On the Problem of Shunt Reactor Tripping during

Single- and Three-Phase Auto-Reclosing

Daniil A. Panasetsky1,2 , Alexey B. Osak1
1

Melentiev Energy Systems Institute of Russian Academy of Sciences

2
Irkutsk National Research Technical University
Irkutsk, Russia
Email:[email protected]

AbstractThe application of shunt reactors (both controlled

and uncontrolled) on HVAC overhead transmission lines improves operational characteristics, but at the same time it brings
the line closer to the resonance. Operation of the shunt reactorcompensated transmission line that is close to the resonance,
causes a large number of problems which in particular can
be solved by temporary tripping of the reactor(s). The article
focuses on some aspects of the SF6 shunt reactor circuit breaker
operations during one-phase and three-phase auto-reclosing. In
particular, the problem of accident-free shunt reactor tripping,
as well as possible ways to solve it are discussed.
Index Termscircuit breakers, energy dissipation, power system simulation, short-circuit currents, transmission lines, zero
current switching.

I. INTRODUCTION
Shunt reactors (SR) (both controlled and uncontrolled) in
HVAC transmission lines are applied to compensate for the
excess reactive power. Being very important components of
electric power systems, they have to be protected against abnormal conditions. From this perspective, the most crucial element is a reactor circuit breaker (CB) whose failure may lead
to serious consequences not only for the nearby equipment,
but for the power system as a whole. The SR circuit breaker
switching characteristics depend on the installation method.
In [1] two most typical cases were studied and compared
installation at line terminals and at a substation busbar.
Reactors compensating the reactive power of a long HVAC
transmission line are usually switched on and off simultaneously with the line, their breakers are connected directly to the
line and not to the substation bus bars (Fig.1). On the one hand,
the application of SR improves operational characteristics, but
on the other hand, it brings the line closer to the resonance.
Operation of the line that is close to the resonance, causes a
large number of problems associated with overvoltages, and
with the necessity for small aperiodic current tripping. The
problem of small aperiodic current tripping is particularly
relevant for the modern SF6 circuit breakers [2], especially of
auto-compression type. The problem of small aperiodic current
tripping may occur during single-phase (SPAR) or three-phase
This work was supported by the Russian Scientific Foundation under project
Development of an intelligent system for preventing large-scale emergencies
in power systems, Grant No.14-19-00054.

(TPAR) auto-reclosing. The resonance overvoltage may occur

only during SPAR.
One of the most effective solutions to both problems is
the resonance detuning by shutting down the shunt reactor(s)
during TPAR and SPAR. However, as is shown below, the
tripping of the shunt reactor(s) in this situation may also be
fraught with difficulties, related to the possible occurrence
of not-breaking currents with a high content of aperiodic
components flowing through the CB of the reactor. And, as
will be shown later, apparently, none of the standard technical
solutions for linear SF6 CB can be directly applied to the
identified problem for reactor SF6 CB. Thus, the objective of
the paper can be summarized as follows:
1) Attract the attention of technicians and equipment
manufacturers. The final answer to the question about
the need for special measures can be provided only by
full-scale experiments, but not by mathematical models.
If necessary it is proposed to introduce a new CB
parameter that would characterize the maximum amount
of energy that can be safely dissipated by one pole of
the breaker.
2) Offer an alternative solution and discuss its advantages and disadvantages. It is proposed to solve the
problem by introducing a power resistor unit into the
circuit of the SR.
II. OPERATING FEATURES OF COMPENSATED LINES
Let us briefly explain the necessity for tripping the SR of
the compensated transmission line during SPAR and TPAR.
The degree of compensation for capacitive reactance of the
transmission line using the SR is defined by the following
formula [3]:

