 50% flashover voltage curves for rod-plane gaps of various lengths d
under positive polarity impulses [4]
Figure 7.5.2 Influence of front-time on 50% positive polarity slow front impulse
flashover voltage of different rod-plane gaps [5]
Page 220
TypethenameofTB
Figure 7.5.3 Influence of front-time on 50% positive polarity slow front impulse
flashover voltage of rod-plane gap [6]
For Rod-plane gap Watanabe results [6] show that critical time to peak is following Tpcr = 55 d (s, m) (Figure
7.5.3), values very similar to Cortina results Tpcr = 50 d (with the per unit value of the conventional deviation =
5%) but Cortina results for conductor plane (more actual substation and OVHL configurations) shows Tpcr = 35 d
(s, m) and only = 3% for conductor length of 50 m [7].
According to Figure 7.5.4, the phase to phase critical time to peak Tpcr of positive polarity slow front impulse
flashover voltage of parallel conductor gap is within 200 s and 300 s. Thus for phase to phase the critical time to
peak coincides with the time to peak of the standard switching impulse voltage: Tcr = 250 s. For such insulations
the distances proposed presently for UHV by IEC 60071-1 Annex A Ed 8.1 should be largely sufficient.
Figure 7.5.4 Influence of front-time on 50% positive polarity slow front impulse
flashover voltage of parallel conductor gap [6]
Page 221
TypethenameofTB
Dependence of the flashover voltage on the impulse shape and gap distance has been investigated also with
reference to rod-rod and conductor-rod configurations. Examples of U - curves are reported in Figure 7.5.5 The
U50 versus gap length characteristics are given in Figure 7.5.6 for rod-rod gap with the comparison of critical to
standard impulse shape. In these figures the numbers within brackets indicate the percentage of discharges to the
earth plane with respect to the total number of discharges. For rod-rod gap the per unit value of the conventional
deviation is 7% - 8%, but for conductor-rod gap is 4%.
Figure 7.5.5 Influence of front-time on 50% positive polarity slow front impulse
flashover voltage of different rod-plane and conductor-rod gaps (with in brackets
the percentage of discharges to the earth plane with respect to he total number of
discharges [7]
Figure 7.5.6 50% positive polarity slow front impulse flashover voltage of rod-rod
for critical time to peak and standard switching impulse with different ratios
"height of the earthed rod to the gap spacing".(With in brackets the percentage of
discharges to the earth plane with respect to the total number of discharges) [7]
Page 222
TypethenameofTB
For UHV tower windows with "V-insulator strings experimental data of the critical time to peak are given in Figure
7.5.7 [4]. The critical time to peak Tpcr is within 200 s and 400 s and the needed air gap spacing is within 8 m and
9 m for the UHV SIWV range.
Figure 7.5.7 50% flashover voltage Curves for tower windows with "V-insulator
strings" of various spacing d under positive polarity impulses [4]
Cortina results [7] show that for conductor-tower arrangements (conductor cross arm, conductor-window, conductor
tower leg in a circus arrangement, see Figure 7.5.8) up to 10 m the critical time to peak is following Tpcr = 35 d (s,
m) and that for positive polarity no influence of insulator strings on the flashover voltage was observed.
Page 223
TypethenameofTB
Figure 7.5.9 50% flashover voltage curves versus the time to peak of the applied
impulse of conductor-structure configurations with post insulators
(six units of 1.33 m each) [7]
Page 224
TypethenameofTB
Regarding insulators Figure 7.5.10 shows additionally that the influence is quite negligible with positive polarity,
while it becomes significant in case of negative polarity, being the larger the shorter the length of the earthed rod
[7].
Page 225
TypethenameofTB
Gap:
10 m
Sphere: 0,75 m radius
Cone: 10 mm tip
Figure 7.5.11 Dependence of U50 from the time to peak for 10 m gap with the
electrode tips of a cone (10 mm Tipp) and a sphere (with 0.75 m diameter) [17]
Page 226
TypethenameofTB
(ii)
Page 227
TypethenameofTB
Figure 7.6.1 A view of mock up tower test facility (left) and a view of the insulator
strings (right)
7.6.1.1 TESTS WITH TOP AND SIDE TOWER
The conductor (eight BERSIMIS bundle with 35 m length) was strung between the end towers of the mock up tower.
The top and side tower was set up with MS and Aluminum channels, with a width of 2 m and a length of 18 m for
the top tower while a width of 5 m for the side tower. The minimum height of the bundle conductor above earth was
kept at about 20 m to avoid any disruptive discharges to earth or any other earthed object other than the top and
side tower. The arrangement for the setup is shown in Figure 7.6.2. The gap between the conductor-side tower and
conductor-top tower was maintained the same. The experiment was carried out with gap clearance 7 m and 10 m
under dry conditions and 50% flashover voltage were obtained with switching impulse voltage of positive polarity
and different times to peak in order to obtain critical time to peak value (Tpcr) for which the 50% flashover voltage is
minimum. up and down method as given in IEC-60060-1 with total number of impulses more than 20 and using one
level of test voltages in steps of less than 3% is used. Four wave shapes with different time to peak including
standard switching impulse voltage wave shape of 250/2500 s were used. Time to half value or Tail time during all
these tests was around 2500 to 2700 s.
18m
5m
2m
Y = X = 7 to 10 m
18m
Conductor
Bundle
20m
Ground
Page 228
TypethenameofTB
The 50% flashover voltage obtained during the test was corrected for reference atmospheric conditions considering
the atmospheric correction factor as per IEC 60060-1. Table 7.6.1 gives the corrected value of 50% flashover
voltage corresponding to different peak time (Tp) of the wave shape applied for 7 m and 10 m gap clearance. The
corrected value of 50% flashover voltage plotted against time to peak (Tp) for the gap distance 10 m and 7 m is
shown at Figure 7.6.3. The result for the per unit value of the conventional deviation was = 0,03.
Gap
Distance
(m)
10
Time
to
peak
Tp
(s)
Temp.
Dry
( C )
Temp.
Wet
( C )
Atm.
Pressure
(mm Hg.)
50%
Flashover
Voltage
U50%
(kV)
Atm.
Corr.
Factor
( Kt )
315
250
220
150
315
250
220
150
29
30
31
30
29
28
31
32
22
22
23
22
21
20
22
21
716
715
716
715
717
715
719
717
2552
2489
2447
2550
2125
2079
2059
2131
0.992
0.991
0.993
0.994
0.984
0.982
0.985
0.927
50%
Flashover
Voltage after
atm.
correction
U50% (kV)
2572
2511
2463
2565
2159
2117
2090
2185
Table 7.6.1 50% flashover voltage with Tower top and one side representation for
10 m and 7 m gap clearance
Figure 7.6.3 50% flashover voltage plotted as a function of Tp for the gaps of 10 m
and 7 m for Top and Side Tower Configuration
From the test results, it is seen that the critical time-to-peak for which the minimal 50% flashover voltages occurred
is between 220 to 230 microseconds, which is slightly lower than the 50% flashover voltage at standard peak time
of 250 s. The lowest value of 50% flashover voltages obtained was 2090 kV for an air gap clearance of 7 m. The
Page 229
TypethenameofTB
withstand voltage is determined as U10 = U50 (1 1.3 ) = 2008 kV where the per unit value of the conventional
deviation, has been considered as 0.03. Even for = 0.06 (IEC 60071-2), the calculated value is 1927 kV. This
value is much higher than the required switching impulse withstand voltage of 1800 kV.
(m)
8
8.5
Time to
peak
(Tp)
Temp.
Dry
bulb
Temp.
Wet
bulb
(s)
160
250
270
350
115
250
270
350
( C )
22
27
27
24
29
25
25
29
( C )
16
17
17
17
19
19
19
20
Atm.
Pressure
(mm.
of Hg.)
718
713
713
718
716
716
716
717
50%
Flashover
Voltage
Atm.
Corr.
Factor
(kV)
1926
1879
1868
1854
2191
1950
1937
1948
(Kt)
0.987
0.94
0.94
0.982
0.98
0.992
0.992
0.988
50% Flashover
Voltage
(corrected)
U50
(kV)
1951
1915
1906
1910
2236
1963
1953
1961
Table 7.6.2 50% flashover voltages as against air gap clearance of 8 m and 8.5 m
Page 230
TypethenameofTB
Figure 7.6.5 shows the 50% Flashover voltage, corrected for atmospheric conditions, plotted as a function of Tp for
conductor-insulator string-tower window configuration and the gap distances 8 m and 8.5 m. The flashover voltages
were minimum for times to peak of 270 s and above.
Figure 7.6.5 50% flashover voltage plotted as a function of Tp for the gaps of 8 m
and 8.5 m conductor-insulator string-tower window configuration
From the experimental results for window clearance of 8 m the 50% flashover voltage obtained was 1906 kV for
which the withstand voltage determined as U10 = U50 (1 1.3 ), for = 0.03 will be 1832 kV. This level is just
about 1.75% higher than the 1800 kV rated standard Switching Impulse Withstand Voltage.
