International Conference on Power Systems Transients – IPST 2003 in New Orleans, USA

Transient Recovery Voltages During the Switching Under

Out-of-Phase Conditions

Dieter Braun1 and Georg Koeppl2

(1) ABB Switzerland Ltd., Dept. PTHG-V, Brown Boveri Strasse 5, CH-8050 Zuerich, Switzerland (e-
mail: [email protected]), (2) KOEPPL POWER EXPERTS, Landstrasse 113, CH-5430 Wettin-
gen, Switzerland (e-mail: [email protected])

Abstract – Out-of-phase synchronising occasionally occurs in tions are defined as 50% of the rated short-circuit breaking
power stations, one of the main reasons being wiring errors current of the circuit-breaker and in [5] as 25%. The TRVs
leading to particular out-of-phase angles of 60°el., 120°el. or in both standards are based on an out-of-phase angle of
180°el.. The relevant standards IEEE Std C37.013 for genera- 90°el. only, which seems to be arbitrary.
tor circuit-breakers and IEC 62271-100 for high-voltage cir-
On the other hand the conditions regarding switching of
cuit-breakers consider only a maximum out-of-phase angle of
90°el.. From the well-defined TRV (transient recovery volt- short-circuit currents are well-founded and very precisely
age) parameters for ?system-source faults? and ?generator- defined.
source faults? also the TRV parameters for out-of-phase con- In [4] two kind of terminal faults are distinguished:
ditions can be deduced. This is done for three examples and a) system-source fault (fault between the generator cir-
the calculated TRV parameters are compared to those given cuit-breaker and the generator)
in IEEE Std C37.013. For a 90°el. out-of-phase angle the rates b) generator-source fault (fault between the generator
of rise of TRV fit rather well; the calculated peak values are circuit-breaker and the step-up transformer)
however smaller and the calculated out-of-phase switching
and corresponding TRV parameters have been assigned to
currents are larger than given in IEEE Std C37.013. This
standard does not at all cover the conditions at an out-of-
these cases.
phase angle of 180°el.. The situation seems to be still worse It seems to be logical that the two short-circuit cases
with IEC 62271-100 for high-voltage circuit-breakers. should be consistent with the out-of-phase switching case
since according to Figure 1 exactly the same components
Keywords – power station, out-of-phase conditions, generator are involved, namely the generator, the generator circuit-
circuit-breaker, high-voltage circuit-breaker, transient recovery breaker, the step-up transformer and the high-voltage sys-
voltage tem, and the same frequencies apply. This means that from
the TRVs at short-circuit cases a) and b) the TRV for out-
I. INTRODUCTION of-phase conditions can be deduced.
It is the purpose of this paper to check by means of a few
Out-of-phase synchronising occasionally occurs in typical examples whether this consistency exists.
power stations [1, 2, 3]. The main reasons for out-of-phase
synchronising are wiring errors made during commission-
ing or during maintenance when connecting voltage trans-
formers and synchronising equipment. These wiring errors
lead to particular out-of-phase angles, i.e. multiples of
60°el.. E.g. polarity errors at a voltage transformer cause
synchronising at 180°el. out-of-phase angle; phase connec-
tion errors lead to 60°el. and 120°el. out-of-phase angles.
Besides these particular out-of phase angles any value
may be caused by inadequate settings of the synchronising
equipment, e.g. due to an incorrect value of the closing
time of the circuit-breaker.
The above mentioned aspects apply especially to gen-
erator circuit-breakers (located between the generator and Fig. 1: Synchronising at MV side of step-up transformer
the step-up transformer) but also to circuit-breakers at the
high-voltage terminals of step-up transformers. All these II. TRVS FOR SYSTEM-SOURCE AND GENERATOR-
circuit-breakers have to have the capability to interrupt out- SOURCE FAULTS
of-phase currents. The corresponding breaking currents and
TRV (transient recovery voltage) ratings are laid down in In power stations the system-source fault (fault between
IEEE Std C37.013 [4] for generator circuit-breakers and in the generator circuit-breaker and the generator) normally
IEC Publication 62271-100 [5] for high-voltage circuit- yields the highest short-circuit current due to the small
breakers. leakage impedance of the step-up transformer and the gen-
In [4] the breaking currents under out-of-phase condi- erally considerably smaller short-circuit impedance of the
high-voltage system (referred to medium-voltage side of