Fig. 1. The test system. 500 kV transmission line

XT L
K=
XSRT

(1)

where XSRT = XSR /N is the total inductive reactance of

N reactors, connected to the transmission line; XT L - total
capacitive reactance of the transmission line. If K is close to
1 then the transmission line is close to the resonance.
Further analysis will be performed using a 500 kV transmission line, presented in Figure 1. Studies were carried out using
MATLAB/Simulink. In our studies we used the simplified
grounded wye connected model of the 160 MVAR shunt reactor with the following parameters [4]: R = 12 M /phase,
L = 5.3 H/phase, r = 30 /phase, C = 14 nF/phase.
There are two main problems associated with the disconnection of compensated lines: the problem of aperiodic current
tripping and the problem of resonance overvoltage during
SPAR.
A. Aperiodic current tripping
This problem is particularly relevant for the modern SF6
circuit breakers, especially of auto-compression type. It is
caused by the fact that the tripped currents of a compensated
line do not cross zero for a long time because of the higher
aperiodic component as compared to the periodic one [5], [6].
The following conditions should be met [7]:
1) there is at least one unfaulted phase in the compensated
line which is switched on only from one side;
2) the phase is switched on at the point closer to voltage
zero crossing;
3) the phase is switched off without time delay.
Such conditions may occur during unsuccessful TPAR.
Figure 2 shows the transmission line currents in the case of
two-phase-to-ground fault with unsuccessful TPAR. SR is in
operation during the transient. In the course of reclosing the
current of the unfaulted phase C does not cross zero for more
than 0.2s, which may result in damages of the circuit breaker.
This problem does not exist, for example, for air circuit
breakers. An analysis carried out by independent experts from
ABB Group [2] and from St. Petersburg [8], shows that the
reasons why tripping of SF6 CB may lead to accident can be
summarized as follows:
1) SF6 CB creates an emergency situation due to the
nature of the switch-on process. The probability of
contact closure for air CB is close to 1 near the voltage
maximum, i.e. the closing results from the air gap
breakdown . In this case the DC component of the
current is minimal. Unlike air CB, the contacts of SF6
breaker can be closed near zero voltage. This leads to a
significant DC component. This behavior is due to high
insulation properties of sulfur hexafluoride.
2) SF6 CB creates an emergency situation due to the
nature of the switch-off process. SF6 CB has closed arc
chutes wherein the intensity of arc extinction depends on
the energy of the arc defined by the breaking current.
SF6 switch cannot turn off low currents until they

cross zero. This feature is evaluated as positive when

disconnecting small inductive currents, because this will
cause the minimum overvoltages (due to the absence of
current chopping). Sometimes this effect is enhanced by
means of synchronizers [9]. However, with respect to
the considered aperiodic current tripping problem, this
positive characteristic is negative, since it leads to a
prolonged arcing. Long arcing leads to the burning of
the contacts, to the overheating of the sulfur hexafluoride
and to the extreme increase of the pressure inside arc
chute. In the absence of a valve that relieves pressure,
there is an explosive destruction of the breaker poles
with damage of the surrounding equipment. Air circuit
breaker has a surplus of air under constant pressure; it
extinguishes the arc with intensity that does not depend
on the current strength, without waiting when the current
goes through zero.
To solve this problem the following methods can be used:
1) Application of a synchronization device that controls
the moment when each phase is reclosed close to the
voltage maximum [10], [11]. However, shunt reactorcompensated transmission lines present an additional
complexity on the voltage waveshape across the open
CB [12], [13], and when 0.7 < K < 1.3, the standard controlled switching devices may not always be
efficient [14].
2) Installation of closing resistors that will accelerate the
decay of aperiodic components. This technical solution
is associated with high financial cost [15], [16].
3) Tripping of the reactor (reactors) during TPAR to
detune the resonance. This solution is effective from a
technical viewpoint and has a low price.
Thus, the shutdown of the reactor is the most appropriate
control action to deal with the problem. Figure 3 again shows

Fig. 2. Line currents. Two-phase-to-ground fault with unsuccessful

TPAR. Without shunt reactor tripping.

the transmission line currents in the case of two-phase-toground fault with unsuccessful TPAR, but this time the reactor
CB is switched off in the dead time.
B. Resonance overvoltage during SPAR
The tuning of the line close to the resonance causes overvoltage that occurs after the extinction of the arc in the dead
time. The overvoltage leads to the restriking of the arc, which,
in turn, prevents the successful implementation of SPAR. A
detailed analysis of this phenomenon is given in [17]. Thus,
for the compensated lines the detuning of the resonance is the
only countermeasure against re-ignition of the arc. To this end
one or all three phases of the reactor should be switched off
in the dead time. Figure 4 shows the line voltages in SPAR
with and without tripping the reactor.
The analysis suggests that in practice there can be situations
that will need the reactor tripping during TPAR and SPAR.