Page 231
TypethenameofTB
Page 232
TypethenameofTB
Figure 7.6.6 Arrangement of side phase air-gap test with I insulator string
Figure 7.6.7 Typical flashover picture for side phase air-gap test with
UHV cathead-tower
Page 233
TypethenameofTB
In the power frequency test, the gap distances between conductor bundle and tower were varied from 2.0 m to 4.5
m and the windage yaw angle of I-insulator string was fixed to 48. The height h of the conductor was 42 m. The
tower width w was 5.5 m. In the switching impulse tests, the gaps were varied from 4.5 m~8.2 m and the windage
yaw angle of I-insulator string was fixed to 18. The height of the conductor h was 38.5 m. The tower width w was
6.5 m. In the lightning impulse tests, the gap distances were varied from 4.0 m to 6.5 m and the windage yaw angle
of I-insulator string was fixed to 10. The h is 38 m and the w is 6.6 m.
The 50% flashover voltage curves as a function of different gap distances are shown from Figure 7.6.8 to Figure
7.6.10.
The per unit value of the conventional deviation, of the positive switching impulse flashover voltage varied within
the range of 4.0%~7.0%; this value range was obtained by analyses of the test results with the 4.5 m~8.2 m air-gap
clearances under standard switching impulse flashover test. Furthermore the value was irregularly rising with
increase of air-gap distance, as well as with the front time of switching impulse voltage. Since the values of were
obtained by up-and-down method rather than multistage method, they are not accurate. Investigations on the
statistics of the conventional deviation of switching impulse voltage with long times to peak will be the future
research work.
1800
U50( kV)
1600
1400
1200
1000
800
2
2. 7
3. 5
d ( m)
4. 5
Figure 7.6.8 50% Flashover voltage curve of side phase air-gap for
UHV cat-head-tower configuration as a function of gap distance
under power frequency voltage
Page 234
TypethenameofTB
U50 ( kV)
2400
2000
1600
1200
4. 5
5. 6
6. 5
7. 5
8. 2
d ( m)
Figure 7.6.9 50% flashover voltage curve of side phase air-gap for
UHV cat-head-tower configuration as a function of gap distance
under 250/2500 s positive switching impulse
3600
U50 ( kV)
3200
2800
2400
2000
3. 5
4. 5
5. 5
6. 5
d ( m)
Figure 7.6.10 50% flashover voltage curve of side phase air-gap for
UHV cat-head-tower configuration as a function of gap distance
under standard lighting impulse
7.6.2.2 FLASHOVER CHARACTERISTIC TESTS OF CAT-HEAD-TOWER MIDDLE PHASE
Since switching overvoltage on middle phase conductor suspended on V-insulator strings controls the cat-headtower windows size, the flashover voltage characteristic curve as a function of different gap lengths and of different
times to peak with a fixed gap length were obtained by tests, as well as lightning impulse flashover voltage for
different gap lengths. The test set up arrangement is shown in Figure 7.6.11. The Figure 7.6.12 shows the typical
flashover path for middle phase air-gap test with the UHV cat-head-tower.
The 50% switching impulse flashover voltage characteristic curve as a function of different gap lengths shown in
Figure 7.6.13 was obtained under the condition that the gap clearance d1 between V-insulator string grading rings
and the tower top were varied from 5.0 m to 8.3 m, d2 from grading ring to the tower side and d3 from conductor
bundle to the tower bottom were fixed to 7.8 m and 7.3 m respectively.
Page 235
TypethenameofTB
Figure 7.6.11 Arrangement of middle phase air-gap test with V-insulator strings
Figure 7.6.12 Picture of the typical flashover path for middle phase air-gap test
with UHV cat-head-tower
Page 236
TypethenameofTB
U 50 (kV)
1850
1700
1550
1400
4
d (m)
Figure 7.6.13 50% flashover voltage curve of middle phase V-insulator string air-gap
for UHV cat-head-tower window configuration as a function of gap distance
under 250/2500 s positive switching impulse voltage
The comparison of Figure 7.6.9 and Figure 7.6.13 shows that the 50% flashover voltage of the middle phase airgap between grading ring and tower top is 13%~15% lower than side phase air-gap between conductor and tower
side. This phenomenon can be explained as follows; the middle phase V-insulator strings and tower window form a
close electric field distribution around the high voltage electrode, so, the electric-field strength on V-insulator string
high voltage electrode is higher than that of the side phase I-insulator string high voltage electrode.
The 50% switching impulse flashover voltage characteristic curve as a function of different times to peak shown in
Figure 7.6.14 was obtained under the condition that the gap spacing between conductor and tower window
crossbeam was 7.9 m and the minimum gap between conductor and the side or the bottom of tower window was
6.7 m.
The gap spacing between high voltage ring and grounding ring of the insulator was 8.6 m. During the middle phase
air-gap test, the numbers of flashovers in paths from high voltage electrodes to different parts of the tower window
were counted. The table 7.6.3 presents the statistical results of flashover numbers in the different paths under
different times to peak of positive switching impulse voltage test. The flashover numbers in the path from (a) ring to
tower top is 38.4% of total flashover numbers, (b) ring to tower side and conductor to bottom are 29.3%
respectively, and (c) high voltage ring to grounding ring is 3.0%. Therefore, the flashover number in the path from
ring to top window was slightly more than that to the other parts of tower window, and the air-gap flashover number
between high voltage ring and grounding ring of the insulator occurred less. The flashover number in the path from
ring to top window significantly decreased when the time to peak of switching impulse voltage increased.
Page 237
TypethenameofTB
S.N.
Time to
peak (s)
Ring to
Tower top
Ring to
Tower side
1
2
3
4
5
120
250
500
1000
5000
11
9
9
8
1
38
38.4%
3
4
5
6
11
29
29.3%
Percentage
Conductor to
Tower
bottom
5
7
6
4
7
29
29.3%
Ring to
Grounding
ring
1
0
0
2
0
3
3.0%
Total
flashover
number
20
20
20
20
19
99
100%
Table 7.6.3 Flashover number statistical result in the paths of middle phase Vinsulator string air-gap for UHV cat-head-tower window under different times to
peak of positive switching impulse voltage test
U 50
kV
2600
2200
1800
1400
10
100
1000
Tp
10000
Figure 7.6.14 The 50% flashover voltage curve of middle phase V-insulator string
air-gap for UHV cat-head-tower configuration
as a function of time to peak of positive switching impulse
7.6.2.3 PHASE-TO-PHASE AIR-GAP FLASHOVER VOLTAGE TEST WITH UHV
TRANSMISSION LINE CONDUCTOR BUNDLES
The phase-to-phase air-gap flashover voltage test configuration is shown in Figure 7.6.15, the two 8-sub conductor
bundles representing two phase electrodes are horizontally suspended 18m high above earth. The two bundles
were injected separately with 250/2500 s positive and negative switching impulse voltage generated by 5400 kV
and 3000 kV impulse generators. The maximum negative switching impulse voltage is adjusted to maintain the
ratio = U- / (U+ + U-)=0.4 during test. The 50% standard phase-phase switching impulse flashover voltage
characteristics curve as a function of gap distance, which is varied from 5 m to 9 m, is shown in the Figure 7.6.16.
Page 238
TypethenameofTB
3500
U50( kV)
3000
2500
2000
1500
1000
3
10
d ( m)
Page 239
TypethenameofTB
Earth electrode
(Plane)
Gap length d
Page 240
TypethenameofTB
2200
d = 10m
2000
1800
d = 8m
1600
1400
d = 4m
1200
1000
10
30
100
300
1000
Time to peak Tf s
Figure 7.6.18 Curves of U50 flashover voltage variation for a rod-plane gap
of different clearances with time to peak of positive switching impulse
7.6.4 COMPARISON OF PAST AND NEW INVESTIGATIONS
Flashover characteristics of long air gap have been investigated under positive switching impulse with rod-plane
electrodes. Figure 7.6.19 shows time to peak Tp dependence of 50% probability flashover voltage for different air
gap lengths under dry conditions by using rod-plane electrodes. The data includes new experiments as shown in
Figure 7.6.18, as well as past experimental data of references [1, 11-14]. The flashover characteristic has the
critical time to peak for which the 50% flashover voltage is the minimum. The critical time to peak increases with the
gap length.
Figure 7.6.20 shows 50% flashover voltage for the critical time to peak and critical time to peak as a function of air
gap length. These experimental data of rod-plane electrodes are read off from Figure 7.6.19. The 50% flashover
voltages of the data agree well with formula (1), especially in the region of gap lengths larger than 8 meters. The
dotted curve is estimated by values of critical time to peak with rod-plane electrodes. Under around 5 or 6 m gap
length, EHV length range, the critical time to peak of rod-plane air gaps becomes the standard value 250 s. While
for the UHV length range (more than about 6 m length), the critical time to peak of rod-plane air gaps became
larger than 300 s.
In Figure 7.6.20 the data of India and Chinese actual UHV towers are also plotted. These experimental data are
shown in Figure 7.6.3 and Figure 7.6.5 for the Indian test results and in Figure 7.6.9, Figure 7.6.13 and Figure
7.6.16 for the Chinese test results.
Tpcr values of UHV tower measured recently in China and India are plotted at the little lower point of the fitting curve
in the figure. This tendency is similar to the past data of parallel conductor and V-insulator string for longer gap
lengths such as in Figure 7.5.4 and Figure 7.5.7. The geometry of air gap electrodes affects the critical time to peak
as well as the 50% flashover voltage. But for the UHV air gap lengths adopted in India and China the critical time to
peak of such actual Tower or substation configurations remain close to the 250 s associated to the Standard
Switching Impulse.
