EUROCON 2007 The International Conference on “Computer as a Tool” Warsaw, September 9-12

Computer Simulation of Transients Due to

Disconnector Switching in Air-Insulated
Substations
Zijad Bajramovic1, Salih Carsimamovic1, Meludin Veledar2, Omer Hadzic3, Adnan Carsimamovic3,
Predrag Osmokrovic4
1 Faculty of Electrical Engineering, Sarajevo, Bosnia and Herzegovina
2
ABB Representation for B&H, Sarajevo, Bosnia and Herzegovina
3
Independent System Operator, Sarajevo, Bosnia and Herzegovina
4
Faculty of Electrical Engineering, Belgrade, Serbia

Abstract—Switching overvoltages due to disconnector voltage collapse, ∆U, Fig. 1., applied to the circuit gives
switching in air-insulated substations (AIS) are presented. rise to a voltage and current damped oscillating wave.
Analysis of shape and levels of switching overvoltages by
means of experimental and computing investigations on 220
kV switchyards of Thermo Power Plant (TPP) Kakanj are
presented. Computer simulations by means of EMTP-ATP
[1] are performed.

Keywords—Switching transients, overvoltages,

substations, electromagnetic compatibility, electromagnetic
interference.

I. INTRODUCTION
As generator stations and substations have become
more complex their required standards of reliability and
stability have become more demanding. These demands
have been met by the widespread use of high-speed and
low-power electronic systems. This progress in the
technology has reduced the level of energy required to
switch between logic states, increased the efficiency with
which these levels of energy can be transferred by Fig. 1. The voltages due to the disconnector switching
unintended coupling path, and lowered the levels of a) Simple pattern of substation, b) Wave shape on disconnector’s
contacts
energy which the components can survive.
Electromagnetic interference problems can manifest
themselves as errors or failures after the equipment has The maximum value of voltages and the maximum
entered service. Every incident of interference involves a value of the slope of the wave front increasing will take
source of disturbances, a coupling mechanisms and path, place at the maximum distance between contacts. For the
and susceptible piece of equipment [2]. Switching purpose of the investigation of the insulation strength and
operation of disconnector in power stations and induction of electromagnetic interferences (EMI), the
substations causes high level of high frequency most important are the first few strikes during the closing
overvoltages that can be coupled with low voltage operation or the last few strikes during the opening
secondary circuits and electronic equipment unless they operation.
are suitably protected. The purpose of this paper is to
suggest computer models of circuits for representative of The function of high-voltage air-break disconnectors is
simulations transients due to disconnector switching in to provide electrical isolation of one part of the
air-insulated substations. switchgear. Disconnector’s standards define a negligible
current interrupting capability (≤ 0.5 A) or a voltage
II. OUTLINE PROBLEM between the contacts if it is not significantly changed.
These values of currents include the capacitive charging
Switching operation of disconnector in power stations currents of bushings, busbars, connectors, very short
and substations give rise to electromagnetic interferences lengths of cables and the current of voltage instrument
because they generate abrupt voltage collapses, ∆U, across transformers. Disconnector’s contacts in air-insulated
the contacts of the switching equipment [3]. The collapse substations (AIS) are moving slowly causing numerous
time, ∆t, is dependent on the distance between the contacts strikes and restrikes between contacts. Strikes and
(several tens or some hundreds of nanoseconds) of restrikes occur as soon as the dielectric strength of the air
equipment in air-insulated substations (AIS) [2]. The between contacts is exceeded by overvoltage. The

