TRANSIENT RECOVERY VOLTAGE (TRV) FOR HIGH-VOLTAGE

CIRCUIT BREAKERS
R.W.Alexander, PPL, Senior Member IEEE

D.Dufournet, Alstom T&D, Senior Member IEEE

1 General
The recovery voltage is the voltage which appears across the terminals of a pole of a circuit breaker.
This voltage may be considered in two successive time intervals: one during which a transient voltage
exists, followed by a second one during which a power frequency voltage alone exists.

CURRENT

TRANSIENT RECOVERY RECOVERY

VOLTAGE VOLTAGE

Figure 1 Current, TRV and Recovery Voltage

During the interruption process the arc rapidly loses conductivity as the instantaneous current
approaches zero. Within a few microseconds after current zero, current stops flowing in the circuit.
The power system response to the current interruptions is what generates TRV. The difference in the
power system response voltage from the source side to the load side of the circuit breaker is the TRV.
The breaking operation is successful if the circuit breaker is able to withstand the TRV and the power
frequency recovery voltage. TRV is then related to the power system response to an interruption of
current in a circuit very close to a power frequency current zero. The nature of the TRV is dependent
on the circuit being interrupted, wether primarily resistive, capacitive or inductive, (or some
combination). Additionally, distributed and lumped circuit elements will produce different TRV
waveshapes.
In principle, the response of the load side and source side of the circuit breaker can be analyzed
separately, and the results subtracted point by point on a time line. The driving voltage will be the
instantaneous power frequency voltage across the circuit elements at the instant of current
interruption. If there is no damping, the highest peak circuit response would be 2 times the driving
voltage. The proportion of the system voltage across each piece of the switched circuit will be
determined by the impedance of each piece at the power frequency.

1
When interrupting a fault at the circuit breaker terminal in an inductive circuit, the supply voltage at
current zero is maximum. The circuit breaker interrupts at current zero, at a time when the power input
is minimum, and the voltage on the supply side terminal reaches the supply voltage in a transient
process called the transient recovery voltage. This is illustrated in Figure 2.

CURRENT

Supply voltage

TRANSIENT RECOVERY
VOLTAGE

Figure 2 Current and TRV waveforms during interruption of inductive current

1
The TRV frequency is , with L = short-circuit inductance, C = supply capacitance.
2π LC
When a pure resistive circuit is interrupted, the supply voltage is zero at the time of interruption,
therefore the recovery voltage has no transient component (see Figure 3).

CURRENT

RECOVERY VOLTAGE

Figure 3 Current and TRV waveforms during interruption of resistive current

A capacitively dominated circuit will have crest voltage across the capacitive elements at the instant of
current interruption, this will give a dc offset on the TRV. In the simplest case, the TRV will be a 1-
cosine of the power frequency, oscilating between 0 and 2 p.u.
In an inductively dominated circuit, the supply voltage at the instant of current interruption will be close
to peak (see figure 2). A pure inductive circuit would then have a TRV which is a 1 p.u. step function,
followed by a cosine wave of power frequency. This would have an infinitely fast rate of rise and could
only be interrupted by an ideal circuit breaker. In real life there is always some shunt capacitance of
the inductor, which in the most onerous case, gives rise to an oscillatory TRV. The frequency of the

2
oscilation will be determined by the L and C involved. The TRV peak will at worst be 2.0 times the
driving voltage. In cases where there is sufficient damping, the inductive circuit will produce a 1- e-αt
wave shape.
Figure 4 illustrates three types of recovery voltages in resistive, inductive and capacitive circuit.

2,5

1,5
CAPACITIVE
CIRCUIT
1

0,5

-0,5

-1
RESISTIVE
CIRCUIT INDUCTIVE CIRCUIT
-1,5
INDUCTIVE CIRCUIT
with stray capacitance
-2

Figure 4 TRV and recovery voltage in resistive, inductive or capacitive circuits

Circuits for mainly active load current switching can be simulated by a combination of resistive and
inductive elements. The recovery voltage is a combination of the shown in figure 4 for resistive and
inductive circuits. It has initially a high frequency transient, due to the voltage drop in the transformer
reactance, followed by a power frequency voltage, the amplitude of which is function of the load power
factor.
In a similar way, interruption in a L-C series circuit produces initially a high frequency TRV of small
amplitude (the voltage prior to interruption tends toward the supply voltage value) called the voltage
jump, followed by the 1 - cos waveshape shown on figure 4 for a capacitive circuit.
An open circuited line will act much like a capacitor (representing the distributed capacitance). If the
line is very long, the Ferranti effect (voltage rise along an unloaded line due to the line’s capacitive
charging current being drawn through its own inductance) can produce a higher voltage at the far end.
After circuit interruption the voltage profile along the line will “travel” and a peak TRV higher than 2.0
p.u. can result across the circuit breaker interrupter.
A short-circuited line will have a “ramp” voltage profile along the line. Upon interruption this ramp will
“travel”. Depending on the electrical circuit discontinuities, there will be reflections. Still a 2.0 p.u.
voltage would be the maximum expected peak (remember that lines can comprise part (or all) of the
source or load impedance or both).
There is a TRV for any interruption not just for fault interruptions. Although fault interruptions are often
considered to produce the most onerous TRVs, there are many exceptions such as shunt reactor
switching (see section 7.3).
TRVs can be oscillatory, triangular, or exponential and can occur as a combination of these forms. A
dc offset may also be present as it is the case for lines with series capacitors.
As shown in annex A.1, a network can be reduced to the simple parallel RLC circuit of figure 5 for TRV
calculations. This representation is reasonably valid for short-time frames until voltage reflections
return from discontinuities in the network.
1
The TRV in the parallel RLC circuit is oscillatory (underdamped) if R 〉 L/C
2
1
The TRV in the parallel RLC circuit is exponential (overdamped) if R ≤ L/C
2

3
 Figure 5 – Equivalent RLC circuit

Figure 6 shows the TRV shape as function of damping in an RLC circuit. From this figure, it can be
seen that by lowering the resistance in the equivalent circuit, for example when adding a resistance of
low ohmic value in parallel to the interrupting chamber(s), it is possible to effectively reduce the rate-
of-rise of TRV (the TRV peak voltage is reduced as well). This method has been widely used for many
years to assist current interruption by air-blast circuit breakers.

TRV (p.u.)
2
0.5
R (C / L) = 10
1,8
4
1,6
2
1,4
1
1,2
0.75
1

0,8

0,6
0.5 critically damped
0,4
0.3
0,2

0
0 1 2 3 4 5 6 7 8 9

t / RC
Figure 6 – TRV shape in RLC parallel circuit
When longer time frames are considered, typically several hundreds of micro-seconds, reflections on
lines have to be taken into account. Lines or cables must then be treated as components with
distributed elements on which voltage waves travel after current interruption. These traveling waves
are reflected and refracted when reaching an open circuit or a discontinuity. The principles are
explained in annex A.3, and a practical application for the determination of short-line fault TRV is
covered in annex A.5.2.
The most severe oscillatory or exponential recovery voltages tend to occur across the first pole to
clear of a circuit breaker interrupting a three-phase symmetrical fault at its terminal when the system
voltage is maximum. A 3 phase fault with no ground return (either an ungrounded fault ,or an
ungrounded system, or both) will produce the highest recovery voltages on the first pole to clear.
TRV peak is function of the grounding arrangement of the system, it is lower in effectively grounded
systems than in ungrounded systems (see section 5 for additional information on the TRV and power
frequency recovery voltage applied on each pole while interrupting a three-phase terminal fault). While
three-phase ungrounded faults produce the higher TRV peaks, the probability of their occurrence is
very low, therefore the TRV ratings are based on three-phase grounded faults (see section 6.1).