1
 International Conference on Power Systems Transients – IPST 2003 in New Orleans, USA

the step-up transformer). This is more pronounced at ductances, paralleled by resistance elements and 10 ca-
smaller MVA ratings of the step-up transformer. pacitance-to-ground elements (all per phase). For calculat-
The TRV parameters for this case (Table 5 of [4]) are ing the TRV shapes the Alternative Transients Program [6]
very severe regarding the steepness (due to high natural has been used. The inductance elements determine the
frequency of transformers) and the steepness increases with short-circuit current. The capacitance and resistance ele-
transformer rating. ments have been adapted to result into the required TRV
Compared to this the generator-source fault (fault be- shape for the first pole-to-clear in case of an interruption of
tween the generator circuit-breaker and the step-up trans- a three-phase ground fault.
former) shows a smaller current and a less critical steep-
ness of the TRV which rises too with the MVA rating of 50
*10 3
the generator (Table 6 of [4]). The smaller steepness of the
generator-side TRV is mainly caused by the large winding 40
capacitance of generators.
As a first step 3 typical and realistic transformer and 30
generator data sets with system-source short-circuit cur-
rents of 80, 100 and 130kA at a rated voltage of 25.3kV
20
have been chosen (Table I). The corresponding generator-
source short-circuit currents are 64, 80 and 104kA.
10

Table I: Data for TRV analysis

0
0.00 0.03 0.06 0.09 0.12 [ms] 0.15
Case 1 2 3
Vrated [kV] 25.3 25.3 25.3 a) TRV for case 1 (envelope acc. to Table 6 of [4])
SGenerator [MVA] 480 600 780 50
Isc, generator.-source [kA] 64 80 104 *10 3

xd? [p.u.] 0.171 0.171 0.171

40
TRV rate [kV/µs] 1) 2.0 2.0 2.0
1
E2-crest voltage [kV] ) 46.6 46.6 46.6
30
T2 [µs] 1) 27.3 27.3 27.3
STransformer [MVA] 480 600 780
20
Isc, system-source [kA] 80 100 130
usc [p.u.] 0.1369 0.1369 0.1369
TRV rate [kV/µs] 2) 5.0 5.0 5.5 10

E2-crest voltage [kV] 2) 46.6 46.6 46.6

T2 [µs] 2) 10.9 10.9 9.9 0
1 0.00 0.03 0.06 0.09 0.12 [ms] 0.15
) according to Table 6 of [4]
2
) according to Table 5 of [4] b) TRV for case 2 (envelope acc. to Table 6 of [4])
50
From Tables 5 and 6 of [4] the corresponding TRV rates *10 3
of rise, E2-crest voltages and T2 are calculated and shown
40
in Table I, too. It is to be noted that these TRV ratings cor-
respond to the first pole to clear a three-phase grounded
fault. The components connected to the other side of the 30

three-phase grounded fault do not play any role for the

TRV, in contrary to a three-phase ungrounded fault which 20
would show a higher steepness of the TRV.
10

0
0.00 0.03 0.06 0.09 0.12 [ms] 0.15

c) TRV for case 3 (envelope acc. to Table 6 of [4])

Fig. 2: Model for generator Fig . 3: TRV shapes for generator-source fault

A three-phase generator model (Figure 2) providing the Figures 3a), b), and c) show the simulated TRV shapes
correct generator-source short-circuit current has been es- for the generator-source fault. The generator model used
tablished consisting of 10 positive and zero sequence in- does not claim to represent the generator correctly in the

2
 International Conference on Power Systems Transients – IPST 2003 in New Orleans, USA

total frequency range but it shows a perfect behaviour re- Similarly as for the generator model it must be empha-
garding the TRV during three-phase fault conditions. The sised that this transformer/system model is valid only for a
model has been deduced from measured TRVs of genera- restricted frequency range, namely up to the frequencies
tors. It should be noted that even the bend during the first associated with TRVs at short-circuit interruptions.
rise of the TRV typical for three-phase grounded faults is
correctly reproduced by this model (Figure 4); the TRV of
ungrounded faults would show a straight line up to the
peak.