1-4244-0813-X/07/$20.00 2007 IEEE. 2051

distance between contacts, the contacts geometry and disposition of the substation, rated voltage and frequency.
relative atmospheric condition defines the overvoltage at Parameters of busbars, overhead lines, and cables are
the instant of strike. Every strike causes high-frequency determined by their geometry characteristics, as lengths,
currents tending to equalize potentials at the contacts. cross sections, type of materials, heights, mutual distances
When the current is interrupted, the voltages at the source and specific resistivity of soil.
side and the loading side will oscillate independently. The Computer simulations of the AIS test circuits at TPP
source side will follow the power frequency while the Kakanj were performed using EMTP-ATP [1] on the
loading side will remain at the trapped voltage. As soon as model of test circuits (Fig. 2. and 5.) containing elements
the voltage between contacts exceeds the dielectric drawn in Fig. 3. and 6., respectively. Overvoltages at
strength of the air, at that distance the restrike will occur, busbars were calculated during disconnector closing
and so on. When closing takes place, the first strike will operations, for the same substation layout on which
occur at the maximum value of the source voltage. Its measurements were carried out.
values can be positive or negative. As the time passes a
series of successive strikes will keep occurring at reduced
amplitude, until the contacts touch. The highest transient
overvoltage therefore occurs during the initial pre-arc.
During the disconnector opening, restrikes occur because
of the very small initial clearance between contacts. At the
beginning of transient the intervals between particular
strikes are in the order of a millisecond, while just before
the last strike the period can reach about half of cycle at
power frequency.
Traveling wave current and voltage transients on each
phase of high voltage bus act as a complex radiating
antenna. These traveling waves are the most important
sources of electromagnetic coupling on secondary circuits.
Bus current transients can also couple into low voltage
circuits that are connected directly to the bus. Common Fig. 2. Test circuit at TPP Kakanj, the first case
impedance coupling (‘parasitic’) occurs inside a piece of CT-current transformer, CL-capacitive load, CVD-capacitive voltage
equipment connected to the bus (for example the internal divider, CB-circuit breaker, DC-disconnector (220 kV, 1250 A)
capacities of the instrument transformers) [4]. Common
mode, CM, voltages, i.e. voltages measured between
conductors and local ground, represent the main parameter network
220kV DISCONNECTOR Current
used for assessing equipment immunity. busbars arc busbars transformer busbars busbars
During switching operation of disconnector (220 kV
U
switchyards of TPP Kakanj) there were registered up to L Lct Ccb
300 restrikes (pulses). Closing time of disconnector lasted C Cdc Cct Lv Rd
for about 35-40 cycles (i.e. 700-800 ms), but opening time
of disconnector lasted for about 40-45 cycles (i.e. 800-900 Lv C1
ms). The maximum values of transient overvoltages at
busbar during operations of disconnectors were recorded CL C2
up to 339 kVp, with the dominant frequency of considered
transient fd equal to 0.8 MHz [5, 6].
Capacitive load DIVIDER
III. MODELING OF THE TEST CIRCUIT
During calculating of overvoltages caused by switching Fig. 3. Model of test circuit, the first case
operation of disconnectors in AIS there is a need to Arc-2 Ω, disconnector’s stray Cdc= 200 pF, busbar Z= 340 Ω, CT
establish appropriate computer model to investigate this current transformer’s stray Cct= 500 pF, circuit breaker’s stray Ccb= 200
pF, CT primary winding inductance Lct= 4 µH, high voltage capacitor
phenomena. As a tool, the worldwide and mostly known C1= 1000 pF, intermediate voltage capacitor C2= 0.445 µF, primary
ATP-EMTP (Alternative Transients Program – resistance of voltage divider Rd= 300 Ω, connection wire inductance of
Electromagnetic Transients Program) has been used. For voltage divider/capacitive load LV= 10 µH, capacitive load CL= 1000 pF
purpose of analyzing switching part of substations by
disconnector, there are needs for appropriate model of
every element of substation propagating electromagnetic Results of measured and calculated trapped charge and
waves. Elements of substation that influence on distortion overvoltage factor for this substation are presented in Fig.
of electromagnetic waves are power transformers, voltage 4.
and current instrument transformers, circuit breakers,
insulators, busbars, connectors, surge arresters, suspension
and support insulators, bushings etc. Busbars, power lines
and cables have been simulated as lines with distributed
parameters by means of wave impedance, Z, and wave
velocity, v. Other elements of AIS have been represented
by their capacitances, C. In addition, to establish computer
model of transient phenomena caused by operation of
disconnectors it is necessary to have data relating to