4
By definition, all TRV values defined in the standards are inherent, i.e. the values that would be
obtained during interruption by an ideal circuit breaker without arc voltage (its resistance changes from
zero to an infinite value at current zero).

2 Exponential (overdamped) TRV

A typical exponential TRV is shown in Figure 7. It is obtained typically on the source side of a circuit
breaker during interruption of a fault at the circuit breaker terminals. This exponential part of TRV
Z1
occurs when the equivalent surge impedance of the n connected lines in parallel (Zeq = α where
n
Z1 is the positive sequence surge impedance of a line and α is a factor equal to 1.5 in the case of
three-phase ungrounded faults and a function of Z0/Z1 in other cases) is lower than 0.5 Leq / Ceq
(Leq = equivalent source inductance, Ceq = equivalent source capacitance). Detailed explanations are
given in A.2 and A.3.
−t /τ
The exponential part of TRV, defined by equation Vcb = Vo (1− e ) , appears also as traveling
waves on each of the transmission lines. Reflected wave(s) returning from open line(s) contribute also
to the TRV as shown on figures 7 and 9.

Figure 7 Exponential TRV characteristic

α Z1 di
The initial rate-of-rise of recovery voltage is given by RRRV = × . As higher short-circuit
n dt
currents are associated with an increasing number of connected lines and the TRV is less influenced
by the transformer natural frequency, the RRRV tends to decrease when the short-circuit current
increases.
As an example, Figure 8 shows the one line diagram of a 550 kV substation. Figure 9 illustrates the
TRV seen by circuit breaker A when clearing the three-phase fault shown in figure 8 (circuit breaker B
is open). This waveform is overdamped and exhibits an exponential waveshape. In figure 9, a
reflection occurs from the end of the shortest line after approximately 535 µs, causing a slight increase
in the TRV crest. Figure 9 also shows the TRV capability curve of a circuit breaker indicating that the
breaker TRV capability exceeds system requirements.

5
 162 km (100 mi) 81 km (50 mi)

550/245 kV

162 km (100 mi)

550/245 kV
162 km (100 mi)

THREE-PHASE TO
GROUND FAULT

Figure 8- System configuration

VOLTAGE (kV)

900

800
SYSTEM TRV
700 TRV CAPABILITY FOR A
STANDARD BREAKER
600

500

400

300

200

100 REFLECTED WAVE

0
0 100 200 300 400 500 600 700 800 900 1000

TIME (µs)

Figure 9 – Comparison of TRV capability for 550 kV circuit breaker and system TRV

3 Oscillatory (underdamped) TRV

The oscillatory TRV shown in Figure 10 occurs generally when a fault is limited by a transformer or a
series reactor and no transmission line or cable surge impedance is present to provide damping.

6
 Figure 10 – Transformer limited fault and oscillatory TRV characteristic

Even when overhead lines are present, it is possible for the recovery voltage to be oscillatory. To be
oscillatory, the surge impedance of a source side line has to be such that the equivalent surge
impedance Zeq is equal or higher than 0.5 Leq / Ceq (Leq = equivalent source inductance, Ceq =
equivalent source capacitance), it follows that the number of lines connected is necessarily small and
that the short-circuit current is generally low, up to 30% of rated short-circuit current. An exception
could be a transformer circuit breaker. in this case, the maximum fault duty could have an oscilatory
TRV. Detailed explanations are given in A.3.
The characteristics (peak value and rate-of-rise) of oscillatory TRVs are often covered by the
rated values defined in ANSI/IEEE Std C37.06 for terminal faults at 10% or 30% of rated
short-circuit current.

Severe TRV conditions occur in some cases, for instance when short-circuit occurs
immediately after a transformer without any appreciable additional capacitance between the
transformer and the circuit breaker. In such cases, both the peak voltage and rate-of-rise of
transient recovery voltage may exceed the values specified in ANSI/IEEE Std C37.06 (see
Figure 12). Such cases are covered by the TRV ratings fast time-to-peak values for definite
purpose circuit breakers given in ANSI C37.06.1-2000.

As an example, Figure 11 shows the case of a 40 kA, 145 kV circuit breaker that has to clear
a three-phase fault at 10% of its rating. The resultant TRV is shown in Figure 12. This TRV is
determined by the inductance and capacitance of the transformer and by the capacitance
between the transformer and the circuit breaker. In this particular case, the circuit breaker
does not have the capability to withstand the imposed system TRV, unless the system is
modified.

Figure 11 - Fault location

7
 300
TRV CAPABILITY FOR A STANDARD BREAKER
250

200
VOLTAGE (kV)

150

100

50 SYSTEM TRV CURVE

0
0 10 20 30 40 50 60 70
TIME (µs)

Figure 12 - Comparison of TRV capability for 145 kV circuit breaker (at 10% of its rated
interrupting current capability) and system TRV

The system TRV curve can be modified by a capacitance and then be within the standard
capability envelope (see section 7).

8
4 Short-line fault – Triangular waveshape

Triangular-shaped TRVs are associated with short-line faults (SLF). After current interruption, the line
side voltage exhibits the characteristic triangular waveshape shown in Figure 13 (see annex A.5.2 for
explanations on the triangular waveshape).

The rate-of-rise of the saw-tooth shaped TRV is a function of the line surge impedance. The rate-of
rise is generally higher than that experienced with exponential or oscillatory TRVs, however the TRV
peak is generally lower.

Because overhead lines have distributed electrical parameters (series resistance, shunt conductance,
shunt capacitance and series inductance), the line side voltage oscillates in the form of a travelling
wave with positive and negative reflections at the open breaker and at the fault, respectively. The line
side component of the recovery voltage has a sawtooth shape and a high rate of rise. Generally, the
source recovery voltage rises much more slowly and only the line side triangular recovery voltage is
important during the early portion of the TRV (see Figure 14). The closer the fault is to the circuit
breaker, the higher the initial rate of rise of the line side recovery voltage due to the higher fault
current, while the crest magnitude of this line side triangular wave decreases due to the shorter time
for the reflected wave to return.
The amplitude and rate-of-rise of TRVs for these short-line faults are determined on a single-phase
basis during their early time periods (typically less than 20 µs) when the source side voltage changes
only slightly. Later in time, the TRV is less severe than for three-phase terminal faults.

Figure 13 - Short-line fault TRV characteristic

9
 Figure 14 - TRV waveshape for short-line fault

VBD = source side voltage, VCD = line side voltage

The fault current for a line side fault is somewhat reduced from that obtained for a bus fault due to the
additional reactance of the line. Let IT be the fault current through the circuit breaker for a single-phase
fault at the breaker terminal, and IL be the reduced current for a line fault.