Fig. 5: Model for step-up transformer

III. CURRENTS AND TRVS FOR OUT-OF-PHASE

CONDITIONS AT MV SIDE OF STEP-UP
TRANSFORMER
Fig. 4: Comparison between measured and calculated TRV
It is obvious that for out-of-phase current switching the
The corresponding model for the step-up transformer is same models must be valid since the only condition
shown in Figure. 5. It consists again of 10 modules per changed is the angle of the voltage and the absence of a
phase, in this case 10 series connected single-phase trans- short-circuit. For a 180°el. out-of-phase angle the TRV
formers paralleled by resistors at the medium-voltage side, plots for the 3 cases are shown in Figures 7a), b), and c),
capacitances to ground both at the high- and medium- respectively, and the corresponding evaluation is given in
voltage side and coupling capacitances between the high- Table II.
voltage and the medium-voltage side. The impedance of
the high-voltage system has been considered to be included Table II: Out-of-phase switching - comparison of TRV
in this scheme. Actually the high-voltage system feeding
the step-up transformer will have a considerably smaller Case 1 2 3
natural frequency than the transformer leading to an TRV data according to [4], Table 9
equivalent circuit with two natural frequencies. This has TRV rate [kV/µs] 4.7 4.7 4.7
been taken into account by a modest amplitude factor of E2-crest voltage [kV] 65.8 65.8 65.8
1.5 for the single frequency circuit specified in [4], Table Out-of-phase current [%] 50 50 50
5. Hence it seems to be justified to include the impedance
TRV data calculated for 180°el. out-of-phase angle
of the high-voltage system in the transformer model. The
high-voltage side neutral of the step-up transformer of TRV rate [kV/µs] 6.7 6.5 7.0
vector group Yd is grounded. E2-crest voltage [kV] 85.7 85.8 84.1
The parallel elements of Figure 5 have been adapted to Out-of-phase current [%] 89 89 89
provide the correct TRV shapes like for the generator- TRV data calculated for 90°el. out-of-phase angle
source fault. TRV rate [kV/µs] 4.5 4.4 4.8
Figures 6a), b), and c) show the simulated TRV shapes E2-crest voltage [kV] 60.6 60.7 59.4
for the system-source fault. Out-of-phase current [%] 63 63 63

3
 International Conference on Power Systems Transients – IPST 2003 in New Orleans, USA

As to be expected far more severe TRV parameters re- phase angle are compared to the TRV parameters given in
sult than those given in Table 9 of [4]. It is interesting to [4]. The steepnesses fit quite well, the voltage peaks of the
note that the out-of-phase current in this case is 89% of the simulations are smaller however. This is caused by the dif-
system-source short-circuit current, i.e. it is quite close to ference in the natural frequency of the generator and the
the rated short-circuit current of the generator circuit- step-up transformer and should be taken into account in a
breaker. future revision of [4].

50 90
*10 3 *10 3

75
40

60
30

45

20
30

10
15

0 0
0.00 0.02 0.04 0.06 0.08 [ms] 0.10 0.00 0.03 0.06 0.09 0.12 [ms] 0.15

a) TRV for case 1 (envelope acc. to Table 5 of [4]) a) TRV for case 1 (envelope acc. to Table 9 of [4])
50 90
*10 3
*10 3

40 75

60
30

45
20

30

10
15

0
0
0.00 0.02 0.04 0.06 0.08 [ms] 0.10 0.00 0.03 0.06 0.09 0.12 [ms] 0.15

b) TRV for case 2 (envelope acc. to Table 5 of [4]) b) TRV for case 2 (envelope acc. to Table 9 of [4])
50 90
*10 3
*10 3

40 75

60
30

45

20

30

10
15

0 0
0.00 0.02 0.04 0.06 0.08 [ms] 0.10 0.00 0.03 0.06 0.09 0.12 [ms] 0.15

c) TRV for case 3 (envelope acc. to Table 5 of [4]) c) TRV for case 3 (envelope acc. to Table 9 of [4])

Fig. 6: TRV shapes for system-source fault Fig. 7: TRV shapes for 180°el. out-of-phase condition

For the case given in [4], i.e. an out-of-phase angle of IV. TRVS FOR OUT-OF-PHASE CONDITIONS AT HV
90°el., the out-of-phase current is 1/√2-times smaller. The SIDE OF STEP-UP TRANSFORMER
TRV shapes are proportional to those of Figures 7a), b),
and c) again with a proportionality factor of 1/√2. In Table Analogous to the case of the generator circuit-breaker
II also the results of the simulation for a 90°el. out-of- the TRV requirements for the circuit-breaker at the high-