Fig. 4. Line voltages during SPAR with and without tripping of the
reactor.

III. T HE PROBLEM
Let us turn analyse the process inside the SR CB. Consider
the following situation (Fig. 5):
1) At some time point there is a single-phase short circuit at
the beginning of the line close to the reactor (Fig.5, A).
2) Linear breakers disconnect the faulted phase from both
sides (SPAR procedure). The arc continues to burn, one
phase of the reactor is discharging to the point of fault
(Fig.5, B).
3) Directly after the disconnection of the faulted phase, the
phase of the reactor must also be switched off (Fig.5, C).
As stated above, this should be done in order to detune
the resonance and thus ensure the extinction of the arc
(Fig.4). In the case of TPAR this should be done to
prevent the damage of the line breaker during reclosing
(Fig.2).

Fig. 5. Explanatory diagram. Tripping of the SR during SPAR.

Fig. 3. Line currents. Two-phase-to-ground fault with unsuccessful

TPAR. SR tripping in the dead time.

The oscillograms of phase currents and voltages of the

reactor in the considered situation are shown in Figure 6.
The closer the short circuit to the reactor, the more aperiodic
the reactor current characteristic, the larger the time constant
value, and the worse the conditions for current tripping.
The main question of the paper can be summarized as
follows. In the considered situation, is it possible to switch off
the current (with high aperiodic and low periodic components)
which flows through the reactor? In other words, is it possible
to perform this operation without the reactor SF6 CB being
damaged?

A. Full-scale experiments
To determine the range of acceptable impacts on a shunt
reactor SF6 CB, the problem can be reformulated as follows.
What is the minimum amount of energy that can be dissipated
by one pole of the CB without the breaker being damaged?
The maximum energy stored in a single phase of a SR can be
calculated using the formula:
2
Umax
(2)
3 2 L
where Umax - maximum operating phase-to-phase voltage
(RMS); = 2f0 ; L - phase inductance of the reactor.
If the experiment confirms the existence of the problem, it
is possible to introduce a new parameter (in addition to the
conventional technical characteristics of the HVAC CB) that
would characterize the maximum amount of energy which can
be dissipated by one phase of the CB without the breaker
being damaged. When selecting a CB, this parameter should
be compared with the value of (2). For instance, for the
considered test system (Fig.1), Wmax 0.2 M J, and only the
experiment can answer the question of whether this amount
of energy can damage the SR CB.

Wmax =

Fig. 6. Phase currents and voltages of the SR.

This question was asked to a number of equipment manufacturers. All of them replied that the final answer will need
a series of expensive full-scale experiments. Finally we were
asked to consider one of the following solutions:
1) Closing resistors. This option has the following disadvantages. First, currently closing resistors are made only
for linear, but not for the SR CB. Second, as a rule, the
closing resistor is put into operation only when CB is
switched on, but in our case, the resistor must be put
into operation when CB is switched off. In addition, as
noted above, the price of the resistor is quite significant,
but the need for its use in our case is not fully justified.
Also, it should be noted that it is unacceptable to install
both a closing resistor and a synchronization device in
CB. However, the latter would be useful to improve the
transients during normal operation or emergency control
switching.
2) Introduction of a delay before the tripping shunt
reactor CB. This measure will reduce the aperiodic
current, and in the delay time most of the energy, stored
in the SR, will dissipate. This solution has the following
disadvantage. If the delay is too long, it may be difficult
to guarantee that SPAR procedure will be successful,
due to the resonance overvoltage to occur (Fig.4). Thus,
the delay should be combined with a means accelerating
the dissipation of the energy stored in SR.

B. Temporary introduction of resistors into the neutral of SR

Initially, the introduction of resistors into the circuit of
SR was proposed in [7] to solve the problem of aperiodic
current tripping (Fig.2). In the normal state the resistors are
bypassed by low voltage CB installed in the neutral (Fig.7).
If necessary, the resistors are temporarily put into operation
thereby improving the quality of transients. Let us show
that the same approach can be used to improve SR tripping
conditions during TPAR and SPAR.
The resistance of the power resistor unit is determined by
the insulation level of the SR neutral:
U
R
p max
UN
2
3 (R + X 2 )

(3)

where Umax - maximum operating phase-to-phase voltage of

the SR (RMS); R - resistance of power resistor unit; X - SR
phase reactance; UN - insulation level of the neutral.

