The International Symposium on High-Voltage Technique "Hfler's Days", 78 November 2013, Portoroz, Slovenia.

Transients Caused by Uncontrolled and

Controlled Switching of Circuit Breakers
Ivo Uglei, Boidar Filipovi-Gri and Sreko Boji

power quality and suppression of transients in transmission

AbstractThis paper describes transients caused by and distribution systems [4].
uncontrolled and controlled switching of high-voltage circuit This paper describes switching transients caused by
breakers. Inrush currents due to shunt reactor energization were uncontrolled and controlled switching of circuit breaker.
analyzed. Switching-off, energization and auto-reclosure of
Switching of shunt reactor and no-load transmission line is
unloaded 400 kV transmission line is presented. Transient
recovery voltage and voltage distribution between breaking analyzed using EMTP-RV software. Field measurements are
chambers of circuit breaker are investigated. Calculation models presented and compared with simulation results.
were developed in EMTP-RV software and some field
measurements are presented. Results show that controlled II. TRANSIENTS CAUSED BY SWITCHING OF SHUNT REACTOR
switching significantly reduces inrush currents and suppresses
switching overvoltages. The function of shunt reactors in transmission networks is to
consume the excess reactive power generated by overhead
Index Termscircuit breaker, controlled and uncontrolled lines under low-load conditions, and thereby stabilize the
switching, shunt reactor, inrush current, no-load 400 kV system voltage. They are quite often switched in and out on a
transmission line, simulations, measurements. daily basis, following the load situation in the system.
Energization of a shunt reactor may cause inrush currents with
I. INTRODUCTION high asymmetry and long time constants. The actual magnitude
of the inrush current is quite dependent on the range of
U NCONTROLLED switching of shunt reactors, shunt
capacitors and transmission lines may cause severe
transients such as high overvoltages or high inrush currents
linearity of the reactor core and on the time instant of circuit
breaker pole operation with respect to the reference signal.
Switching operations at unfavorable instants can cause
[1]. Conventional countermeasures such as pre-insertion
currents that may reach high magnitudes and have long time
resistors, damping reactors or resistors, or arresters are used to
constants. At shunt reactors with solidly grounded neutral,
limit the magnitude and effect of the switching transients, after unsymmetrical currents cause zero-sequence current flow
they have occurred. In addition, system and equipment which can activate zero-sequence current relays [5].
insulation may be upgraded to withstand the stresses. These
methods, however, may be inefficient, unreliable or expensive, A. Model Description
and do not treat the root of the problem [2]. The inrush currents due to energization of 150 MVAr shunt
Controlled switching is a method for eliminating harmful reactor in 110 kV network have been observed. Both
transients via time controlled switching operations. Closing or uncontrolled and controlled circuit breaker switching is
opening commands to the circuit breaker are delayed in such a simulated. Shunt reactor manufacturer data are shown in Table
way that switching will occur at the optimum time instant I.
TABLE I 150 MVA SHUNT RECTOR DATA
related to the voltage phase angle. Controlled switching has
Rated power 150 MVAr
become an economical substitute for a closing resistor and is
Rated frequency 50 Hz
commonly used to reduce switching surges. The number of Rated voltage 110 kV
installations using controlled switching has increased rapidly Rated current 787 A
due to satisfactory service performance since the late 1990s Core type Five limb
[3]. Currently, it is often specified for shunt capacitor and Total losses (at 110 kV) 363 kW
shunt reactor banks because it can provide several economic Impedance in saturation 38
benefits such as elimination of closing resistors and extension Zero sequence impedance 80,7 per phase
of a circuit breaker nozzle and contact maintenance interval. It Stationary magnetic flux 1,32 T
also provides various technical benefits such as improved Capacitance of winding to
1,36 nF per phase
ground