Figure 15 illustrates the single-phase circuit where the short-circuit current is limited by the source
reactance (XS) in series with the line reactance (XL λ):

V LG
I L=
XL λ + XS

Figure 15 Single-phase circuit with short-line fault

The source side reactance is given by

V
X S = LG
IT
The fault current is then

10
 V LG
IL =
X L λ + V LG / I T
where
XL is (2 L1w + L0w ) ω / 3
XL is the reactance of the line to the fault point per unit length
L1w is the positive sequence power frequency line inductance per unit length
L0w is the zero sequence power frequency line inductance per unit length
VLG is the system line-ground voltage
λ is the distance from the opening circuit-breaker to the fault

In general, it is not necessary to calculate the SLF TRV as long as the terminal fault TRVs are within
rating and transmission line parameters are within the values specified in ANSI/IEEE Std C37.04. The
transmission line parameters are given in terms of the effective surge impedance, Zeff, of the faulted
line and the peak factor, d, defined as:

Z eff
d = 2ω
XL v
where:
Zeff = (2Z1 + Z0)/3
v is the velocity of light
Zeff is the effective surge impedance of the line
Z1 is the positive sequence surge impedance
Z0 is the zero sequence surge impedance
ω is 2 π × system power frequency (377 rad/s for a 60 Hz system)

Annex A.5 gives equations for the calculation of the peak factor d as function of system parameters.
The rate of rise of recovery voltage on the line side (RL) is given by the following formula, as function
of the fault current (IL) and the line surge impedance (Z):
RL = 2 ω I L Z
The first peak of TRV seen across the circuit-breaker terminals is the sum of a line-side contribution
(eL) and a source-side contribution (eS):
e = e L + eS

e L = d (1 − M ) 2 / 3 E max kV
e S = 2 M (TL − t d )
where td is the time delay of TRV on the source side
e
TL is the time to peak TL = L
RL
M is the ratio of the fault current to the rated short circuit current
Emax is the rated maximum voltage (kV)

This rate-of-rise during SLF is significantly higher than the values that are met during terminal fault
interruption:

10.8 kV/µs for SLF with IL = 45 kA, Z = 450 Ω and a 60 Hz system

8.64 kV/µs for SLF with IL = 36 kA, Z = 450 Ω and a 60 Hz system

3 kV/µs for Terminal fault at 60 % of rated breaking current

2 kV/µs for Terminal fault at 100% of rated breaking current

11
The high rate-of-rise of SLF associated with high fault currents (of 45 kA or higher) can be difficult to
withstand by circuit breakers. In order to assist the circuit breaker during the interruption, a phase to
ground capacitor, or a capacitor(s) in parallel to the interrupting chamber(s), can be used to reduce the
rate-of-rise of recovery voltage (RRRV).

When a phase-to-ground capacitor is used, the reduction of line side RRRV is given by

Z ω I L 2 CL
RRRVL =
C L + 2.5 Cadd

where
λ XL
CL = is the total line capacitance
ω Z2
Cadd is the additional phase-to-ground capacitance

5 First pole-to-clear factor

The first–pole–to-clear factor (k pp ) is a function of the grounding arrangements of the system.

It is the ratio of the power frequency voltage across the interrupting pole before current
interruption in the other poles, to the power frequency voltage occurring across the pole or
poles after interruption in all three poles.

For systems with ungrounded neutral, k pp is or tends towards 1.5. Such systems can be met
with rated voltages less than 245 kV, however at transmission voltages, i.e. greater than
72.5kV, it is increasingly rare and effective grounding is the norm.

For effectively grounded neutral systems, the realistic and practical value is dependent upon
the sequence impedances of the actual earth paths from the location of the fault to the
various system neutral points (ratio X 0 /X 1 ). For these systems the ratio X 0 /X 1 is taken to be
≤ 3.0. Three-phase to ground faults are the basis for rating because it is recognised that
three-phase ungrounded faults have a very low probability of occurrence. For voltages, less
than 100 kV, the case of three-phase ungrounded faults is uncommon, however the situation
is automatically covered in the Standards since a first pole to clear factor of 1.5 is specified to
cover three-phase faults in non effectively grounded systems.
For special applications in transmission systems with effectively grounded neutral where the
probability of three-phase ungrounded faults cannot be disregarded, a first-pole-to-clear factor
of 1.5 may be required.

For rating purposes, two values of the first-pole to clear factor are defined for the three-phase
short-circuit condition. The choice between these two values is dependent on the system
grounding arrangement:

a) systems with ungrounded neutral: a standardised value for k pp of 1.5 is used;

b) effectively grounded systems: for standardisation purposes the value for k pp used is 1.3.

A third condition does exist, this is where the fault is single-phase in an effectively grounded
system and the last-pole-to-clear is considered. For this k pp is 1.0.

Generally it will not be necessary to consider alternative transient recovery voltages as the
standard values specified cover the majority of practical cases.

- Formula for the first-pole-to-clear factor

3X 0
k pp = ,
X1 + 2X 0

where X 0 is the zero sequence, and X 1 the positive sequence reactance of the system.

12
If X 0 >> X 1 , as in ungrounded systems then: k pp = 1.5
If X 0 = 3.0 X 1 , as in effectively grounded neutral systems then: k pp = 1.3

- Equations for the other clearing poles

a) Systems with ungrounded neutrals
As illustrated in Figure C.1, after interruption of the first phase (A), the same fault current is
carried in phases B and C (but with opposite sign). This current is interrupted by the last two
poles in series under the phase-to-phase voltage (E B – E C ) equal to 3 times the phase-to-
voltage. Each pole shares ½ of the phase-to phase voltage, so that for each pole,

3
k pp = = 0.87
2

EA

EB

EB - EC I

EC
I

Figure C.1 Ungrounded system after interruption of the first phase

b) Systems with effectively grounded neutrals

In systems with effectively grounded neutrals, the second pole clears a three-phase to ground
fault with a factor of

3 X 02 + X 0 X 1 + X 12
k pp =
X 0 + 2X1
This formula can be expressed as a function of the ratio X 0 /X 1 :

3 α 2 +α +1
k pp = where α = X 0 /X 1
2 +α
If α = 3.0, the second pole to clear factor is 1.25.
For the third pole-to-clear: k pp = 1

Table C.1 gives k pp for each clearing pole as a function of X 0 /X 1 as appropriate.

13
 Table C.1 - Pole-to-clear factors (kpp) for each pole when clearing three-phase to ground faults

neutral X 0 /X 1 Pole-to-clear factor k pp

Ratio First Second Third
ungrounded Infinite 1.5 0.87 0.87
Effectively 3.0 1.3 1.25 1.0
grounded
See note 1.00 1..0 1.00 1.0

Note: values of the pole-to-clear factor are given for Xo/X1 = 1.0 to indicate the trend in the special
case of networks with a ratio Xo/X1 of less than 3.0.
It is important to note that as the amplitude factor is the same for each pole, the multiplying factors of
table C.1 are applicable to the power frequency voltages and to the TRV on each pole.
In the special case of three-phase ungrounded faults, the pole-to-clear factors are as defined in a) for
three-phase faults in ungrounded systems.