None of the proposed solutions can answer all the questions.

The problem remains unsolved.
IV. WAYS OF SOLVING THE PROBLEM
The following solutions to the problem can be proposed:
1) Full-scale experiments for the identification of a range
of acceptable impacts on the SR CB.
2) Temporary introduction of resistors into the SR neutral.

Fig. 7. Explanatory diagram. Connection of power resistor units to

the circuit of shunt reactor.

For the equipment of 500 kV the typical insulation level

of the neutral is 35 kV. Thus, for the considered test system
(Fig.1), Rmax 200 . Taking into account the possibility of
resonance overvoltage (if arc extinguished before the reactor
is tripped off) we can finally take R = 150 . Figure 8 shows
the current of the SR faulted phase during SPAR with and
without temporary introduction of a 150 power resistor.
Maximum energy consumed by the power resistor is determined by the SR maximum current flow:
WRN =

2
Umax
Rt
3 (R2 + X 2 )

(4)

where Umax - maximum operating phase-to-phase voltage of

the SR (RMS); R - resistance of power resistor unit; X - SR
phase reactance; t - operation time of the resistor.
Advantages. The proposed solution is universal. It allows
solving not only the possible problem of SR tripping during
SPAR and TPAR, but also the problem of aperiodic current
tripping. At the same time, its implementation cost is lower in
comparison with the cost of preinsertion resistors. The reason
for this is that the closing resistors are installed on the high
voltage side of the SR (built into the CB), and their price
is determined by the need to ensure the thermal stability in
case of a close fault on the high voltage side. Maximum
energy consumed by the closing resistor is determined by the
following equation:
WR =

2
t
Umax
R

(5)

where Umax - maximum operating phase-to-phase voltage of

the SR (RMS); R - resistance of closing resistor; t - operation
time of the closing resistor.
The price of power resistors and CB installed in the neutral
is in turn determined by the thermal stability in the case of SR
load current flow (4). The load current is much smaller than
the short circuit current.
Compare maximum energies consumed by the closing resistor and by the resistor installed in the neutral. If Umax =
525 kV , R = 150 , X = 1570 , t = 0.02 sec, than
WRN 0.11 M J and WR 12 M J, which is about
a hundred times more. Given such a big difference, it can
be assumed that the difference in the prices of the resistors
(including installation and accessories) may be less than the

Fig. 8. SR faulted phase current with and without the introduction of

a 150 power resistor.