I. Uglei and B. Filipovi-Gri are with the University of Zagreb,

Faculty of Electrical Engineering and Computing, Unska 3, 10000 Zagreb, Calculation of inrush currents requires an adequate
Croatia (e-mail: [email protected]; [email protected]). modeling of the reactor nonlinear flux-current curve. The
S. Boji is with Energy Institute Inc., Ul. grada Vukovara 37, 10000 nonlinearity is caused by the magnetizing characteristics of the
Zagreb, Croatia (e-mail: [email protected]).
shunt reactor iron core. Recorded RMS voltage-current curve
 The International Symposium on High-Voltage Technique "Hfler's Days", 78 November 2013, Portoroz, Slovenia. 2

obtained from manufacturer test data is converted into switching occurs at an instant near the voltage zero-crossing in
instantaneous flux-current saturation curve (Fig. 1) which is phase A, since it results with the maximum DC component of
used in the nonlinear inductance model in EMTP-RV [6] and current.
approximated with two segments (linear area A-B, below
knee of the saturation curve and saturation area B-C).
550
500 L'air=96.997 mH C'
450
400 B'

U (kV)
350
(Vs)

300
250
Ln=256.77 mH
200
150
100
50 A'
0
0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000
i (A) Fig. 3. Shunt reactor voltages in case of reactor uncontrolled energization
Fig. 1. Instantaneous flux-current saturation curve of 150 MVAr shunt
reactor

Model for analysis of shunt reactor switching in EMTP-RV is

shown in Figure 2.
A m2
+VM
?v
BUS1
R2

a
+

RCu
R1

?i

AC1 b
+

RL1 SW1 ?i
+ + RFe
+

6.66ms|15|0 Lnonl1
110kVRMSLL /_0 10ms|15|0
c
?if
L
13.33ms|15|0

B m3
+VM
?v
R4
+

RCu
+
L1

3L0
R3

Tr0_2
?i
+

?i

RFe + Fig. 4. Shunt reactor currents in case of reactor uncontrolled energization:

+

Lnonl2
1 2
?i

?if
L IAmax= 3455 A (3.10 p.u.), IBmax=-2982 A (2.68 p.u.), ICmax=-2494 A
1
(2.24 p.u.)
m4
C +VM
?v Conducted simulation shows that transient inrush current
R6
+

RCu with amplitude of 3.10 p.u. and high DC component can last
Tr0_1
R5

?i

up to 3.5 seconds (Fig. 5). This could cause difficulties, such

+

+
RFe
+

Lnonl3
1 2
?if
L as unwanted operation of the overcurrent relay protection.
1
+
?i

ZVJ
A
m1

Fig. 2. Model for analysis of shunt reactor switching in EMTP-RV

Each phase of a three phase shunt reactor is modeled as a

nonlinear inductance with serially connected resistance
RCu=146.5 m, representing copper losses and parallel
connected RFe=133.3 k, representing iron losses. Magnetic
coupling among the three star connected phases is represented
with zero sequence inductance L0=257 mH which provides a
path for the zero sequence current [7]. Equivalent 110 kV
network is represented with positive (R1=265 m, L1=7 mH)
and zero (R0=216 m, L0=5,7 mH) sequence impedances,
determined from single-phase and three-phase short circuit Fig. 5. Shunt reactor current in phase A in case of reactor uncontrolled
energization
currents: I sc1 = 33.8 kA 83.1o , I sc3 = 31.7 kA 83.1o .

B. Uncontrolled Energization Zero sequence current occurs in case of uncontrolled reactor

energization (Fig. 6) as a consequence of asymmetry. This may
By uncontrolled energization the following instants of pole cause the false operation of relay protection, used for detecting
closing were considered: tA=15 ms, tB=13 ms and tC=17 ms. single phase-to-ground faults.
Figures 3 and 4 show shunt reactor voltages and currents,
respectively. The highest inrush current happened when
 The International Symposium on High-Voltage Technique "Hfler's Days", 78 November 2013, Portoroz, Slovenia. 3

Magnetic flux (Vs)

I (A)

Fig. 9. Magnetic flux in case of reactor controlled energization

Fig. 6. Zero-sequence current in case of reactor uncontrolled energization

Imax=-2052 A (1.84 p.u.)

C. Controlled Energization
Figures 7 and 8 show shunt reactor voltages and currents in
case of controlled energization at optimum instants of circuit
breaker poles closing at peak voltages (tA=10 ms, tB=6.66 ms
and tC=13.33 ms).