Page 241
TypethenameofTB
4000
3500
Past experimental
data *1
d=25m
2500
2000
d=10m
New experimental
data *2
1500
d=4m
1000
d=2m
500
0
10
100
1000
10000
Figure 7.6.19 Relationship between time to peak and 50% flashover voltage for
rod-plane gaps of lengths between 0.5 m to 25 m (*1 past and *2 new
experimental data)
*1
*1
*1
*2
*2
2400
Formula (2)
[Paris's formula]
2200
2500
2000
Formula (1)
[Critica sparkover voltage]
2000
1800
1600
3000
1400
1500
1200
1000
1000
800
600
500
400
200
0
0
0
10
12
14
16
18
20
22
24
26
28
Page 242
TypethenameofTB
2400
2200
2000
U10 [kV]
1800
1600
ph-e_Conductor-Structure (IEC)
1400
1200
1000
4000
5000
6000
7000
8000
9000
10000
d [mm]
Page 243
TypethenameofTB
In the same way as before (U50 to U10 and only standard wave shapes) the results from the configurations with
tower windows have been gathered in Figure 7.6.22. Here insulators were partly involved. The following results
have been used:
from Indian measurements, the Tower Window values according to Table 7.6.2 (with Tp = 250 s and 270 s)
from Chinese measurements, the Tower-Middle Phase values according to Figure 7.6.13 (all with Tp = 250 s)
Figure 7.6.22 shows much lower U10 values in comparison to the conductor-tower configurations of Figure 7.6.21.
The reason for that may be manifold. There is a more concentrated electrical field around the conductor due to the
tower window which leads to a higher electric-field strength on the high voltage electrode. Additionally there were
insulator strings in the windows; but it must be seen that the insulators only in some cases were directly involved in
the flashovers. Nevertheless they are a further increase in field-stress due to the armature and fixing points of the
insulators on the conductor.
The lower U10 values of the Tower Window configurations are compared in Figure 7.6.22 with the clearances for
the inhomogeneous Rod-Structure air gaps in IEC 60071-1. In the upper range the Chinese measurements are
slightly below the Rod-Structure values, but the Indian measurements fit very well. Therefore taking the RodStructure values for most unfavorable configurations should be handled with care. But this statement should not be
generalized due to the fact that there are not enough test results and due to the above mentioned uncertainty of
these considerations.
2400
2200
2000
U10 [kV]
1800
1600
ph-e_Rod-Structure (IEC)
1400
1200
1000
4000
5000
6000
7000
8000
9000
10000
d [mm]
Page 244
TypethenameofTB
Page 245
TypethenameofTB
Figure 7.7.1 Ratio of the withstand voltages for Standard Switching Impulse
(derived with Paris formula) and
Critical time to peak impulse (derived with Watanabe formula)
versus the air gap spacing
For phase to phase insulation involved in UHV overhead lines or substations, the critical time to peak is in the
range of 250 s whatever the air gap distance is. So for phase to phase distances of IEC there is no
underestimation of the distance needed.
In order to take into account the 1.06 difference for phase to earth air gap; instead of increasing the specified
substation or overhead line distances, post type insulators may be applied which flatten the U curve and reduces
the flashover voltage. Otherwise, a test could be carried out using rated standard switching impulse voltage.
Representative Overvoltages
Ucw :
Urw :
Ks :
Safety Factor
Kc :
Co-ordination Factor
Page 246
TypethenameofTB
Urp2/SI = U2
NOTE: Usually the Standard Switching Impulse is referred as SI but in this table to explicitly refers to the
Standard Switching Impulse an S has been added before: SSI. This is to clearly makes here (paragraph 7.7.2)
the difference with switching impulses of different time to peak (and even time to half value) called here SITpcr
as they refer to the critical time to peak of a given air gap.
In conclusion for the above example in case of a rod-plane gap configuration if distances above 10 meters are
actually needed for that gap type insulation then:
Urw (critical time to peak > 250 s) = (1.1 or 1.2 ) Urw (250 s time to peak)
dmin(critical time to peak > 250 s) = (1.17 or 1.35) dmin (250 s time to peak)
Of course in case of air gap configurations having Ktc(SSI / SITpcr) "conversion factor" between voltage
with critical time to peak (SITpcr) and the Standard SI voltage (SSI) lower than 1, the conclusion would be of
course the reverse in that cases.
Page 247
TypethenameofTB
7.8 Conclusion
In the following the conclusions out of Chapter 7 are gathered as hints and recommendations for IEC TC 28
Insulation Coordination concerning the special aspects of air gaps in the UHV range.
In this Chapter 7 the knowledge about air gap clearance and flashover characteristics of switching impulses is
refreshed, based on IEC Insulation Coordination and studies from the past. Recent investigations give some new
input. The only relevant voltage stress for air gaps in the UHV range is the positive switching impulse voltage.
Some recent investigations from Japan, India and China are presented.
Japanese investigations support the accuracy of Equation (1) for the calculation of the 50% flashover voltages for
rod-plane gaps with a decrease of the flashover voltage with increasing time to peak.
The systematic Indian investigations on a tower configuration show that the selected gap distance of 8 m is slightly
below the rod-structure value of the IEC-Table value of 8.3 m to meet the required switching impulse withstand
voltage of 1800 kV. The conductor-structure value of the IEC-Table with 6.2 m seems to be too small for general
dimensioning.
The Chinese results indicate that for a tower configuration with V-insulator-strings it is sufficient to consider the
standard impulse wave shape to find the lowest flashover voltage for the investigated air gap in the tower. Longer
times to peak lead to higher flashover voltages.
Transferring the U50 measurements from India and China to withstand voltages U10 by calculation and comparing
these results with the recommended UHV minimal air gap clearances of IEC 60071-1 show that the ConductorStructure clearances are rather close to the measurements and it supports the validity of the minimal air gap
clearances selected in IEC 60071-1 for the standard UHV switching withstand voltages. The much larger minimal
clearances for UHV Rod-Structure air gaps in IEC fit more or less to the measurements of tower window
configurations, but such a correlation should be handled with care due to various parameters which cannot be
figured out from these measurements. Special care has to be taken also for these comparisons because of the
uncertainties due to the calculation and the low number of test series.
The u-curve behavior of flashover voltages with non standard switching impulses is valid for rod-plane
configurations. From investigations in the past as well as from the recent measurements from India and China it
can be seen that configurations with a more homogeneous field distribution and configurations including insulators
do not show this behavior. For actual configurations in a tower with and without insulators the critical time to peak is
near to the peak of the standard wave shape. The necessity to use impulses with longer front times in tests seems
not to be compulsory so as for the calculations.
If longer front times are expected and unfavorable configurations have to be considered the relation between
standard and non-standard wave shape can be included in the insulation coordination process by means of the
coordination factor Kc.
The IEC-equations for the calculation of gap clearances include uncertainties in the gap factor K and the per unit
value of the conventional deviation . The gap factor can only be estimated with low accuracy based on the
examples given in IEC and the literature. The influence of the selected deviation cannot be neglected but the
recommended IEC value of 6% may be in most cases on the safe side. Therefore the calculations can only give
some support. For to achieve a reliable gap design tests are necessary.
Page 248
TypethenameofTB
7.9 References
1. L. Paris,Influence of Air Gap Characteristics on Line-to-Ground Switching Surge Strength, IEEE Trans. on
Power Apparatus and Systems, vol. 86, no.8, pp. 936947, 1967.
2. L. Paris and R. Cortina, Switching and lightning impulse discharge characteristics of large air gaps and long
insulator strings, IEEE Transactions on Power Apparatus and Systems, Vol. PAS-87, No.4, April 1968, pp. 947-957.
3. W.C. Pokony and R.W. Flugun, UHV tower insulation parameters determined by full scale testing, IEEE
Transactions on Power Apparatus and Systems, Vol. PAS-94, No.2, Mar-April 1975, pp. 518-525.
4. G. Gallet., G. Leroy, R. Lacey and I. Kromer, General expression for positive switching impulse strength valid up
to extra long air gaps, IEEE Transactions on Power Apparatus and Systems, Vol. PAS-94, No.6, Nov./Dec, 1975,
pp.1989-1993.
5. A. Pigini, G. Rizzi, R. Brambilla, and E. Garbagnati,Switching Impulse Strength of Very Large Air Gaps, 3rd
I.S.H. Milan 1979, article 526-15.
6. I. Kishizima, K. Matsumoto, Y. Watanabe, New facilities for Phase-to-Phase Switching Impulse Tests and Some
Tests Results, IEEE Transactions on Power Apparatus and Systems, Vol. PAS 103, No. 6, June 1984, pp. 12111216.
7. R. Cortina, E. Garbagnati, A. Pigini, G.Sartorio and L.Thione, Switching impulse strength of phase-to-earth UHV
external insulation - Research at the 1000kV project, IEEE Transactions on Power Apparatus and Systems, Vol.
PAS-104, No.11, November 1985, pp. 3161-3168.
8. Yoshibumi Yamagata, Akio Oe, Kunihiko Miyake, Yoshinori Aihara, and Takatoshi Shindo, Phase-to-ground and
phase-to-phase sparkover characteristics of external insulation at the entrance of a UHV substation, IEEE
Transactions on Power Delivery, Vol.17, No.1, January 2002. pp. 223-232.