- TRV on each pole to clear

Figure 17 shows the TRV on each pole to clear during interruption of a three-phase to ground terminal
fault in an ungrounded system. The same TRV would be applied in the case of a three-phase
ungrounded fault.

Figure 17 TRV on each pole to clear during a three-phase to ground terminal fault in an
ungrounded system.

Figure 18 shows currents and TRV on each pole to clear during interruption of a three-phase to
ground terminal fault in an effectively grounded system.

14
Figure 18 TRV on each pole to clear during a three-phase to ground terminal fault in an
effectively grounded system.

15
6 Rating and testing
6.1 Terminal fault
The TRV ratings for circuit breakers are applicable for interrupting three-phase to ground faults at the
rated symmetrical short circuit current and at the maximum rated voltage of the circuit breaker. For
values of fault current other than rated and for line faults, related TRV capabilities are given. Rated
and related TRV capabilities are described in ANSI/IEEE Std C37.04 and given in detail in ANSI
C37.06.
While three-phase ungrounded faults produce the highest TRV peaks, the probability of their
occurrence is very low. Therefore, as described in ANSI/IEEE Std C37.04, the TRV ratings are based
on three-phase grounded faults with the TRV peaks established based on the grounding
arrangements prevalent at the respective system voltages.

For circuit breakers applied on systems 72.5 kV and below, the TRV ratings assume that the systems
can be operated ungrounded. The first pole-to clear factor kpp is 1.5.

For circuit breakers applied on systems 245 kV and above, the TRV ratings assume that the systems
are all operated effectively grounded: the first pole-to-clear factor kpp is 1.3. For systems 100 through
170 kV the systems can be operated either ungrounded or effectively grounded so two TRV ratings
are available for these systems (kpp = 1.3 or 1.5). In addition, for special applications in transmission
systems where the probability of three-phase ungrounded faults cannot be disregarded, a first pole-to-
clear factor of 1.5 may be required.

The two-parameter and four-parameter envelopes, illustrated in Figures 19 and 20, have been
introduced in ANSI/IEEE Std C37.04 in order to facilitate the comparison of a TRV obtained during
testing and a specified TRV. In a similar way it is possible to compare a circuit-breaker specified TRV
capability and a system TRV obtained by calculation.

Two-parameter and four-parameter envelopes are used respectively for oscillatory (underdamped)
and exponential (damped) TRVs. For standardization purposes, two-parameter envelopes are
specified for circuit breakers rated less than 100 kV, at all values of breaking current, and for circuit
breakers rated 100 kV and above if the short-circuit current is equal or less than 30% of the rated
breaking current. Four-parameter envelopes are specified in other cases.

Figure 19 - Example of inherent test TRV with two-parameter envelope

16
 Figure 20 - Example of inherent test TRV with four-parameter envelope

The peak value of TRV is defined as follows

uc = k pp × k af × 2 / 3 ×U r
where
kpp is the first pole to clear factor (see section 7).
kaf is the amplitude factor (ratio between the peak value of TRV and the peak value
of the recovery voltage at power frequency)
Typical values of kaf are 1.4 and 1.7 respectively at 100% and 10% of rated breaking current.

A circuit breaker TRV capability is considered to be sufficient if the two or four parameter envelope
drawn with rated parameters is higher than the two or four parameter envelope of the system TRV.
This procedure is justified as it allows comparison of the circuit breaker TRV capability and the system
TRV in the two regions where a reignition is likely, i.e. during the initial part of the TRV where the
RRRV is maximum and in the vicinity of the peak voltage (uc). Generally, if the circuit breaker can
withstand the initial TRV rate of rise, and the TRV peak, it will sucessfully interrupt. These are the most
critical areas to check.

The general characteristics of the TRV envelopes defined by ANSI/IEEE Std C37.04 are illustrated in
Figures 21 and 22 as a function of the fault current magnitude.

17
 VOLTAGE

0.1 I

0.3 I 0.6 I
1.0 I

TIME
Figure 21 - TRV envelopes, 100 kV and above
(I is the rated short-circuit current)
VOLTAGE

0.1 I
0.3 I
0.6 I 1.0 I

TIME

Figure 22 - TRV envelopes, 72.5 kV and below

As explained in section 2, the rate-of-rise of recovery voltage (RRRV) decreases when the short-circuit
current is increased. For circuit-breakers rated 100 kV and above, standard values of RRRV are
2kV/µs and 3 kV/µs respectively for terminal fault tests at 100 % and 60 % of rated short-circuit
current.
Tests are required at 100%, 60%, 30% and 10% of rated short-circuit current with the corresponding
TRVs and recovery voltages [8].
Tests are generally performed with symmetrical and asymmetrical currents, as both conditions are
possible in service. However when interrupting asymmetrical currents, the rate-of-rise and peak value
of TRV are reduced (see annex A.6).

18
In a network, the initial part of the TRV may have an initial oscillation of small amplitude,
called ITRV, due to reflections from the first major discontinuity along the busbar. The ITRV is
mainly determined by the busbar and line bay configuration of the substation. The ITRV is a
physical phenomenon that is very similar to the short-line fault (see 6.2). Compared with the
short-line fault, the first voltage peak is rather low, but the time to the first peak is extremely
short, within the first microseconds after current zero. If a circuit breaker has a short-line fault
rating, the ITRV requirements are considered to be covered. Since the ITRV is proportional to
the busbar surge impedance and to the current, the ITRV requirements can be neglected for
all circuit-breakers with a rated short-circuit breaking current of less than 25 kA and for circuit-
breakers with a rated voltage below 100 kV. In addition the ITRV requirements can be
neglected for circuit-breakers installed in metal enclosed gas insulated switchgear (GIS)
because of the low surge impedance

6.2 Short line fault

The rated values for the line surge impedance Z and the peak factor d as defined in ANSI/IEEE Std
C37.04 are as follows:

Z = 450 Ω
d = 1.6

The line side contribution to the initial part of TRV is defined as a triangular wave in ANSI/IEEE Std
C37.09 as follows:

e L = d (1 − M ) 2 / 3 Ur kV

R L = 2ω M I Z .10 −6 kV / µs
TL = eL / RL µs
where

eL is the peak value of TRV on the line side (kV)

RL is the rate-of-rise (kV/µs)
TL is the time to peak (µs)
M is the ratio of the fault current to the rated short circuit current
Ur is the rated maximum voltage (kV)
I is the rated short circuit current (kA)

As can be seen in Figure 14, the TRV across the interrupter is in reality the difference between the
transient recovery voltages on the supply and on the line side. As illustrated on Figure 23, the variation
of the source side voltage increases the first peak value of TRV by es.
etotal = es + e
As a first approximation, the contribution from the source side voltage can be estimated by considering
that the variation of the source side voltage is zero until time td = 2 µs (standard value of time delay),
and then changes with a slope of 2 kV/µs (standard value of RRRV for terminal fault) up to time TL.
In this approximation, it is considered that the RRRV is the same as for three-phase terminal faults. In
reality it is reduced by the factor M as RRRV is proportional to the fault current (on the source side
RRRV = 2 × M).
es = (TL-2µs) x 2M

Tests are required at 90% and 75% of rated short-circuit current with the corresponding TRVs and
recovery voltages [8].