Fig. 10. The distribution of the magnetic flux in the five limb core of shunt
reactor - phase A (1), B (2) and C (3)

Conducted simulation shows that amplitudes and DC

components of inrush current (Fig 11) and zero-sequence
U (kV)

current (Fig 12) are significantly lower in case of controlled

switching. As a consequence, successfuly controlled switching
reduces the mechanical and electromagnetic stresses of the
high voltage equipment and also prevents the unwanted
operation of relay protection.

Fig. 7. Shunt reactor voltages in case of reactor controlled energization

I (A)
I (A)

t (s)
Fig. 11. Shunt reactor current in phase A

Fig. 8. Shunt reactor currents in case of reactor controlled energization,

IAmax=-1260 A (1.13 p.u.), IBmax=1090 A (0.98 p.u.), ICmax=1123 A (1.01 p.u.)

The current in phase A (Fig. 8) is slightly larger than in the

other two phases, due to the appearance of the DC component,
I (A)

which is caused by initial magnetic flux in the core limb of that

phase at the moment of energization. This initial magnetic flux
is a part of a magnetic flux from the phase B, which is firstly
being switched on (Fig. 9). The final distribution of the
magnetic flux in the reactor core is shown in Fig. 10. Due to
the air gaps utilized in shunt reactor core there are no severe
saturation effects [8]. Fig. 11. Zero-sequence current, Izmax=1122 A (1.01 p.u.)
 The International Symposium on High-Voltage Technique "Hfler's Days", 78 November 2013, Portoroz, Slovenia. 4

III. TRANSIENTS CAUSED BY SWITCHING UNLOADED 400 KV B. Model Description

TRANSMISSION LINE Model for the analysis of switching the 231 km long 400 kV
unloaded transmission line [9] is shown in Fig. 14.
A. Measurements of switching transients
The equipment in high voltage substation was represented
The measurements of switching transients caused by by surge capacitances, whereas transmission lines, busbars and
uncontrolled switching of the 231 km long line were connecting leads by a frequency depending line model [10].
performed. Figures 12 and 13 show recorded TRV and phase The phase transpositions of the transmission lines have been
voltages on line side in case of switching-off line by circuit
taken into account. The MO surge arresters were modeled with
breaker equipped with grading capacitors.
nonlinear U-I curves for switching overvoltages.
800 The model of CB with two breaking chambers for switching
600 the transmission line 2 is shown in Fig. 15.
Napon izmeu kontakata prekidaa (kV)

400

200
+ +
0.5 0.5
0 0.010nF 0.010nF
+ + + +
-200
1uH 1uH
-400

-600
SW8 ?vi SW9 ?vi
-800
+ +
-0.05 0 0.05 0.1 -1ms|10ms|0 -1ms|10ms|0
Vrijeme (s)
Fig. 12. Recorded TRV in case of regular switching-off: UL1= -686 kV (2.08
+

+
p.u.),UL2=-657 kV (1.99 p.u.), UL3=641 kV (1.94 p.u.)

0.01nF 0.06nF 0.01nF

400
C14
Fig .15. Model of circuit breaker with C11
two braking chambers C13
300

200
The capacitance between the open contacts of breaking
Napon na NMT (kV)

100
chambers is 10 pF and inherent earth capacitances were taken
0
into account as depicted in Fig. 15.
-100

-200
The equivalent networks were represented with a voltage
-300
source in series with sequences impedances, which are
-400
-0.04 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 0.04 0.05 obtained from short circuit currents in case of switching state
Vrijeme (s)
prior to a fault (Table II).
Fig. 13. Recorded phase voltages on line-side after regular switching-off
UL1=360 kV (1.09 p.u.), UL2= 351 kV (1.06 p.u.), UL3= -348 kV (1.06 p.u.)