9. IEC 60071 : Insulation co-ordination : Part 1: Definitions, Principles and Rules. Edition 8.1 - 2011-03
10. IEC 60071-2 Insulation co-ordination : Part 2: Application guide, Edition 3, 1996-12.
11. I.Kishizima, et al "New facilities for phase-to-phase switching impulse tests and some test results, IEEE,
Transactions on, Power Apparatus and Systems, Volume: PAS-103 Issue:6 pp.1211 1216, (1984)
12. R.Cortina et al.; Switching Impulse Strength of Phase-to-Earth UHV External Insulation -Research at the 1000
kV Project-, IEEE Transactions on, Power Apparatus and Systems, Volume: PAS-104, Issue:11, pp.3161-3168,
(1985)
13. R.Cortina et al.; Air Density Influence on the strength of external insulation under positive impulse:
Experimental investigation up to an altitude of 3000m a.s.l., IEEE Transactions on, Power Apparatus and Systems,
Volume:5, Issue:2, pp.730-737 (1990)
14. G.Galet et al.; General expression for positive switching impulse strength valid to extra long air gaps, IEEE
Transaction on Power Apparatus and Systems, Volume: 1. PAS-94, No.6, (1975)
15. IEC 61083-1 (2001) Instruments and software used for measurements in high-voltage impulse tests, Part 1:
Requirements for instruments
16. IEC 61083-2 (1996) Digital recorders for measurements in high-voltage impulse tests-, Part 2: Evaluation of
software used for the determination of the parameters of impulse waveforms
17. Les Renardieres Group: Research on long air gap discharges at Les Renardiers - 1973 Results; Electra No 35;
1974
Page 249
TypethenameofTB
18. CIGRE Technical Brochure 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
19. CIGRE Technical Brochure 400; Technical requirements for substations exceeding 800 kV; Working Group
B3.22; December 2009
Page 250
TypethenameofTB
System analysis
Insulation characteristics
Performance c riterion
Statistical distribution (+)
Inaccuracy of input deta (+)
(+) Effects combined in a co-ord ination
factor Kc
Test conditions
Test conversion factor Ktc
Standard withstand voltages
Rages of Um
Rated or standard insulation level : set of Uw
NOTE In brackets the subclauses reporting the definition of the term or the description of the action.
Sided bo xes refer to required input
Sided bo xes refer to performed actions
Sided bo xes refer to obtained results
Figure 8.1.1 IEC 60071-1 Flowchart for the determination of rated or standard
insulation levels
Page 251
TypethenameofTB
The following steps are followed in the determination of rated or standard insulation levels:
Step 1: Determination of the representative voltages and overvoltages (Urp)
These voltages and overvoltages are defined in IEC 60071-1as follows:
The voltage and the overvoltages that stress the insulation shall be determined in amplitude, shape and duration by
means of a system analysis which includes the selection and location of the overvoltage preventing and limiting
devices. For each class of voltages and overvoltages, this analysis shall then determine a representative voltage
and overvoltage, taking into account the characteristics of the insulation with respect to the different behaviour at
the voltage or overvoltage shapes in the system and at the standard voltage shapes applied in a standard
withstand voltage test.
The representative voltage and overvoltages may be characterized either by:
- an assumed maximum, or
- a set of peak values, or
- a complete statistical distribution of peak values.
Using the waveform evaluation described in Section 5.3, the voltage waveforms generated in the field can be
converted to the standard voltage shapes to determine these values.
Step 2: Determination of the co-ordination withstand voltages (Ucw) that satisfy the performance standard required
for the system
The co-ordination withstand voltages are defined in IEC 60071-1 as follows:
The determination of the co-ordination withstand voltage consists of determining the lowest values of the withstand
voltage of the insulation meeting the performance criterion when subjected to the representative overvoltages
under service conditions.
The co-ordination withstand voltages of the insulation have the shape of the representative overvoltages of the
relevant class and their values are obtained by multiplying the values of the representative overvoltages by a coordination factor. The value of the co-ordination factor depends on the accuracy of the evaluation of the
representative overvoltages and on an empirical, or on a statistical appraisal of the distribution of the overvoltages
and of the insulation characteristics. The co-ordination withstand voltages can be determined as either
conventional assumed withstand voltages or statistical withstand voltages. This affects the determination procedure
and the values of the co-ordination factor.
Simulations of overvoltage events combined with the simultaneous evaluation of the risk of failure, using the
relevant insulation characteristics, permit the direct determination of the statistical co-ordination withstand voltages
without the intermediate step of determining the representative overvoltages.
Step 3: Determination of the required withstand voltage (Urw)
The determination of the required withstand voltages of the insulation consists of converting the co-ordination
withstand voltages to appropriate standard test conditions. This is accomplished by multiplying the co-ordination
withstand voltages by factors which compensate the differences between the actual in-service conditions and those
in the standard withstand voltage tests.
The factors to be applied shall compensate atmospheric conditions by the atmospheric correction factor Kt and the
effects listed below by a safety factor Ks.
Effects combined in a safety factor Ks;
- the differences in the equipment assembly;
Page 252
TypethenameofTB
Page 253
TypethenameofTB
Page 254
TypethenameofTB
Page 255
TypethenameofTB
where
Urw: Required withstand voltage
Ka: Altitude correction factor
Ks: Safety factor
Ucw: Co-ordination withstand voltage
Urp: (LIPL+2 ST) for fast-front overvoltage, SIPL for slow-front overvoltage
Kcd: Deterministic co-ordination factor (1 for LIPL, 1.0 ~1.1 for SIPL as per IEC 60071-2 Figure 6)
S: Steepness of impinging surge
T: Travelling time of lightning surge from arrester point to the protected equipment terminal such as
transformer.
These basic rules and equations can be used for preliminary estimations. For more complex and/or practical
applications, results provided by these simplified formulas are considered to be less precise. For such cases, better
results are usually obtained from detailed computer simulations. It may incorporate such important features as
realistic line and station configurations with tower and ground transients, accurate models such as multi-phase
coupled lines with frequency-dependent parameters and corona, capacitances from major equipment such as
CVTs, non-linear behaviour of arresters plus many more which are described in IEC TR 60071-4.[4]
REFERENCES
[1] IEC 60071-1, Insulation co-ordination Part 1: Definitions, principles and rules, 2006-01.
[2] IEC 60071-2, Insulation co-ordination Part 2: Application guide, 1996-12.
[3] IEC 60099-4, Surge arresters Part 4: Metal-oxide surge arresters without gaps for a.c. systems, 2009-05.
[4] IEC TR 60071-4, Insulation co-ordination Part 4: Computational guide to insulation co-ordination and modeling
of electrical networks, 2004-06.
Page 256
TypethenameofTB
800kV
800kV
800kV
800kV
800kV
800kV
Country/Utility
Brazil/
Furnas
China/
SGCC
India/
Power Grid
Korea/
KEPCO
South
Africa
United
States/
AEP
According
to IEC and
ANSI
standard at
that time
According
to
IEC60071-2
(1996);
1.15, 1.05
Considering
for the
sever
condition*
1.12-1.2
N/A
N/A
1.25
N/A
1.32
N/A
N/A
1.15
N/A
LIWV
SIWV
800kV
765kV
Venezuela Canada
1200kV
1200kV
1100kV
1100kV
1050kV
Country/Utility
India/Power
Grid
Russia/
China/SGCC
Japan/
TEPCO
Italy/CESI
1.25
According to
IEC60071-2
(1996);
1.15, 1.05
N/A
N/A
LIWV
SIWV
According
to
IEC600712 (1996);
1.15, 1.05
1.15
Page 257
TypethenameofTB
was proposed. In the meetings of the WG, no document on the numerical computation of the safety factor was
found.
In the draft IEC 71-1 submitted to National Committees in 1987, the safety factor was defined as follows:
The safety factor is a factor based on experience, that is introduced to account for all the elements which can
influence negatively the insulation performance, and which are not taken into consideration in the insulation coordination procedure, since they are either unknown, or difficult to be quantified, or related to insufficient knowledge
of the values introduced in the calculation.
IEC TC-28 WG02 studied the insulation coordination procedure and discussed overvoltages, withstand voltages,
and various factors. During this work, in order to clarify the details of safety factor, it was explained that the safety
factor would compensate for the following influences:
(1) dispersion in production
(2) contamination during the erection on site (GIS)
(3) aging of liquid and solid insulation
(4) parallel insulation
The draft report in 1989 indicated the following:
Safety factor
The value of 1.02 for external insulation is different from 1.05 of IEC 60071-2(1996). However, no document
explaining the numerical value was found.
While the details of the safety factor were defined in these two working groups, it seems that not all the elements of
the safety factor were evaluated at that time. It is assumed that sufficient data was not yet available and individual
effects could not be quantified. Consequently, the safety factor was proposed as the overall value based on the
margin empirically used at that time, and that value was adopted.
Page 258
TypethenameofTB
obvious ageing of the internal insulation. The results show that the dielectric performance of the current transformer
can be considered as similar to that of a new unit.
Other special tests were performed to characterise the performance of a 17 years old composite polymer insulator.
The composite insulator also showed excellent performance.
(2) Performance of epoxy insulating spacer
Ageing and residual performance of insulating spacers made of epoxy resin was studied, using the actual spacers
after long-term field operation in Japan.
The reference reports the test results on the ageing of 154kV insulating spacer after 17 years operation since 1971.
The verification tests by lightning impulse withstand voltage (LIWV), AC withstand voltage (ACWV) and PD
measurement have been conducted, which shows satisfactory insulating performance. The test results also
suggest that there was no deterioration after 17 years of operation. The interface between the high-voltage
embedded electrode and the epoxy resin spacer was observed, reporting that no sign of deterioration was found at
the interface.