4
 International Conference on Power Systems Transients – IPST 2003 in New Orleans, USA

voltage terminals of the step-up transformer may be de- V. CONCLUSIONS

duced. The same synchronising errors can occur at this
location, hence a 180°el. out-of-phase angle is possible as The TRV requirements of IEEE Std C37.013 [4] re-
well. The corresponding scheme is shown in Figure 8. garding out-of-phase switching have been analysed based
The highest short-circuit current through the circuit- on three examples. The following conclusions may be
breaker is normally delivered by the system (fault between drawn from the results obtained:
the high-voltage circuit-breaker and the step-up trans- § The out-of-phase TRV parameters and switching
former) which corresponds to test duty TD100 and TRV- currents specified in [4] are restricted to an out-of-
requirements according to Tables 1b, 1c and 1d of [5]. For phase angle of 90°el. From the point-of-view of
rated voltages of 100kV and above the rate of rise of the possible wiring errors causing synchronising faults
TRV is given as 2kV/µs and the TRV shape as a four- this angle seems to be arbitrary and inadequate. The
parameter curve with a peak value of 1.4·1.3·v2Un/v3. most severe case is an out-of-phase angle of 180°el..
A generator-source fault (fault between the high-voltage § Since the TRV parameters for system-source faults
circuit-breaker and the high-voltage system) comprises in and generator-source faults are well defined in [4],
this case the generator impedance and the transformer im- the corresponding TRV parameters for out-of-phase
pedance leading to a substantially smaller short-circuit cur- switching follow necessarily from that data.
rent than for the system-source fault. This case is covered § For the three examples appropriate models have
by the test-duties TD10 and TD30 of [5], i.e. short-circuit been chosen to represent the TRV behaviour of the
currents of 10% and 30% of the rated short-circuit breaking system source faults and the generator-source faults.
current. TRV calculations for these models at 90°el. out-of-
The corresponding TRV parameters are laid down in phase conditions show a good fit with [4] regarding
Tables 14a and 14b of [5]. Especially high rate of rise val- steepness; the peak value is only about 90% of [4]
ues and high amplitude factors are assigned to test duty however. At 180°el. out-of-phase angle the steep-
TD10; they are meant to represent the high natural fre- nesses and peak values are v2-times higher than at
quencies and the low damping of transformers. These high 90°el..
rate of rise values and even higher amplitude factors would § The out-of-phase switching current specified in [4]
also prevail in case of out-of-phase switching, since the amounts to 50% of the short-circuit breaking cur-
voltage drop mainly appears across the generator and the rent. The examples show higher values of 63% at
step-up transformer. 90°el. and 89% at 180°el..
The out-of-phase breaking current according to [5] is § Based on these findings it seems to be advisable to
25% of the rated short-circuit breaking current which is revise the out-of-phase switching requirements
fairly in line with the considerations above. In contrast to given in [4].
this the TRV parameters for out-of-phase conditions given Similar considerations apply to IEC 62271-100 [5]. This
in [5] are extremely modest (1.54kV/µs in the voltage standard applies to high-voltage circuit-breakers, i.e. in the
range =100 kV) and do not seem to be adequate. case considered to the circuit-breaker at the high-voltage
Basically the same procedure as described in Section III terminals of the step-up transformer. The required out-of-
could be applied for the determination of the TRV pa- phase switching current of 25% of the short-circuit break-
rameters for out-of-phase switching with high-voltage cir- ing current fits with practical data. The TRV requirements
cuit-breakers. however seem to be far too low regarding steepness as well
Due to a multitude of cases and combinations to be con- as amplitude.
sidered such an analysis is beyond the scope of this paper.
REFERENCES
[1] B.M. Pasternack, J.H. Provenzana, L.B. Wagenaar, "Analysis of
a generator step-up transformer failure following faulty synchro-
nisation", IEEE Transactions on Power Delivery, Vol. 3 (3), pp.
1051-1058, 1988.
[2] C.M. Arturi, "Transient simulation and analysis of a three-phase
five-limb step-up transformer following an out-of-phase syn-
chronisation", IEEE Transactions on Power Delivery, Vol. 6 (1),
pp. 196 -207, 1991.
[3] J.V. Mitsche, P.A. Rusche, "Shaft torsional stresses due to asyn-
chronous faulty synchronization", IEEE Paper F 80 203-0, IEEE
PES Winter Meeting, New York, 1980.
[4] IEEE Std C37.013-1997: IEEE Standard for AC High-Voltage
Generator Circuit Breakers Rated on a Symmetrical Current.
[5] IEC 62271-100 2001-05 and Amendment 1 2002-05: High-
Fig. 8: Synchronising at HV side of step-up transformer Voltage Switchgear and Controlgear - Part 100: High-Voltage
Alternating-Current Circuit-Breakers.
[6] Canadian/American EMTP User Group: Alternative Transients
Program (ATP) Rule Book, 1987-92.