Capacitor voltage transformer 4400 pF

Capacitor voltage transformer 4400 pF
Equipment in transformer bay: Current transformer 680 pF
Current transformer 680 pF
Autotransformer 2700 pF Disconnector 200 pF
Disconnector 200 pF
Current transformer 680 pF Post insulators in line bay 2x120 pF
Post insulators in line bay 2x120 pF
+

17.960nF
+

Circuit breaker 60 pF 5.52nF

+

0.68nF
Disconnector 200 pF Transmission line 2
8 x post insulator 120 pF
Capacitor voltage transformer 4400 pF a a
a a SW8 ?tT b + FD b
b + FD b
+ c c
c c 10ms|1|0
+

12.640nF
Equivalent network 110 kV Connecting leads 510 m
ZnO +

ZnO +

672000 ?vi
RL3 684000 ?vi
AC3
+ ZnO1
+

ZnO3
418kVRMSLL /_0 Main busbars 630 m
a a
AC2 RL1 b + FD b 400 kV network equivalent
+ c c Transmission line 1
+

RL5 418kVRMSLL /_0

a a
418kVRMSLL /_0 +
b + FD b
+

c c
AC5
a a 418kVRMSLL /_0
RL6
+

4.08nF 34x120 pF b b
ZnO +

+
c c
+

672000 ?vi Busbar post insulators

ZnO4 AC6

Transmission line 3 400 kV network equivalent

400 kV network equivalent Transmission line 4
RL9 418kVRMSLL /_0
AC4 RL4 a a a a
+ + +
+ b b b
+

FD FD
+

c c c c
AC9
418kVRMSLL /_0

Fig. 14. Model for analyzes of circuit breaker switching no-load 400 kV transmission line in EMTP-RV
 The International Symposium on High-Voltage Technique "Hfler's Days", 78 November 2013, Portoroz, Slovenia. 5

TABLE II between breaking chambers is very important for CB dielectric

SHORT CIRCUIT CURRENTS IN 400 KV SUBSTATION
stresses during switching operations. Simulation results show
Connections: I3ph (kA/) I1ph (kA/) pretty unequal voltage distribution between breaking chambers
3.4/-85.5 2.4/-82.2 in phase L1 where the maximum peak of TRV occurs (Fig.
Transmission line 1 (86.3 km)
3.4/-85.5 2.4/-82.2 18).
Transmission line 2 (231 km) 2.4/84.8 1.8/-80.0
Transmission line 3 (91.5 km) 4.1/-84.9 3.3/-79.9 m10a@vn
m10a@vn m7a@vn
SW8@vb
87
Transmission line 4 (152 km) 3.9/84.9 3.0/-80.1 7

Power transformer 6
1.2/-81.5 1.3/-84.8 6
TR 1 (400/110 kV) 5
5
Power transformer
1.1/-81.5 1.3/-84.8 4
4
TR 2 (400/110 kV)
33
Total: 19.5/-84.7 15.6/-81.6
22

C. Switching-off unloaded 400 kV transmission line

11

When an unloaded line is regularly switched-off the electric 0

0
arc distinguishing occurs at current natural zero-crossing. Fig. 1
1 0
16 shows phase voltages on line-side after regular switching- 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1

off. Voltage is highest in the phase which is firstly switched-

off due to the electromagnetic coupling of the other phases and Fig. 18. Distribution of CB recovery voltage between breaking chambers
due to Ferranti effect.
The voltage distribution between the breaking chambers
L3
m7c@vn L2 L1
m7b@vn m7a@vn could be improved by installing the grading capacitors, which
4004
are especially important in cases when switching-off relatively
30034
3
long lines. Further analyses show the influence of grading
2002
2 capacitors of 500 pF on voltage distribution. Simulation results
1001
1 show significant improvement of voltage distribution and
000 reduction of TRV amplitude (Fig. 19).
-10011
-20022 4.5
m10a@vn
m10a@vn m7a@vn
8
4.5
-30033 4
4
-40044 3.5
5 3.5
-5005 00 0.01
0.01
0.02
0.02
0.03
0.03
0.04
0.04
0.05
0.05
0.06
0.06
0.07
0.07
0.08
0.08
0.09
0.09
0.1
0.1
3
0 10 20 30 50 4060 70 80
Time (s) 90 100 3

2.5
2.5
t (ms)
22
Fig. 16. Phase voltages on line-side after regular switching-off
1.5
1.5

11
Since the capacitor voltage transformers are installed on 0.5
0.5
both sides of the line the discharging of trapped charge is slow. 0
0
Such discharging depends on weather conditions, mainly on 0.5
0.5 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
humidity. So, the trapped charge has a very significant 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1

influence on CB transient recovery voltage (Fig. 17).