Similar tests using 72/84kV epoxy insulating spacers after 23 years of in-service operation were conducted in terms
of the lifetime assessment. First, PD measurements at AC 128 kVrms and AC withstand tests at 160 kVrms for one
minute were conducted at 0.4 MPa for 12 spacers of 23 years old. No PD detection and no breakdown were
confirmed. Secondly, the ageing of the 23 years old spacers are accelerated by additional 1.5 times AC stress of
operating voltage during 5.5 months, which corresponds to 67 years of operating voltage. After the AC voltage
acceleration, PD measurements and breakdown tests at LI voltage are performed and no PD detection and no
reduction of LI breakdown voltages are confirmed again. Lastly, the interface between the high-voltage embedded
electrode and the epoxy resin is observed in detail. It is reported that there is no sign of ageing of epoxy resin
around the maximum electric field portion. Therefore, the series of the tests verified that the spacers have at least
90 years lifetime.
(3) Performance of a 123 kV GIS
The literature reports that overall the entire insulation system does not exhibit an increase in the failure rate with
time due to the failure rate of 123kV GIS in Germany since 1967 as exhibited in Figure 8.1.3
There is a general reduction of the failure rate with time, which proves that all further failures were caused by
teething faults during the first 6 years of operation. However, even after 25 up to 30 years of service no significant
increase of the failure frequency was noted. Therefore, the literature concludes that after 30 years of operation
there is no general ageing which recognizably affects the long-term performance.
Figure 8.1.3 Failure rate of 123kV GIS in Germany due to dielectric faults
Page 259
TypethenameofTB
(4) Summary
The investigation results above mentioned can be summarized as follows.
Equipment
(Insulated part)
Voltage
class
Age in the
field
CT
245 kV
17 years
Epoxy spacer
154 kV
17 years
Epoxy spacer
72/84 kV
GIS
123 kV
Test method
Results
No PD up to 460 kV
No PD, No breakdown,
PD test, AC voltage withstand test,
No PD even after the accelerated
Accelerated
deterioration
test
23 years
ageing test,
equivalent to 67 years, Investigation
No LI breakdown voltage decrease,
of the boundary face
No change in the boundary surface
Change in the failure rate over years
No ageing in 30 years
Page 260
TypethenameofTB
Page 261
TypethenameofTB
Switching impulse
withstand voltage
Insulation
External insulation
air clearances and clean insulations, dry
clean insulators, wet
Internal insulation
GIS
liquid-immersed insulation
solid insulation
NOTE The test conversion factors include a factor of
1,4
1,7
1,6
2,3
2,0
2 to convert from r.m.s to peak value.
Table 8.2.1 Test conversion factors for range II to convert required short-duration
power-frequency withstand voltages to switching impulse withstand voltages
(According to Table 3, IEC60071-2 1996).
8.2.2 STANDARD RATED WITHSTAND VOLTAGES FOR UHV SYSTEMS
This WG conducted a research into the lightning and switching impulse withstand voltages. Table 8.2.2 and Table
8.2.3 summarize the research results for the insulation design of substations (only UHV systems were studied in
this draft).
Page 262
TypethenameofTB
Highest voltage
1200kV
Country / Utility
Russia/
Substation type
AIS
Representative
overvoltage (kV)
1100kV
1100kV
1050kV
GIS
GIS
N/A
N/A
1700
1700
1796
2040
1896
2208
N/A
N/A
Calculation method
Special
programme
Simplified
method
EMTP
EMTP
TNA,
EMTP
Safety factor
1.15
1.15/1.05*
1.15/1.05*
1.00
N/A
Tr.
Other
2250
2400
2250
2400
2250
2400
1950
2250
2250
2250
Tr.
N/A
1.32
1.25
1.03
N/A
Other
N/A
1.41
1.18
1.02
N/A
LIWV (kV)
LIVW/
Representative
overvoltage
Tr.
Other
1200kV
India/Power
Grid
AIS,
Hybrid-IS
Table 8.2.2 LIWV of substation equipment for UHV systems and determination
process
8.2.2.2 SWITCHING IMPULSE WITHSTAND VOLTAGES
Table 8.2.3 indicates that all countries set either 1425 kV or 1800 kV for the SIVW of transformers, and 1550 kV or
1675 kV or 1800kV for the SIWV of other equipment. Russia, India/Power Grid and China/SGCC set 1800 kV for
both transformers and other equipment. The bottom two rows of the table indicate the ratios between SIWV and
representative overvoltage. The smallest value of 1.09 is used by TEPCO (Japan) for transformers, and the largest
value of 1.29 is used by SGCC of China for transformers. Differences among countries are less significant than
those for the lightning impulse withstand voltage. In p.u., the representative overvoltages in Table 8.2.3 range
between 1.46 and 1.69, which are significantly lower than those in lower voltage systems in general. This is
because switching overvoltages are effectively suppressed in UHV systems by means of circuit breakers with
resistors, shunt reactors, higher performance surge arrestors, etc. Switching overvoltages are explained in detail in
Section 4.3.
Page 263
TypethenameofTB
Highest voltage
1200kV
Country / Utility
Russia/
Substation type
AIS
Representative
overvoltage (kV)
1050kV
Italy/CESI
GIS
Tr.
1570
(Maximum)
1500
(2% value)
1392
(2% value)
1309
(Maximum)
1450
(Maximum)
Other
1570
(Maximum)
1500
(2% value)
1437
(2% value)
1400
(Maximum)
1450
(Maximum)
Special
programme
Simplified
method
EMTP
EMTP
TNA,
EMTP
Tr.
Other
Tr.
1.15
1800
1800
1.15
1.15/1.05*
1800
1800
1.20
1.15/1.05*
1800
1800
1.29
1.0
1425
1550
1.09
N/A
1800
1675
1.24
Other
1.15
1.20
1.25
1.11
1.16
Calculation method
Safety factor
SIWV (kV)
(Phase-to-ground
SIWV/
Representative
overvoltage
1200kV
1100kV
1100kV
India/Power
China/SGCC Japan/TEPCO
Grid
AIS,
GIS,
GIS
Hybrid-IS
Hybrid-IS
Table 8.2.3 SIWV of substation equipment for UHV systems and determination
process
8.2.2.3 VERIFICATION OF SWITCHING IMPULSE WITHSTAND VOLTAGES FOR TOV
As explained in Section 8.2.1, it should be taken into account in determining the standard withstand voltage that the
standard switching impulse withstand voltage is required to cover short-duration power-frequency withstand
voltages in UHV systems. Hence, estimation was carried out to evaluate the possibility of using the switching
impulse withstand voltages for verifying the power-frequency withstand voltages (PFWV) in the standard procedure
of IEC60071-1 2006 and IEC60071-2 1996.
Table 8.2.4 lists the converted switching impulse withstand voltages equivalent to the power-frequency withstand
voltages.
Here, the maximum value of the temporary overvoltage was assumed to be the representative temporary
overvoltage. Co-ordination factor Kc = 1, that is the co-ordination withstand voltage is equal to the representative
temporary overvoltage, according to Section in 3.3.1 of IEC 60071-2 1996. Similarly, safety factor Ks is 1.15, and
conversion factor is 2.3 (Table 8.2.1) according to IEC 60071-2.
The bottom row of the table indicates the switching impulse withstand voltages estimated using a standard
conversion factor in IEC 60071-2 1996. These values, ranging from 2383 to 2566 kV, are considerably greater than
SIWVs in Table 8.2.3, and require the switching impulse withstand voltages of at least 2400 kV in UHV systems.
Since the SIWVs greater than the lightning impulse withstand voltages are economically and technically
disadvantageous, an alternative test is available to verify the withstand of internal insulation to TOV, as mentioned
in Section 6.4 of IEC60071-1 2006. Therefore, the power-frequency withstand voltage tests, which is explained in
Section 8.4, are provided for verifying withstand voltages by the relevant apparatus committees.
Page 264
TypethenameofTB
Highest voltage
1200kV
1200kV
1100kV
1100kV
1050kV
Country / Utility
Russia/
India/Power
Grid
China/SGCC
Japan/TEPCO
Italy/CESI
Substation type
AIS
AIS, Hybrid-IS
GIS, Hybrid-IS
GIS
GIS
1.3 p.u.
(901kV r.m.s)
1.4 p.u.
(970kV r.m.s)
1.4 p.u.
(889kV r.m.s)
1.5 p.u.
(953kV r.m.s)
N/A
EMTP
EMTP
TNA / EMTP
2521
N/A
Representative
overvoltage
(Maximum TOV)
Calculation method
Co-ordination factor
1.0*
Safety factor
1.15*
Conversion factor
2.3*
Equivalent SIWV
2383
2566
2351
Page 265
TypethenameofTB
Voltage [pu]
4.0
3.0
GIS
GIS
GIS
2.0
MTS
MTS
1.0
Japan
South Africa
China
0.0
300
500
700
900
Rated voltage Ur [kV]
1100
Figure 8.3.1 Dependency of rated withstand voltages and calculated VFTO on rated
voltage as per IEC 62271-203 , calculated values from [2, 4, 5, 6]
Page 266
TypethenameofTB
Page 267
TypethenameofTB
Page 268
TypethenameofTB
YES
System analysis
VFTO calculation
(verified calculation
method)
Maximum
calculated VFTO
Umax_VFTO
charge behavior
Insulation characteristic
Statistical distribution
Inaccuracy of simulation
KC
Co-ordination factor KC
Correction factors
Atmospheric correction
factor Kt
Aging in service
Quality of installation
Co-ordination withstand
voltage
Ucw_VFTO
KS
Application of factors to
account for the
differences between
type test and actual
service conditions
Safety factor KS
External
insulation
YES: Kt
Required withstand
voltage
Urw_VFTO
Test conditions
Test conversion factor Ktc
Ktc
Standard lightning
impulse withstand voltage
(LIWV)
NO
Definition of required
damping measures
YES
No damping measures
necessary
Page 269
TypethenameofTB
Voltage [kV]
1100
900
700
500
300
Measurement
Calculation
100
-100
1.8
1.9
2.1
2.2
2.3
2.4
2.5
Time [us]
Figure 8.3.3 VFTO calculation and measurement when switching busbars with a GIS
disconnector as per IEC 62271-102 without pre-charging (left), test setup (right)
[16]
Step 2 Required VFTO withstand level - comparison with LIWV
Basis for the insulation coordination is calculation of the required VFTO withstand voltage. The insulation withstand
strength under VFTO stress must be compared with the LIWV. The necessary safety depends on many factors and
could be different for the different kinds of equipment. Generally, the VFTO withstand voltage is influenced by the
breakdown behaviour of the insulating material, the frequency of occurrence and the probability of trapped charge
voltages as a basis for the simulation. Further aspects should not be disregarded.