2052
 network
220 kV DISCONNECTOR Current
L busbars busbars transformer
arc
U

C Ccvt Cdc Cct Lct

Lv Lv
Capacitive
voltage
transformer Rd Rd
busbars
C1 C1
Ccb
busbars
C2 C2

busbars Cdc

a) Trapped charge on busbars ( ___ measured, ___ calculated) DIVIDERS Cmosa

Fig. 6. Model of test circuit

Arc-2 Ω, disconnector’s stray Cdc= 200 pF, busbar Z= 340 Ω, CT
current transformer’s stray Cct= 500 pF, circuit breaker’s stray Ccb= 200
pF, metal-oxide surge arrester’s stray Cmosa= 200 pF, CT primary
winding inductance Lct= 4 µH, high voltage capacitor C1= 1000 pF,
intermediate voltage capacitor C2= 0.445 µF, primary resistance of
voltage divider Rd= 300 Ω, connection wire inductance of voltage
divider 10 µH, capacitive voltage transformer Ccvt= 2200 pF

The wave shape of measured and calculated surge

overvoltages at source side is given in Fig. 7.
b) Surge overvoltages p.u. ( ___ measured, ___ calculated)

Fig. 4. The wave shape of measured and calculated overvoltages on

load side – the first case

Comparison of measured and calculated surge

overvoltages and frequencies are given in Table I.

TABLE I.
COMPARISON OF MEASURED AND CALCULATED SURGE OVERVOLTAGES
LOAD SIDE
Difference, ∆
Parameter Measured Calculated
(%)
Overvoltage factor k
1.55 1.62 4.52
(p.u.)
Frequency f (kHz) 390 400 2.56

a) Overvoltage on source side ( ___ measured, ___ calculated)

Fig. 5. Model of test circuit, the second case

b) Surge overvoltages p.u. ( ___ measured, ___ calculated)
CT-current transformer, CVD-capacitive voltage divider, CB-circuit
breaker, DC-disconnector (220 kV, 1250 A)
Fig. 7. The wave shape of measured and calculated overvoltages on
source side – the second case

2053
 Comparison of measured and calculated surge Comparison of measured and calculated surge
overvoltages and frequencies are given in Table II. overvoltages and frequencies are given in Table III.

TABLE II. TABLE III.

COMPARISON OF MEASURED AND CALCULATED SURGE OVERVOLTAGES AT COMPARISON OF MEASURED AND CALCULATED SURGE OVERVOLTAGES AT
SOURCE SIDE LOAD SIDE

Difference, ∆
Parameter Measured Calculated Parameter Measured Calculated Difference, ∆ (%)
(%)

Overvoltage factor k Overvoltage factor

1.31 1.33 1.53 1.34 1.36 1.49
(p.u.) k (p.u.)
Frequency f (kHz) 412 420 1.94 Frequency f (kHz) 412 420 1.94

The wave shape of measured and calculated surge

overvoltages at load side is given in Fig. 8. IV. CONCLUSIONS
The differences between magnitudes of measured and
calculated overvoltages are up to 4.52 % and up to 2.56 %
on frequency.
Comparison of the transient computer simulations with
the field measurements showed that calculations could be
used for assessment of the transient overvoltages caused
by disconnector switching.

REFERENCES
[1] EMTP_ATP, European EMTP-ATP Users Group e.V.
[2] Guide on EMC in Power Plants and Substations, CIGRE WG
36.04, Dec. 1997
[3] R.Cortina, A.Porrino, P.C.T. Van Der Laan, A.P.J. Van Deursen,
‘Analysis of EMC problems on auxiliary equipment in electrical
a) Trapped charge ( ___ measured, ___ calculated) installations due to lightning and switching operations’, CIGRE
1992, paper 36-302
[4] H.Remde, H.Schwarz, ‘Transient overvoltages in CT and VT
secondary circuits in high-voltage substations’, ABB Review 1/91
[5] ‘Overvoltages in primary and secondary circuits in RP Kakanj
due to disconnector switching’, Faculty of Electrical Engineering,
Report No.L1410020, Sarajevo, 2005
[6] S.Carsimamovic, R.Gacanovic, Z.Bajramovic, M.Veledar,
M.Ljevak, P.Osmokrovic, ‘Switching Overvoltages in Air-
Insulated Substations (AIS) due to Disconnector and Circuit
Breaker Switching’, CIGRE 2006, paper C4-301

b) Surge overvoltages p.u. ( ___ measured, ___ calculated)

Fig. 8. The wave shape of measured and calculated overvoltages on load

side – the second case

2054