price of the necessary 35kV circuit breaker installed in the

neutral (Fig.7).
Disadvantages. Due to the introduction of resistors, tripping
of the reactor should be delayed. For the considered test system
the delay should be about 0.05-0.1 sec. Note that the minimum
duration of the dead time during SPAR is about 0.8-0.9 sec.
It is likely that even if the delay is taken into account, the
SR will be switched off before the arc is extinguished and the
overvoltage occurs. The minimum duration of the dead time
during TPAR is about 0.3-0.4 sec. and there is no overvoltage
after the arc extinction. Hence, it is guaranteed that the reactor
will switch off before reclosing. Thus, despite the fact that the
disconnection process is slightly extended in time, the delay
will not affect the reliability of the relay protection operation.
However, it should be noted that the delay alone without the
introduction of resistors cannot be considered as an adequate
measure.
V. CONCLUDING REMARKS
The research has demonstrated that on the one hand switching of shunt reactor during single and three-phase autoreclosing is an effective measure for the resonance detuning,
but on the other hand, this measure may be associated with
some difficulties. As the computer simulations prove, temporary introduction of power resistors into the neutral of SR
will resolve not only the problem of shunt reactor tripping
during SPAR and TPAR, but also the aperiodic current tripping
problem. Only full-scale experiments but not mathematical
models can give the final answer to the questions posed in
the paper.
R EFERENCES
[1] Ching-Yin Lee, Chang-Jhih Chen, Chao-Rong Chen, Yen-Feng Hsu,
Comparison of transient phenomena when switching shunt reactors on
the lines two terminals and station busbar, Power System Technology,
2004. PowerCon 2004. 2004 International Conference on, vol.2, no.,
pp.1255-1260, Vol.2, 2124 Nov. 2004.
[2] Report ABB Power Technologies, Investigation of Breaker Failure
when Disconnecting a Highly Compensated Line, no. 1JNR100009-697
Rev. 01.
[3] I. Naumkin, M. Balabin, N. Lavrushenko, R. Naumkin, Simulation of
the 500 kV SF6 circuit breaker cutoff process during the unsuccessful
three-phase autoreclosing, Proceedings of the IPST2011, Delft, The
Netherlands, June 14-17, 2011.
[4] L. Prikler, G. B`an, G. B`anfai, EMTP models for simulation of shunt
reactor switching transients, International Journal of Electrical Power &
Energy Systems, Volume 19, Issue 4, May 1997, Pages 235-240.
[5] V. E. Kachesov , D. V. Kachesov, Requirements for Switching Algorithms of EHV Shunt Compensated OHL by SF6 Circuit Breakers,
Proceedings of the IPST2011, Delft, The Netherlands, June 14-17, 2011.
[6] I. Naumkin. Emergency failures of SF6 circuit breakers during the
commutations of the compensated 500-1150kV transmission lines, (in
Russian). Elektrichestvo, no.10, 2012, p.22-32.
[7] I. Naumkin, V. Podyachev, L. Sarin, D. Kochura, Methods of performance assurance for SF6 circuit-breakers at switchings of compensated
500-1150 kV overhead power lines, Proceedings of the IPST2013,
Vancouver, Canada, July 18-20, 2013.
[8] G. Evdokunin, M. Dmitriev, S. Goldstein, U. Ivanitskii, High-voltage
overhead lines. Switching and their impact on circuit breakers, (in
Russian). Novosti Elektrotehniki, no.3, 2008.
[9] R. J. Rajotte, C. Charpentier, S. Breault, H. H. Le, Hieu Huynh,
J. Desmarais, Field tests of a circuit breaker synchronous control, Power
Delivery, IEEE Transactions on, vol.10, no.3, pp.1301-1309, Jul 1995.

[10] CIGREWorking Group 13.07, Controlled switching of HVAC circuit

breakers planning, specification and testing of controlled switching
systems, ELECTRA, no.197 Aug.2001, pp.23-33.
[11] R. Alvinsson, Controlled Switching of Circuit Breakers: A Technology
Status Review, ABB Switchgear AB, BC 96-027, Sweden.
[12] K. Froehlich, A. C. Carvalho, B. L. Avent, C. Hoelzl, W. Hofbauer,
D. F. Peelo, M. Stanek, P. Hoegg, and J. H. Sawada, Controlled closing
on shunt reactor compensated transmission lines Part I: Closing control
device development. Part II: Application of closing control device for
high-speed autoreclosing on BC hydro 500 kV transmission line, IEEE
Trans. Power Del., vol.12, no.2, pp. 734-746, Apr. 1997.
[13] P. M. Mestas, M. C. Tavares, A. M. Gole, Implementation and Performance Evaluation of a Reclosing Method for Shunt Reactor-Compensated
Transmission Lines, Power Delivery, IEEE Transactions on , vol.26, no.2,
pp.954-962, April 2011.
[14] M. Dmitriev, 110-750kV SF6 circuit-breakers. Methodology for the
selection of measures against aperiodic currents, (in Russian). Novosti
Elektrotehniki, no.4, 2013.
[15] A. C. Legate, J. H. Brunke, J. J. Ray, and E. J. Yasuda, Elimination
of closing resistors on EHV circuit breakers, IEEE Trans. Power Del.,
vol.3, no.1, pp.223-231, Jan.1988.
[16] A. C. Carvalho, M. Lacorte, and O. Knudsen, Improved EHV line
switching surge control by application of MO-Arrester and controlled
switching, in Proc. Int. Conf. Energy Management and Power Delivery,
Nov.1995, pp.292-297.
[17] M. Dmitriev, G. Evdokunin, EMTP Simulation of the Secondary
Arc Extinction at Overhead Transmission Lines under Single Phase
Automatic Reclosing, Proceedings of the IEEE St.Petersburg PowerTech,
St.Petersburg, Russia, June 27-30, 2005.