The absolute number of occurrence of VFTO processes caused by disconnector switching during the total lifetime
of the equipment is in the range of some thousands up to ten thousand whereas the number of overvoltage
processes due to lightning is in the range of some tens up to a few hundred.
An accurate modelling of each individual GIS component makes it possible to reproduce VFTO waveforms with a
relatively high precision (differences lower than 5 %), especially in short GIS structures or test equipment. In case
Page 270
TypethenameofTB
of a proved simulation tool a coordination factor of 1.05 is recommended. Exceptions are possible in case of
unsecure simulation.
Sometimes, variations of more than 10 % depending on the calculation methods are reported. Generally, it is
important to verify the simulation results by measurements. But in the described case a coordination factor of 1.1 is
recommended. Otherwise, if it is verified that the simulation results are higher in any case compared to the
measured VFTO peak values a lower a coordination factor of 1 could be used.
The maximum value of the VFTO depends on the voltage difference between the contacts of disconnector just
before striking and on the location considered. For the calculation of VFTO stresses, the trapped charge charges
remaining on the load side of the disconnector must be taken into consideration.
Switching by a slow acting disconnector generates numerous re-strikes between the moving and fixed contacts, but
99 % of the trapped charge voltages were limited to 0.4 p.u. [13]. For these cases, the resulting VFTO is in the
range of 1.7 p.u. and reach 2.0 p.u. for very specific cases. Fast operating disconnector on the other hand can
leave residual charge levels corresponding to 1 p.u. in a no negligible number of cases [16, 17]. The trapped
charge voltage behaviour strongly depends on the contact speed, as shown in the simulation plot (see Figure
8.3.4) [16]. A lower trapped charge voltage gives a higher safety margin compared to the calculation based on a
trapped charge voltage of 1 p.u. For the insulation coordination this additional margin has to be considered.
1
calculated
90% probability
95% probability
99% probability
measured
0.8
0.6
0.4
0.2
0
0
0.5
1.5
2.5
3.5
Page 271
TypethenameofTB
to defects give considerably lower breakdown values compared to a sound system. But also in case of
inhomogeneous fields the minimum breakdown voltage occurs for lightning impulse waveform having a front time
of 5 s. For times shorter than 1s, the breakdown voltage increases with decreasing time caused by the changing
leader inception conditions. Moreover, fixed protrusions on live parts are usually avoided by a proper design,
quality control and adequate testing in both factory and on-site. They can be detected by sensitive diagnostic
measurements under AC voltage stress.
The test conversion factor Ktc describes the comparison between the breakdown strength of the insulating system
between VFTO and LIWV stress for a given equipment or insulation configuration. The factor to be applied to the
required withstand voltage, which describes the different withstand behaviour under VFTO stress compared to the
stress with standard LI voltages. For SF6 insulated systems like GIS or GIL the recommended Ktc is 0.95. The
recommended test conversion factor Ktc for oil or oil/solid insulated systems is 1.0.
Step 3 Measures according to the insulation coordination
If the required VFTO withstand voltage is higher than LIWV, it is necessary to define measures reducing the risk of
failures. There are two possibilities: an increase of the LIWV or a mitigation of VFTO. The first choice is easy to
realize, but cost-intensive. Nevertheless in some cases this solution has advantages. The second choice aims for
mitigation of amplitudes of VFTO and finally for a reduction of the effect of VFTO on the equipment [2].
The damping of VFTO by integration of a damping resistor is a well proven technology. Service experiences exist
since more than 10 years [2]. Generally, the mitigation effect of the damping resistor depends on the value of the
resistance. To verify the mitigation effect, calculations and measurements were performed for the Japanese and
Chinese UHV projects.
Figure 8.3.6 shows a relation between resistance and VFTO peak for a typical GIS layout with 8 bays and a double
busbar scheme. The calculation is based on the assumption that re-striking occurs when the trapped charge
voltage at the load side and the voltage at the source side were -1 p.u. and +1 p.u. respectively. Without damping
resistor, the VFTO amplitude reaches a value of 2.49 p.u. This exceeds the LIWV according to the insulation
coordination. In case of integration of a damping resistance higher than 200 , the VFTO amplitude can be
suppressed below 1.5 p.u. A damping resistance more than 500 mitigates the maximum VFTO amplitude to a
level lower than 1.13 p.u. [20]. Consequently the resistance of damping resistor could be chosen and defined
according to the maximum calculated VFTO and the required mitigation effect. A 110 damping resistor was used
in the Italian 1000 kV project [2]. For the Chinese, Japanese and Korean UHV and EHV projects, it was decided to
use a 500 damping resistor [2]. An example of GIS disconnector structure with damping resistor is shown in
Figure 8.3.7.
1.13 pu
500
Page 272
TypethenameofTB
The damping resistor has to withstand the dielectric stress during striking. The highest voltage across the resistor
occurs shortly after the first pre-strike during close operation. Therefore, it is necessary to prove the voltage
withstanding characteristic and energy absorption capability of the resistor in case of re-strikes and pre-strikes
between the moving contact and the arcing electrode of the resistor. A flashover across the resistor may lead to
high VFTO comparable to a disconnector without damping resistor and has to be avoided.
A higher resistance value leads to a higher voltage stress across the damping resistor and can reach values in the
range of 2 p.u. Moreover the rate of rise of the voltage across the resistor could be very high and depends on the
test setup and the capacitances on the load and source side. The rate of rise of the voltage across the resistor has
to be considered especially during testing.
The absorption energy strongly depends on the load side capacitance and the voltage across the disconnector.
The required capacity of thermal energy absorption for the resistor could be calculated by summing up all closeopen operations containing a high number of strikes. For typical applications the required energy absorption for a
500 resistor ranges between 20 kJ and 35 kJ for one close-open operation [2]. Mostly the thermal absorption
capability is defined to withstand the thermal stress for one close-open operation. The possibility to operate more
than one close-open operation within some minutes which corresponds to the thermal time constant of the damping
resistor is estimated to be very short.
Page 273
TypethenameofTB
BIBLIOGRAPHY
[1] CIGR Working Group A3.22: Technical Requirements for Substation Equipment Exceeding 800 kV - Field
experience and technical specifications of Substation equipment up to 1200 kV, Brochure 362, December 2008
[2] CIGR Working Group A3.22: Background of Technical Specifications for Substation Equipment Exceeding
800 kV AC, CIGR Brochure No. 456, April 2011
[3] Riechert, U.; Neumann, C.; Hama, H.; Okabe, S.; Schichler, U., on behalf of CIGR WG D1.36 and AG D1.03:
Basic Information and Possible Counter Measures Concerning Very Fast Transients in Gas-Insulated UHV
Substations as Basis for the Insulation Co-ordination, CIGR SC A2 & D1 Joint Colloquium 2011, Kyoto, Japan,
PS3-O-5
[4] Riechert, U.; Neumann, C.; Hama, H.; Okabe, S.; Schichler, U., CIGR SC D1: Very Fast Transient
Overvoltages (VFTO) in Gas-Insulated UHV Substations, CIGR TB, accepted for publication in 2012
[5] Hemmi, R.; Shinohara, R.; Kitazumi, Y.; Yatsuzuka, H.; Hirasawa, K.; Yamagiwa, T.: Evaluation of VFTO (Very
Fast Transient Overvoltage) and Its Reduction by Parallel Resistor during Switching Operation of Disconnecting
Switch in Future UHV Gas Insulated Substations, Int. Conference of UHV Power Transmission Technology,
Beijing, China, 2006, pp. 460-466
[6] Ban Liangeng, Xiang Zutao, Wang Sen, Lin Jiming, Wang Xiaogang, Zheng Bin, Wang Xiaotong: Estimation of
VFTO for GIS and HGIS of China 1000 kV UHV pilot project and its suppressing countermeasures, IEC - CIGR
UHV Symposium, Beijing, China, 2007, paper 2-3-4
[7] KOBAYASHI, T.; SUN, G.; YOKOTA, T.; SASAMORI, K.; YAMAGIWA, T.: Compactness and High-reliability
Technology for UHV Circuit-breakers, disconnectors and earthing switches, The second IEC CIGR
International Symposium on International Standards for UHV Transmission, 29-30 January 2009, New Delhi, India,
proceedings pp. 169-177
[8] YAMAGATA, Y.; OKABE, S.: Utility's experience on design and testing for UHV equipment in Japan, The
second IEC CIGR International Symposium on International Standards for UHV Transmission, 29-30 January
2009, New Delhi, India, proceedings pp. 142-154
[9] Yamagata, Y.; Nakada, Y.; MuraYama, Y.; Hirasawa, K.; Yoshizumi, T.: Field Test of 1000 kV Gas Insulated
Switchgear, CIGR Report 13-209, 38th CIGR Session, Aug. 27 to Sept. 1, 2000, Palais des Congrs de Paris,
Paris, France
[10] Nakamura, A.; Taniguchi, H.; Yokoyama, A.: 1100kV AC Transmission Project in Japan, IEC/CIGR UHV
Symposium, Beijing, China, July 18-21, 2007, paper 1-2-1
[11] YAMAGATA, Y., KOBAYASHI, A.;YOSHIZUMI, T.; TSUBAKI, T.: Development of 1100kV GIS Gas Circuit
Breakers, Disconnectors and High Speed Grounding Switches, CIGR Report 13-304, 36th CIGR Session, Aug.