19
 Voltage
kV es
Slope =
2M

td = TL Time
2 µs TL - 2 µs µs

Figure 23 - Contribution of the source side voltage on TRV

6.3 Out-of-Phase

TRVs for out-of-phase conditions are specified as for terminal fault, except that the pole to clear factor
is equal to 2.0 and 2.5 respectively for systems with effectively grounded neutral and for systems with
ungrounded neutral.
Tests are required at 25% and 5-10% of rated short-circuit current with the specified TRV and
recovery voltage [8].

7 Application considerations
IEEE C37.011 (under revision) covers procedures and calculations necessary to apply the standard
TRV ratings for ac high-voltage circuit breakers rated above 1000 V.

7.1 Transformer fed fault

The standard capability curve, shown in Figure 24, is defined by a two-parameter envelope
where u c and t 3 are defined in tables 3 and 6 of ANSI C37.06 (with 10% and 30% of rated
short-circuit current).

Severe TRV conditions may occur in some cases, for instance when short-circuit occurs
immediately after a transformer without any appreciable additional capacitance between the
transformer and the circuit breaker. In such cases, both the peak voltage and rate-of-rise of
transient recovery voltage may exceed the values specified in ANSI/IEEE Std C37.06 (see
Figure 12). Such cases are covered by the TRV ratings fast time-to-peak values for definite
purpose circuit breakers given in ANSI C37.06.1-2000.

These values should be specified only when the rate of rise of the system TRV is higher than
the rate of rise of the standard capability curve defined in ANSI C37.06.

The system TRV curve can be modified by a capacitance and then be within the standard
capability envelope. Figure 24 illustrates the modified system TRV for the condition of Figure
12, but with additional capacitance assumed between the transformer and the circuit breaker.

20
 300

250 TRV CAPABILITY FOR A STANDARD BREAKER

200
VOLTAGE (kV)

150

100
SYSTEM TRV CURVE
MODIFIED BY CAPACITANCE

50

0
0 5 10 15 20 25 30 35 40 45
TIME (µs)

Figure 24 Comparison of TRV capability for 145 kV circuit breaker (at 10% of its rated
interrupting current capability) and system TRV modified by additional capacitance
between circuit breaker and transformer

The contribution of transformers to the short-circuit current is relatively larger at smaller values of
short-circuit current as in T30 and T10 conditions. However, most systems have effectively grounded
neutrals at ratings of 100 kV and above. With the system and transformer neutrals effectively
grounded, the first-pole-to-clear factor of 1.3 is applicable for all terminal fault test duties. In some
systems for rated voltages of 100 kV up to including 170 kV, transformers with ungrounded neutrals
are in service, even though the rest of the system may be effectively grounded. Such systems are
considered special cases and are covered in ANSI/IEEE C37.04 and ANSI C37.06 where test duties
T30 and T10 are based on a first-pole-to-clear factor of 1.5. For rated voltages above 170 kV, all
systems and their transformers are considered to have effectively grounded neutrals.

For currents between 10% and 30% of rated short-circuit current, values of uc and t3 can be obtained
by linear interpolation.

7.2 Series reactor limited fault

Series reactor are used to limit the short-circuit current in a line.
When line side series reactors are used, high rate-of-rise TRVs can result in much the same ways as
for transformer limited faults.
An example of a series reactor used on a 230 kV system is illustrated in Figure 25. The resultant TRV
for the case described is shown in Figure 26. The same TRV is obtained if the reactor is on the bus
side of the circuit breaker and the fault is on the line side of the circuit breaker.
This system TRV may exceed the standard capability curve, which is described by a two-parameter
envelope where uc and t3 are defined in ANSI C37.06 for 10% short-circuit breaking capability,
maximum voltage (Emax = 245 kV)
uc = 398 kV
t3 = 57 µs
For currents between 10% and 30% of rated short-circuit current, values of uc and t3 can be obtained
by linear interpolation.

Wavetraps used in transmission line communication systems may also add a high frequency

21
component to the TRV although of a lesser magnitude than a transformer or a current limiting reactor.
However, under some circumstances, wavetraps can substantially increase the TRV over that present
without the trap. A wavetrap is usually a parallel L-C device that is placed between the line and circuit
breaker.

When the system TRV exceeds a standard breaker capability, the user has two possibilities
- specify a Definite Purpose circuit breaker for fast transient recovery voltage rise times, as defined
in ANSI C37.06.1-2000. In some cases their higher TRV withstand capability will be sufficient.
- Add a capacitance in parallel to the reactor in order to reduce the TRV frequency and have a
system TRV curve within the standard capability envelope.

Figure 25 – Series reactor limited fault

400

350 TRV CAPABILITY FOR A STANDARD BREAKER

300
VOLTAGE (kV)

250

200

150

100

50 SYSTEM TRV CURVE

0
0 10 20 30 40 50 60
TIME (µs)

Figure 26 - Comparison of TRV capability for 245 kV circuit breaker (at 10% of its rated
interrupting current capability) and system TRV

7.3 Shunt reactor switching

When switching a shunt reactor, a circuit breaker interrupts an inductive current of small amplitude. As
a first approximation, the load can be simulated by an inductance L1 in parallel with a capacitance C1
(Figure 27a).
An ideal circuit breaker, i.e. without arc voltage, interrupts at current zero, at a time when the supply
and load voltages are maximum (Figure 27b).

22
 1
After interruption, the load side voltage oscillates towards zero with a frequency f L = L 1C 1
2π
and the supply side voltage varies at power frequency, like the source voltage. The TRV is the
difference between the load side and the supply side voltages.
As the frequency fL is typically in the range 1 to 5 kHz, reignitions are possible when the circuit breaker
interrupts with short arcing times. These reignitions can lead to overvoltages.
During interruption, current can be forced prematurely to zero due to arc instability. This phenomenon
called current chopping can also generate overvoltages (Figure 27c).

Figure 27 Shunt reactor current interruption a) Simplified circuit

b) Interruption without arc voltage c) Interruption with current chopping
Overvoltages can be limited to acceptable values when one of the following methods is used:
- synchronised switching,
- metal oxide varistor across the circuit breaker terminals,
- opening resistor.

Figure 28 shows interruption of a 55 MVAR shunt reactor current by an air-blast circuit breaker with
opening resistor in the 735 kV network of Hydro-Québec.

23
 Figure 28 Interruption of shunt reactor current
Upper trace = load side voltage Lower trace = TRV
Limitation of overvoltages and circuit breaker specification are covered in ANSI/IEEE Standard
C37.015 [11].

7.4 Generator circuit breakers

Special TRV requirements are applicable for generator circuit breakers installed between a generator
and a transformer.
As illustrated on Figure 29, two types of faults need to be considered.

Figure 29 Fault conditions for Generator circuit breakers

A1 System-source fault

B1 Generator-source fault

For the two types of fault, theTRV has an oscillatory waveshape and the first pole to clear factor is 1.5
in order to cover three-phase ungrounded faults.