Fig. 19. Distribution of CB recovery voltage between breaking chambers in
case with grading capacitors
m10c@vnm7c@vn m10b@vnm7b@vn m10a@vn
8

8
6 Fig. 20 shows the comparison of calculated TRV peak
6
4
values in cases with and without grading capacitors.
4
2
2,5
2
2,25 Maximum withstand voltage 2,4 p.u. (823 kV) Total TRV
0
0
2
2
2 Without grading capacitors
1,75
(Chamber 1)
1,5
U (p.u.)

4
4
Without grading capacitors
6 1,25 (Chamber 2)
6
1 With grading capacitors
8
8 0
0
0.01
0.01
0.02
0.02
0.03
0.03
0.04
0.04
0.05
0.05
0.06
0.06
0.07
0.07
0.08
0.08
0.09
0.09
0.1
0.1 0,75 (Chamber 1)
0,5 With grading capacitors
(Chamber 2)
0,25
Fig. 17. TRV in case of regular switching-off: UL1=758 kV (2.32 p.u.),
0
UL2=713 kV (2.18 p.u.), UL3=692 kV (2.12 p.u.) 1 2 3
Phase
Besides the peak value of TRV the voltage distribution Fig. 20. Calculated TRV when switching-off unloaded 400 kV line
 The International Symposium on High-Voltage Technique "Hfler's Days", 78 November 2013, Portoroz, Slovenia. 6

The CB should withstand the maximum voltage of 823 kV TABLE III

SWITCHING OVERVOLTAGE AMPLITUDES AT THE END OF UNLOADED LINE
(2.4 p.u.) across the open contacts [11]. Compared to this, the
IN CASE OF ENERGIZATION WITHOUT TRAPPED CHARGE
computed value of recovery voltage is only 10 % smaller Voltage amplitude (kV)
(758 kV/2.2 p.u.). However, the greatest part (nearly 80 %) of Uncontrolled switching Controlled switching
TRV stresses the breaking chamber closer to the substation. UL1 UL2 UL3 UL1 UL2 UL3
Without surge
752 803 723
D. Energization of unloaded 400 kV transmission line without arresters
392 401 397
trapped charge With surge
634 630 632
arresters
Energization and auto-reclosure of long transmission lines
can cause undesirable overvoltages in the transmission Controlled energization at optimum instants of circuit breaker
network, so special overvoltage mitigation measures are poles closing at voltage zero-crossing is simulated. Figure 23
employed to meet the insulation coordination considerations. shows voltages at the end of the line for this case.
The most common practice has been to use metal-oxide surge
arresters and circuit breakers equipped with closing resistors,
but this solution is relatively expensive.
Worst case of uncontrolled energization at peak voltage in
all phases is analyzed. Figure 21 shows voltages at the end of
transmission line without surge arresters in case of
uncontrolled switching.

Fig. 23. Voltages at the end of transmission line in case of controlled

switching

E. Auto-reclosure of unloaded 400 kV transmission line

In the case of transmission line with a capacitive potential
transformers connected at both ends, no leakage path exists for
the trapped charge. Figure 24 shows voltages at the end of
transmission line without surge arresters in case of
Fig. 21. Voltages at the end of transmission line without surge arresters in
case of uncontrolled switching uncontrolled switching.

Figure 22 depicts voltages at the end of transmission line

with surge arresters Ur=342 kV in case of uncontrolled
switching.

Fig. 24. Voltages at the end of transmission line without surge arresters in
case of uncontrolled switching

Contact closing occurs at voltage peak on the source-side of

Fig. 22. Voltages at the end of transmission line with surge arresters in case
of uncontrolled switching the opposite polarity to the trapped charge. This represents the
most severe case of uncontrolled auto-reclosure.
Table III shows the comparison between uncontrolled and Figure 25 shows voltages at the end of transmission line
controlled switching regarding the amplitudes of switching with surge arresters in case of uncontrolled switching. The
overvoltages at the end of the unloaded line. It can be seen that optimum instant of circuit breaker switching is the voltage
switching overvoltages are significantly lower in case of peak on the source-side of the same polarity as the trapped
controlled switching. charge (Fig. 26).
 The International Symposium on High-Voltage Technique "Hfler's Days", 78 November 2013, Portoroz, Slovenia. 7

protection.