25 to 31, 1996, Palais des Congrs de Paris, Paris, France
[12] Kobayashi, T.; Ono, M.; Shimomura, T.; Yokota, T.: Basic design / specifications of GIS for UHV AC and its
verification test at site, IEC/CIGR UHV Symposium, Beijing, China, July 18-21, 2007, paper 2-3-3
[13] Kobayashi, A.; Murayama, Y.; Ohyama, S.; Kan, M.: Development and Verification Test of 1100 kV Gas
Insulated Switchgear, Int. Workshop of UHV Transmission Technology, Beijing, China, 2005, pp. 94-99
[14] Zaima, E.; Neumann, C.: Insulation Coordination for UHV AC Systems based on Surge Arrester Application
(CIGR C4.306), IEC CIGR International Symposium on International Standards for UHV Transmission, New
Delhi, India, 2009, pp. 108-118
[15] Riechert, U.; Holaus, W.: Ultra High Voltage Gas-Insulated Switchgear A Technology Milestone, Euro.
Trans. Electr. Power (2011), (wileyonlinelibrary.com). DOI: 10.1002/etep.582
Page 274
TypethenameofTB
[16] Riechert, U.; Krsi, U.; Sologuren, D.: Very Fast Transient Overvoltages during Switching of Bus-Charging
Currents by 1100 kV Disconnector; 43rd CIGR Session, 2010, Paris, France CIGR Report A3-107
[17] Szewczyk, M.; Piasecki, W.; Stosur, M.; Riechert, U.; Kostovic, J.: Impact of Disconnector Design on Very
Fast Transient Overvoltages in Gas-Insulated UHV Switchgear, Proceedings of 17th International Symposium on
High Voltage Engineering (ISH), August 22nd 26th, Hannover, Germany
[18] CIGR Working Group 15.03: GIS Insulation Properties in Case of VFT and DC Stress, 36th CIGR Session,
1996, Paris, France, CIGR Report 15-201
[19] CIGR Working Group 33/13-09: Monograph on GIS Very Fast Transients, Brochure No. 35, July 1989
[20] Ikebe, K.; Shimomura, T. ; Iyoda, I.: Simulation Technology for 1000 kV Power Systems, Mitsubishi Electric
Advance, Vol.77/Dec. (1996)
Page 275
TypethenameofTB
3)
Page 276
TypethenameofTB
Highest
Voltage for
equipment
winding
Um kV
Full Wave
Lightning
Impulse
(LI) kV
Chopped Wave
Lightning
Impulse
(LIC) kV
Switching
impulse
(SI) kV
1950
2250
2250
2145
2475
2475
1425
1800
1800
1100
1200
Applied
voltage or line
terminal AC
withstand
(AV) (LTAC)
kV
-
Clause 7.3.3 Tests on Transformers with Um > 170 kV, 7.3.3.1 Routine tests
4) Induced voltage with partial discharge measurement (IVPD)
A test shall be carried out using the method given in section 11.3, with and enhancement (phase to earth) voltage
level of 1.8 Ur / 3 and a one hour PD measurement voltage of 1.58 Ur / 3. Alternative higher voltage levels
may be used if specified by the purchaser, in particular an enhancement voltage of 3 Um / 3 and a one hour
PD measurement voltage of 1.5 Um / 3 may be used if higher.
Clause 11.3 Induced voltage with partial discharge measurement (IVPD), 11.3.2 Test duration and
frequency
The test time at the enhancement voltage shall be 60 s in case Um 800 kV and 300 s in case Um > 800 kV for any
test frequency up to and including twice the rated frequency, unless otherwise specified. When the test frequency
exceeds twice the rated frequency, the test time in seconds of the test shall be:
600
rated frequency
, but not less than 75 s for Um > 800 kV
test frequency
The duration of the test, except for the enhancement level, shall be independent of the test frequency
Clause 11.3.3 Test sequence: Figure 1 in the CDV (14/716/CDV)
Test time
Enhancement voltage
60 min.
1.2 Ur/ 3
One hour PD
1.2
0.4
U r/ 3
U r/ 3
Page 277
TypethenameofTB
This clause describes the theory for the long-duration induced AC voltage test.
The long-duration induced AC voltage test can be used to verify the insulation performance of the equipment
across its lifespan and the insulation strength when subjected to temporary AC overvoltages with high accuracy.
The test is understood to be effective for very high voltage class of equipments. In IEC 60076-3, long-duration
induced AC voltage test are specified for equipment with Um > 170kV. The long-duration induced AC voltage test is
also specified for equipment with nominal system voltage of 187 kV and above in Japan.
The main features of the long-duration induced AC voltage test as a quality control test are listed below:
(1) The long-duration induced AC voltage test is totally low level of the test voltages and is applied over a long
period, so the equipment can be absolved of the damage due to an undue stress from the test. It is intended to
verify the insulation performance of equipment across its lifespan and its insulation strength against temporary AC
over voltages.
(2) The test voltages and duration should be calculated using temporary AC overvoltages and the insulation
characteristics of the equipment (V-t insulation characteristics) for evaluation of reliability of equipment using the
Weibull distribution function.
(3) The pattern of the test voltages, as shown Figure 8.4.1, is a combination of the lower voltages, V1 and V3, and
relatively higher voltage, V2. V2 is applied over a short period whereas V3 is applied over a longer period. V1 is set
slightly higher with a longer period in Japan, and a shorter period in China. The aim of V1 is to monitor the test
voltage and keep it constant. It is not intended to verify the insulation performance of the equipment.
(4) V2 is used to verify the insulation strength against temporary overvoltages and V3 is intended to verify the
insulation performance against the operating voltage across the entire lifespan of the equipment. Partial discharge
measurement should be done along the entire span of the test period.
Page 278
TypethenameofTB
where, m=n x a
m: parameter of voltage form
n: gradient of V-t characteristic
a: parameter of time form
If the probability with the no partial discharge in the equipment in a test is Rt, the stress voltage in service is Vs, its
service period is ts and I iss applied N number of times, and Rs is the probability with no partial discharge in
operation, then the test voltage, Vt and its applied period, tt should be fixed. This relationship is be shown in
Equation (b) as
Vt m t ta N Vsm t sa
lnR t
--- (b)
lnR s
8.4.1.4 EXAMPLE OF EVALUATION FOR THE UHV TESTING VOLTAGES (JAPANESE CASE)
The calculated example for the UHV testing voltages in Japan is shown below.
(1) Characteristic of insulation for the equipment
Figure 8.4.2 and Table 8.4.1 show the necessary data of the insulation characteristics for the calculation of the
testvoltages [3] [6].
It should be applied to the insulation characteristics on the barrier-oil-duct models to evaluate the UHV testing
voltages.
120
PD inception time
50% PD time
110
* A number in circle
represents a number of
models without PD
n=44.2
n=323
100
(70 days)
90
1m
80
10-3
1h
1d 4d
10-2 10-1
100
101
102
103
104
105
106
107
108
Page 279
TypethenameofTB
Structure part of
transformer
Turn-to-turn
44.9
0.42
18.9
292
0.31
90.2
Coil-to-coil
39.9
0.40
16.0
200
0.35
70.0
Barrier-oil-duct
44.2
0.39
17.2
323
0.38
122.7
Operating
voltage
1.5pu
1.0pu
Parameter
One-line ground fault
Over-voltage Vs
1.1pu
1.1pu
Duration Ts
0.1s
2s
0.83s
Frequency N
90
10
30years
-
* Considering the influence to the insulation characteristic by the voltage shape of trapezoid
[3]
1.5E
Applying periods
V2
3E
V3
t1
t2
t3
1.5E
5 min
5 min
60 min
Table 8.4.3 Long-duration AC voltage test mode for UHV Power Transformers
8.4.2 OTHER SUBSTATION EQUIPMENT
Page 280
TypethenameofTB
the open switching device and/or isolating distance can be done with two power-frequency sources in out-of-phase
condition across the gap. In both cases the test voltage has to be applied for 1 minute. No flashovers are allowed.
The test procedure is described in IEC 60060-1.
Since the short-duration power-frequency voltage test is not necessary for insulation coordination purposes, it is
used more for quality assurance concerning line-to-earth voltages. The values serve as a basis for routine tests
and they are also the pre-stress voltage for the partial discharge measurement.