24
TRV parameters, i.e. peak voltage uc, rate-of-rise (RRRV) and time delay, are listed in ANSI/IEEE
C37.013.
- TRV for system-source faults
RRRV for system-source faults is 3-5 times higher than the values specified for distribution or sub-
transmission circuit breakers. This is due to the fact that the TRV frequency is dominated by the
natural frequency of step-up transformer if one considers the more severe case where the capacitance
on the high side of the transformer can be neglected.
After reviewing the available transformer data of many power plants, the IEEE standardization group
has defined TRV parameters in several ranges of transformer rated power.
The RRRV can be significantly reduced if a capacitor is installed between the circuit breaker and the
transformer. It is also reduced in the special cases where the connection between the circuit breaker
and the transformer(s) is made by cable(s) [14]. This will be covered in an amendment to ANSI/IEEE
C37.013.

TRV RATE FOR SYSTEM FED FAULTS TRANSFORMER 50MVA<<=100MVA

3,6

3,4

3,2
TRV RATE (kV/µs)

81MVA
100MVA
2,8

2,6

65,5MVA
2,4

2,2

2
0 2000 4000 6000 8000 10000 12000

CABLE CAPACITANCE (pF)

Figure 30 RRRV for system-source faults – Transformers 50 MVA to 100 MVA

- TRV for generator-source faults

RRRV for generator-source faults is roughly 2 times higher than the values specified for distribution or
sub-transmission circuit breakers. The values were defined after reviewing the data of many
generators.
Standardized values are currently revised to cover applications with generators rated 10 to 100 MVA.
Due to the large time constants of generators and transformers (high X/R), Generator circuit breakers
are required to interrupt currents with a high percentage of dc component (high asymmetry). The rate-
of rise and peak value of TRV during interruption of currents with large asymmetry are reduced as
explained in annex A.6.

7.5 Selection of a circuit breaker

The TRV ratings define a withstand boundary. A circuit TRV that exceeds this boundary at rated short
circuit current, or the modified boundary for currents other than rated, is in excess of the circuit
breaker’s rated or related capability. Either a different circuit breaker should be used, or the system
should be modified in such a manner as to change its transient recovery voltage characteristics when
the withstand boundary is exceeded. The addition of capacitors to a bus or line is one method that can
be used to improve the system’s recovery voltage characteristics.

25
In special cases where the terminal fault TRV capability at 60% or 100% of short-circuit capability is
higher than rated, a breaker with a higher rated interrupting capability could be used (see ANSI/IEEE
C37.011).

8 Bibliography

[1] Naef, O., Zimmerman, C. P., and Beehler, J. E., “Proposed Transient Recovery Voltage Ratings for
Power Circuit Breakers,” IEEE Transactions on Power Apparatus and Systems, vol. PAS-84, no. 7, pp.
580–608, July 1965.
[2] Transient Recovery Voltages on Power Systems (as Related to Circuit Breaker Performance). New
York: Association of Edison Illuminating Companies, 1963.
[3]] Wagner, C.L., and Smith, H.M., “Analysis of Transient Recovery Voltage (TRV) Rating Concepts,” IEEE
Transactions, vol. PAS-103, pp. 3354-3363, Nov. 1984.
[4] ANSI C37.06, AC High-Voltage Circuit Breakers Rated on a Symmetrical Current Basis—Preferred
Ratings and Related Required Capabilities (under revision).
[5] ANSI C37.011, Application Guide for Transient Recovery Voltage for AC High-Voltage Circuit
Breakers. (under revision).
[6] ANSI C37.06.1-2000, Guide for High-Voltage Circuit Breakers Rated on a Symmetrical Current Basis
Designated “Definite Purpose for Fast Transient Recovery Voltage Rise Times”.
[7] ANSI/IEEE Std C37.04, IEEE Standard Rating Structure for AC High-Voltage Circuit Breakers (under
revision).
[8] ANSI/IEEE Std C37.09, IEEE Standard Test Procedure for AC High-Voltage Circuit Breakers (under
revision).
[9] ANSI/IEEE Std C37.013, IEEE Standard for AC High-Voltage Generator Circuit Breakers Rated on a
Symmetrical Current Basis.
[10] A.Greenwood, “Electrical Transients in Power Systems”, Second Edition, John Wiley & Sons Inc,
1991.
[11] ANSI/IEEE Std C37.015, IEEE Application Guide for Shunt Reactor Switching.
[12] Harner R.and Rodriguez J., “Transient Recovery Voltages Associated with Power System, Three-phase
transformer secondary faults”, IEEE Transactions, vol. PAS-91, pp. 1887–1896, Sept./Oct. 1972.
[13] Colclaser R.G. and Buettner D.E., “The traveling wave Approach to Transient Recovery Voltage”,
IEEE Transactions on Power Apparatus and Systems, vol. PAS-88, N°7, July 1969.
[14] Dufournet D. and Montillet G., “Transient Recovery Voltages Requirements for System Source Fault
Interrupting by Small Generator Circuit Breakers”, IEEE Transactions on Power Delivery, Vol.17, N°2,
April 2002.
[15] Griscom S.B., Sauter D.M. and Ellis H.M., “Transient Recovery Voltages on Power Systems – Part II-
Practical Methods of Determination”, AIEE Transactions, Vol. 77, August 1958, pp 592-606.
[16] Hedman, D. E., and Lambert, S. R., “Power Circuit Breaker Transient Recovery Voltages,” IEEE
Transactions on Power Apparatus and Systems, vol. PAS-95, pp. 197–207, Jan./Feb. 1976.

26
 ANNEX A

A.1 Three-phase terminal fault

During the interruption of a three-phase terminal fault, the circuit shown in Figure A.1 defines the
general electrical equivalent network for the first phase to clear. The reduced circuits are valid for short
time frames until reflections return from remote buses. Reflections are covered in A.4.
Figure A.1a) shows the corresponding one-line diagram representations, while Figure A.1b) indicates
the three-phase representation.
The equivalent circuit given by Figure A.1c) shows that it is reduced to a simple parallel RLC circuit
with components defined as follows:
- Equivalent inductance
3 L 0 L1 (A-1)
L eq =
L1 + 2 L 0

For three-phase to ground faults in effectively grounded systems, i.e. with L0 = 3 L1,
Leq = 9 L1/ 7 = 1.3 L1
For three-phase to ground faults in ungrounded systems (L0 infinite), Leq = 1.5 L1
- Equivalent surge impedance
3 Z 0 Z1 (A-2)
Z eq =
n Z1 + 2Z 0

where Z0 = 1.6 Z1 Zeq = 1.14 Z1 / n

- Equivalent capacitance
Ceq = C0 + 2 (C1 – C0)/3 = (C0 + 2C1)/3
if C0 = C1 then Ceq = C0 = C1
where
Z1 is the positive-sequence surge impedance of the transmission lines terminating at the station
Z0 is the zero-sequence surge impedance of the transmission lines terminating at the station
n is the number of lines
L1 is the positive-sequence inductance, representing all other parallel sources terminating at
the station (transformation to lower or higher voltage systems, generation, etc.)
L0 is the zero-sequence inductance, representing all other parallel sources terminating at the
station
C1 is the positive-sequence capacitance
C0 is the zero-sequence capacitance

For the special case of three-phase ungrounded faults on effectively grounded systems, Leq = 1.5 L1,
Zeq = 1.5 Z1/n, Ceq = 2/3 C1.