REFERENCES
[1] I. Uglei, S. Hutter, B. Filipovi-Gri, M. Krepela, F. Jakl, Transients
Due to Switching of 400 kV Shunt Reactor, International Conference
on Power System Transients (IPST), Rio de Janeiro, Brazil, June 24-28,
2001.
[2] Karcius M. C. Dantas, Washington L. A. Neves, Damsio Fernandes Jr.,
Gustavo A. Cardoso, Luiz C. Fonseca, On Applying Controlled
Switching to Transmission Lines: Case Studies, International
Conference on Power Systems Transients (IPST), Kyoto, Japan June 3-
6, 2009.
[3] CIGRE TF13.00.1, Controlled Switching, State-of-the-Art Survey,
Part 1: ELECTRA, No.162, pp. 65-96, Part 2: ELECTRA No.164, pp.
Fig. 25. Voltages at the end of transmission line with surge arresters in case 3961, 1995.
of uncontrolled switching [4] Mitsubishi Electric Advance: Controlled Switching System, Vol.117,
ISSN 1345-3041, Japan, 2007.
[5] Z. Gaji, B. Hillstrom, F. Meki, HV shunt reactor secrets for
protection engineers, 30th Western Protective Relaying Conference,
Washington, 2003
[6] EMTP-RV, documentation, WEB site www.emtp.com.
[7] Vernieri, J; Barbieri, B; Arnera P: Influence of the representation of the
distribution transformer core configuration on voltages during
unbalanced operations, International Conference on Power System
Transients (IPST), Rio de Janeiro, 2001.
[8] ABB, Controlled Switching, Buyers & Application Guide, Edition 4,
2013.
[9] S. Boji, I. Uglei, B. Filipovi-Gri, Switching Transients in
400 kV Transmission Network due to Circuit Breaker Failure,
International Conference on Power Systems Transients (IPST),
Vancouver, Canada, July 18-20, 2013.
Fig. 26. Voltages at the end of transmission line in case of controlled [10] Ali F. Imece, D. W. Durbak, H. Elahi, S. Kolluri, A. Lux, D. Mader, T. E.
switching McDemott, A. Morched, A. M. Mousa, R. Natarajan, L. Rugeles, and E.
Tarasiewicz, "Modeling guidelines for fast front transients", Report
Table IV shows the comparison between uncontrolled and prepared by the Fast Front Transients Task Force of the IEEE Modeling
controlled switching regarding the amplitudes of switching and Analysis of System Transients Working Group, IEEE Transactions
overvoltages at the end of the unloaded line. It can be seen that on Power Delivery, Vol. 11, No. 1, January 1996.
[11] IEC 62271-100: High-voltage switchgear and controlgear; High-voltage
switching overvoltages are significantly lower in case of alternating-current circuit-breakers, 2003.
controlled switching.
TABLE IV
SWITCHING OVERVOLTAGE AMPLITUDES AT THE END OF UNLOADED LINE
IN CASE OF AUTO-RECLOSURE
Voltage amplitude (kV)
Uncontrolled switching Controlled switching
UL1 UL2 UL3 UL1 UL2 UL3
Without surge
1027 1115 1363
arresters
386 375 368
With surge
663 671 609
arresters

IV. CONCLUSION
This paper describes switching transients caused by
uncontrolled and controlled switching of high-voltage circuit
breaker. Switching of shunt reactor and no-load transmission
line was analyzed using EMTP-RV software.
Amplitudes and DC components of inrush currents and
zero-sequence current are significantly lower in case of reactor
controlled switching compared to uncontrolled switching.
Switching-off, energization and auto-reclosure of unloaded
400 kV transmission line was presented. Controlled switching
generates significantly lower overvoltages.
As a consequence, controlled switching reduces the
mechanical and electromagnetic stresses of the high voltage
equipment and also prevents the unwanted operation of relay