Concerning the AC test across the open switching device and/or isolating distance, the phase-opposition conditions
may be the background, but this is nowhere stated. For UHV systems (see Table 8.4.4), the values for lightning
impulse withstand voltages are considerably higher in comparison to the test voltages of the lower voltage systems.
Beside the pure power-frequency voltage tests, the AC voltage is also used for the combined voltage tests for the
longitudinal insulation - across the open gap of the switching devices. In this case the impulse voltage is
synchronized in the way that the peak-value of the impulse reach the peak-value of the power-frequency voltage in
the opposite polarity. The sum of both voltages is the test voltage across the gap.
In the combined voltage test, the test voltage is generally 0.7 p.u. (1 p.u. = peak of line-to-ground voltage) of the full
lightning impulse level and full 1 p.u. against a reduced switching impulse level.
Highest
voltage of
equipment
Ur
(kVr.m.s.
value)
Phase-toearth and
between
phases
1100
1200
Across
open
switching
device
and/or
isolating
distance
Phase-toearth and
across
open
switching
device
1100
Between
phases
Across
isolating
distance
Phase-toearth and
between
phases
Across open
switching
device and/or
isolating
distance
1550
2635
1550+(900)
2250
2250+(630)
1100+(635)
1800
2880
1675+(900)
2400
2400+(630)
1200
1800
2970
2400
2400+(685)
2550
2550+(685)
1100
1200
1675+(980)
1200+(695)
1950
3120
Table 8.4.4 Extension of highest voltages of equipment for UHV-Systems in the IEC
62271-1 A1 Ed.1: [7]
The situation in the standard for GIS (IEC 62271-203) is a little bit different concerning the power-frequency test
voltages. Generally the power-frequency test voltages are higher for line to ground and across the open switching
devices in comparison to IEC 62271-1. The reason for choosing higher test values was to achive higher reliability.
The values have been fixed with the factor of 0.45 in relation to the lightning impulse withstand voltage values.
Combined voltage tests in IEC 62271-203 are similar to those of IEC 62271-1.
Page 281
TypethenameofTB
The latest revision of IEC 62271-203 does not have any standardised values for UHV. The standard only lists
insulation withstand values as applied in some countries. Compared with systems in the lower voltage ranges (as
described before) the withstand values of the different countries are increased especially for the longitudinal
insulation.
Table 8.4.5 Insulation levels used for GIS in UHV systems (Annex G) [8]
8.4.2.2 LONG-DURATION OVER VOLTAGE TEST FOR GIS, CB, ETC. (JAPANESE CASE)
In Japan, gas insulated apparatus (GIS, GCB, etc.) were also examined with the same technique as abovementioned transformer [3]-[5]. The outline is described below.
(1) V-t characteristics of GIS
Tests were carried out as part of an investigation of the long-term V-t characteristics by using a 72kV single-phase
bus bar as shown in Figure 8.4.3[9]. Figure 8.4.3 also shows the results of the long-term V-t characteristics of 72kV
busbar. As the 50% breakdown V-t characteristics, they were expressed as inversely proportional to 82nd power of
the time. These results demonstrated that gas-gap insulation did not much deteriorate even after long-term voltage
application.
Table 8.4.6 show the necessary data of the insulation characteristics for the calculation of the testing voltages
obtained from the V-t characteristics. In addition, with GIS, there are no differences in short- and long-time
characteristics as the only insulation medium is gas.
Page 282
TypethenameofTB
n = 82
GIS
82
0.232
19
82
0.232
19
Operating
voltage
Parameter
One-line ground fault
Over-voltage Vs
1.1pu
1.1pu
1.5pu
1.0pu
Duration Ts
0.1s
2.0s
0.2s
30years
Frequency N
90
10
Page 283
TypethenameofTB
For GIS, because the insulation characteristics of the equipment do not differ between short- and long-term zones,
verification as long-term zones seem to be permitted with short-term (several minutes) voltage application.
However, long-term test are preferable since problems would occur under relatively long-term voltage application.
Test voltages
V1
GIS
1.5E
V2
3E
Applying periods
V3
t1
t2
t3
1.5E
30min
1min
30min
Page 284
TypethenameofTB
8.5 Conclusion
Section 8.1 describes the procedure for insulation coordination, specifying the simplified method and the safety
factor. Section 8.2 shows concretely the process of determining the LIWV and SIWV, and verification of SIWV for
TOV. Section 8.3 covers the consideration of insulation coordination for VFTO and Section 8.4 explained the power
frequency voltage tests for substation equipment with reference to the relevant IEC standards.
The simplified method provides a preliminary estimation based on the residual voltage with the nominal discharge
current of a surge arrester. The residual voltage comes from V-I characteristics with the nominal discharge current.
The nominal discharge current for UHV surge arresters is studied in IEC TC37 MT4. Detailed computer simulations
are generally carried out for UHV systems.
The safety factor compensates for the differences in the equipment assembly and the dispersion in the product
quality. The current safety factor in IEC 60071-2 was adopted 20 years ago as an overall value obtained from field
experiences.
The safety factors used in practice range from 1.00 to 1.32 and those recommended in IEC 60071-2 are 1.05 or
1.15. Data on insulation deterioration characteristics due to ageing have been accumulated and also the quality
control, and assembly and installation technology of equipment components and equipment manufacturing have
been significantly improved. Based on such a situation, it might be one option to allow some range for the
recommended value of the safety factor instead of using it a constant fixed value.
According to IEC 60071-1 TOVs should be basically covered by SIWV. Following this way of IEC 60071-1 the
standard switching impulse voltages equivalent to TOV range from 2383 kV to 2566 kV in the UHV projects and
they require a standard switching impulse withstand voltages of at least 2400 kV in UHV systems. Since SIWVs
greater than LIWVs are economically and technically unrealistic and not feasible, the power-frequency voltage test
as an alternative test is available to verify the withstand capability concerning TOV. The power-frequency withstand
voltage tests are provided by the relevant apparatus committees.
In cases where the calculated VFTO caused by GIS-disconnector switching is higher than the withstand level,
special measures to mitigate the VFTO are required. Japanese and Chinese experiences are introduced. Generally
a three-step procedure is proposed as a general insulation coordination approach: 1) Calculation of VFTO (peak
and rise time), 2) Comparison of calculated VFTO values with LIWV level by using coordination factor, safety factor
and test conversion factor, and 3) Decision and definition concerning the necessity of measures to reduce the
VFTO level.
Japanese and Chinese UHV projects apply the long-duration induced AC voltage test (ACLD) for UHV power
transformers, whose time sequence is the same for both projects. The test voltages and time period are estimated
by the Weibull distribution function in consideration of reliability. In the meantime, the maintenance work of the IEC
60076-3 has been started and the description of the CDV of IEC 60076-3 Ed.3 (August, 2012) is introduced in the
chapter.
For other equipment the standard Common clauses of substation equipment IEC 62271-1 specifies a shortduration power-frequency withstand voltage for phase-to-earth and between phases and for across the open
switching devices. The test across the open switching device can be done with two power-frequency sources in
out-of-phase condition across the gap. In both cases the test voltage has to be applied for 1 minute. No flashover is
allowed. According to national requirements in Japan, gas insulated apparatus (GIS, GCB, etc.) were also
examined with the same technique as the above described procedure for transformers.
Page 285
TypethenameofTB
9.1 Recent Practices on insulation coordination for UHV and 800 kV system
Economical and highly reliable transmission lines and substations equipment with environmental considerations
are essential in the UHV system. UHV clearance in the air may require much longer distance than those applied to
EHV systems, because the dielectric strength for switching overvoltage does not increase linearly with an increase
of the air gap. Therefore reducing the size of transmission lines and substation equipment are practical
countermeasures.
In Chinese, Indian, Japanese UHV projects, suppressing overvoltage by higher performance surge arresters is a
common countermeasure, and additional countermeasures, such as suppressing overvoltage by the circuit
breakers with closing and/or opening with pre-insertion resistors, are adopted in each project. In these projects,
overvoltages are simulated by the latest analyzing technology such as EMTP.
Page 286
TypethenameofTB
Page 287
TypethenameofTB
Page 288
TypethenameofTB
(4) VFTO
In cases where the calculated VFTO level due to GIS-disconnector switching is higher than the withstand level,
special measures to mitigate the VFTO are required. Japanese and Chinese experiences and approaches are
presented A three-step procedure is proposed as a general insulation coordination approach: 1) Calculation of
VFTO (peak and rise time), 2) Comparison of calculated VFTO values with LIWV level by using coordination factor,
safety factor and test conversion factor (the exact values of these factors cannot be given at the moment), and 3)
Decision and definition concerning the necessity of measures to reduce the VFTO level.
(5) Power frequency voltage test
Japanese, Chinese and Indian UHV projects apply the long-duration induced AC voltage test (ACLD) for UHV
power transformers, whose time sequence is nearly same for these projects. According to IEC 60071-1 TOVs
should be basically covered by SIWV. Following this way of IEC 60071-1 the standard switching impulse voltages
equivalent to TOV range from 2383 kV to 2566 kV in the UHV projects and they require a standard switching
impulse withstand voltages of at least 2400 kV in UHV systems. Since SIWVs greater than LIWVs are economically
and technically unrealistic and not feasible, the power-frequency voltage test as an alternative test is available to
verify the withstand capability concerning TOV. The power-frequency withstand voltage tests are provided by the
relevant apparatus committees. Japanese, Chinese and Indian UHV projects apply the long-duration induced AC
voltage test (ACLD) for UHV power transformers, whose time sequence is nearly same for these projects.
Page 289