ANSI/IEEE C37.011 provides methods for the determination of parameters (such as L1) and gives
typical values of Z1, Z0 and capacitances of various equipment.

27
 FAULT

N
TRANSMISSION
LINES

a) Single line diagram

A B C
Z1

(Z0 - Z1)/3
(Vcb) FAULT
L1

(L0 - L1)/3
(C1-C0)/3
C0

b) Three-phase diagram

(Vcb)

Leq
Zeq Ceq

c) Equivalent circuit

Figure A.1— Circuit definition—interruption of a three-phase to ground fault

28
A.2 Exponential (overdamped) TRV
Current injection techniques can be used to solve for the circuit breaker TRV and, because the time
span of interest is short (microseconds), the interrupted current can be represented by a ramp. The
solution for the parallel RLC network as shown in Figure A.1c) is

Vcb = E1 ( 1-e-αt (coshβt + α/β sinhβt )) kV (A-3)

where
Vcb is the voltage across the open circuit breaker contacts

E1 is 2 I ω Leq in kV
ω is 2 π f= 377 rads for 60 Hz systems
I is short circuit current in kA, rms
α is 1/(2ZeqCeq)

β is α 2 − 1 /( Leq Ceq )
Zeq is in ohms
Leq is in henrys
Ceq is in farads

For many systems the circuit will be overdamped by the parallel resistance of the line surge
impedances, thus the capacitance can be neglected as a first approximation. The solution to the
simple RL circuit is then
Vcb = E1 (1 – e-t/τ) kV (A-4)
where
τ is Leq / Zeq s
The derivative of equation (A-4) at time zero is the rate of rise of the recovery voltage and is given by

dVcb /dt = R = 2 I ω Zeq 10-6 kV/µs (A-5)

The above exponential expressions [see equations (A-3), (A-4), (A-5)] describe the component of the
TRV until reflections return from remote stations associated with the transmission lines connected to
the faulted station.

A.3 Oscillatory (underdamped) TRV

If there are no lines on the bus, then the resistance is removed from the equivalent circuit in Figure
A.1c), and the TRV will be oscillatory. An approximate expression for the voltage is given in equation
(A-6). The expression is approximate because of neglecting the source impedances behind the
transformers.

Vcb = E1 [1 − cos(t / LeqC )] kV (A-6)

eq
Even when lines are present, it is possible for the recovery voltage to be oscillatory. To be oscillatory,
the surge impedance of a source side line has to be such that

Z eq ≥ 0.5 Leq / Ceq

With Zeq, Leq and Ceq as defined in A.1.
This formula shows that as the number of transmission lines is increased, the circuit is likely to be
nonoscillatory, i.e., overdamped. In most cases, however, even N=1 makes the circuit overdamped.

A.4 Reflected waves

The initial wave that was calculated in equation (A-4) appears across the breaker pole. It also appears

29
as traveling waves on each of the transmission lines. When one of these waves reaches a
discontinuity on the line such as another bus or a transformer termination, a reflected wave is
produced, which travels back towards the faulted bus. The time for a wave to go out and back to a
discontinuity is

T = 6.65 l µ k µs (A-7)

l is the distance to the first discontinuity (in kilometer)

µ is the magnetic permeability

k is the dielectric constant

for overhead lines

µk = 1.0 and Z1 about 400 ohms (less for bundled conductor 250<Z1<350)

for cables, typically

Paper insulated fluid filled k = 4 , µ = 1.0 and µk =2

PPP insulated fluid filled or EPR k = 3 , µ = 1.0 and µk = 1.73

Polyethylene k = 2.3 , µ = 1.0 and µk = 1.52

Cable Surge Impedance (ohms) 20 < Z1 < 50

At a discontinuity transmitted and reflected waves can be described by equations (A-8) and (A-9) and
Figure A.2.

Transmitted wave
et = ei [2Zb/(Za + Zb)] (A-8 )
Reflected wave

er = ei [(Zb - Za)/(Za + Zb)] (A-9)

where

ei is the incident wave

Za and Zb are the equivalent surge impedances on either side of the discontinuity

Returning to the bus, the reflections are in turn reflected to begin the process again. A typical TRV,
including the first reflection, is shown in Figure A.3. A reflected wave returning from an open ended
line will contribute to the bus side TRV as follows:

Erl = E1 (2 Zeq t/Leq) e-Zeq t/Leq kV (A-10)

Z 3L0 L1
where Z eq = Z a = 1.14 1 and Leq =
n L1 + 2 L0

30
From (A-10) it can be shown that the maximum reflected voltage is Erl max = 0.7 E1/n. The more lines
connected, the lower the magnitude of the reflected wave.

Za Zb

Figure A.2— Traveling waves at discontinuity

Figure A.3— Typical TRV including the first reflection

31
A.5 Short-line fault
Figure A.4 illustrates the single-phase circuit where the short-circuit current is limited by the source
reactance (XS) in series with the line reactance (XL λ):

Figure A.4 Single-phase circuit with short-line fault

The fault current is given by:

V LG
I L=
XL λ + XS

The source side reactance is given by

V
X S = LG
IT
The fault current is then
V LG
IL = (A-11)
X L λ + V LG / I T
where
XL is (2 L1w + L0w ) ω / 3 (A-12)
XL is the reactance of the line to the fault point per unit length
L1w is the positive sequence power frequency line inductance per unit length
L0w is the zero sequence power frequency line inductance per unit length
VLG is the system line-ground voltage
λ is the distance from the opening circuit-breaker to the fault

A.5.1 Calculation of peak factor (d)

The peak factor (d) is the ratio of amplitudes of the first peak of the TRV (UL = VCDo + VCDp as shown
on Figure 14) to the peak voltage on the line side prior to interruption (Uo = VCDo):
d = UL / Uo (A-13)
The rate of rise (R) of the line side TRV is equal to the effective line surge impedance (Zeff) multiplied
by the slope of current at current zero (di/dt) as follows:

R= 2 ω IL Zeff
with di/dt = 2 ω IL

32
The time to the first peak UL is equal to two times the travel time from the circuit breaker to the fault
point (time necessary for the traveling wave to reach the fault and be reflected back to the circuit
breaker) : 2 λ / v = tL

where
λ is the distance from the opening circuit-breaker to the fault
v is the velocity of light

It follows that the first peak of the line side TRV is given by :

UL = Zeff di/dt 2 λ /v (A-14)

Uo = XL λ IL 2 (A-15)

From (A-13), (A-14) and (A-15):

Z eff
d = 2ω (A-16)
XL v
where
XL = (2 L1w + L0w ) ω/3
Zeff = (2Z1 + Z0)/3 (A-17)

Z1 = L1 / C1 (A-18)

v = 1 / L1C1 (A-19)
Z1 = positive sequence surge impedance
Z0 = zero sequence surge impedance
L1 = high-frequency positive sequence line inductance per unit length
C1 = high-frequency positive sequence line capacitance per unit length

The effective surge impedance Zeff is influenced by bundle and tower configuration.

From (A-18) and (A-19):

Z1 / v = L1 (A-20)
From (A-16), (A-17), (A-18) and (A-20):
(2Z1 + Z 0 ) L1 3
d = 2ω
(3Z1 )(2 L1ω + L0ω ) ω
(2 + Z 0 / Z1 ) L1
or d =2 (A-21)
(2 L1ω + L0ω )
If the high-frequency inductance L1 is equal to the power frequency inductance of the line L1w, formula
(A-21) simplifies to:
(2 + Z 0 / Z1 )
d =2 (A-22)
(2 + L0ω / L1ω )
if L0w/L1w ≈ 3 is assumed for high-voltage networks:

d = 0.4 (2 + Z 0 / Z1 ) (A-23)

In practice the high frequency inductance L1 is lower than the power frequency value L1w., and losses
which are always present have been neglected in the calculation. For these reasons the peak factor
value obtained by (A-23) is conservative.

As the ratio Z0/Z1 is always lower than 2 for HV networks, 72.5 kV to 550 kV, values of the peak factor
(d) is equal or lower than 1.6. Therefore the standardised value of 1.6 is conservative.

Using (A-17) and (A-23) the peak factor is

33
 d = 1.2 Z eff / Z1
Typical values of Zeff and Z1 are given in Table A.1 taken from ANSI/IEEE C37.011.

The line side recovery voltage crest is then equal to

U L = d U0 = d X L λ IL 2 (A-24)

Table A.1— Typical surge impedances

System Z0 Z1* Zeff‡
(kV) (Ohms) (Ohms) (Ohms)
Overhead lines 145 560 350 420
245 525 375 425
362 430** 280** 330**
550 430** 280** 330**
800 400** 265** 310**
Cables 72.5 Cable surge impedance depends on the cable type and
145 configuration. Typically Z1 and Zeff range from 50 to
245 75 Ω with Z1 ≅ Zeff
362

SF6 buswork All voltages 55 55

* Used for three-phase grounded terminal faults

** Bundled conductors assumed for 362 kV class lines and above
‡
Used for short line faults where Zeff = (2 Z1 + Z0)/3 and Z0 is determined at switching surge
frequencies.
These values do not take into account conductor clashing. Calculations performed on 420 kV lines
have shown that the effective surge impedance is between 434 Ω and 450 Ω when bundle
contraction during fault is considered.

34
A.5.2 Graphical determination of line side voltages by traveling waves
At the time of interruption (t= 0) the instantaneous voltages on the circuit breaker terminals (points B
and C in figure A.4) are at a maximum. The voltage decreases linearly along the line and is zero at the
location of the fault (point D).
As shown on the upper part of Figure A.5, the voltage distribution can be considered as the sum of two
waves, of equal amplitude, that are traveling in opposite directions.
In accordance with the theory given in A.4, when a traveling wave reaches the open circuit (Fig A.4
point B), the reflected wave is of same amplitude and with the same polarity.
When a traveling wave reaches the short-circuit (fig A.4 point D), the reflected wave is of the same
amplitude as the incident wave but of opposite polarity.
The application of these two basic rules leads to Figure A.5 which shows the distribution of voltages
on the faulted line at times 0, tL/4, 3tL/8, tL/2, 3tL/4 and tL. Time tL represents the time necessary for
the traveling wave to reach the fault and be reflected back to the circuit breaker.

V O LT AG E (p.u.)

2 x1
T IM E = 0

x = 0 x = L
2 x 0.5

x = 0.5 L
T IM E = t L /4

T IM E = 3 t L /8
2 x 0.25

x = 0.75 L

T IM E = t L /2
0

T IM E = 3 t L /2

2 x -0.5 x= L
x= 0.5 L
T IM E = t L
0

2 x -1

Figure A.5 Traveling waves on a faulted line after current interruption

Views in space with time “frozen”
Blue = wave (or second reflection) moving to right
Light Blue = first reflection of blue wave moving to left
Red = wave or second reflection moving to left
Pink = first reflection of red wave moving to right

35
The total voltage at each point of the line, and at any given time, is the sum of all waves (Blue, red, light
blue, pink). Figure A.6 shows the resulting distribution of voltage on the line at several times after
current interruption.

Figure A.6 Voltage distribution on the line at different times after current interruption
Figure A.7 shows the corresponding time-variation of voltages at different locations on the line: at the
circuit breaker terminal (fig A.4 point C, (x=0)), half-way to the fault (x=0.5L) and three-quarters of the
way to the fault (x=0.75L).

VOLTAGE (p.u.)
2
VOLTAGE AT CIRCUIT BREAKER
TERMINAL POINT C x = 0

tL/4 0.5 tL 3 tL/4 tL 1.5 tL

0
TIME

VOLTAGE 3/4 OF WAY

TO FAULT x= 0.75 L

VOLTAGE HALF WAY

TO FAULT x = 0.5 L
-2

Figure A.7 Time variations of voltages at three locations on the faulted line

36
The voltage at the circuit breaker terminal has the well known triangular waveshape. If the voltage on
the supply side terminal (figure A.4 point B) is assumed to be constant, since the TRV frequency on
the supply side is very low when compared with the line side TRV, the TRV across the circuit breaker
terminals has the waveshape shown on figure 13. The forgoing treatment of short line fault TRV has
ignored the effects of damping. Damping occurs due to line losses, and imperfect reflections. The
fault is not usually a complete short circuit. The open circuit breaker is not an infinite impedance.
(especially in the first microsecond) Damping will give rise to lower peak factors than those used in
standards.

A.5.3 Calculation of Short-line fault quantities

In this sub-clause we give practical applications of the theory developed in sections 4 and 6.2.
We consider the case of a 245 kV circuit breaker application in a network with a single-phase short-
circuit current of 40 kA and a power frequency of 60 Hz.
According to the notations in 4 and 6.2, we have

Ur 245
V LG = = = 141.5 kV
3 3
I = I T = 40 kA

A.5.3.1 Calculation of the line length corresponding to a fault with 90% of rated short-
circuit current

M = 0.9

I L = 0.9 I T = 0.9 × 40 = 36 kA
V LG 245
Source side reactance XS = = = 3.536 Ω
IT 3 40
As explained in section 4 :

V LG 245 / 3
I = 36 = =
XL λ + XS X L λ + 3.536
it follows that
245
36 ( X L λ + 3.536) =
3
245
XL λ = − 3.536 = 3.929 − 3.536 = 0.393 Ω (1)
36 3
where
XL is the reactance of the line to the fault point per unit length
XL is (2 L1w + L0w ) ω/3
L1 is the positive sequence power frequency line inductance per unit length
L0 is the zero sequence power frequency line inductance per unit length
VLG is the system line-ground voltage
λ is the distance from the opening circuit-breaker to the fault
ω is 2 π × system power frequency (377 rad/s for a 60 Hz system)

37
 ω 2 X1 + X 0
X L = (2 L1 + L0 ) =
3 3
with

X 1 = L 1ω = 0.5 Ω / km
X 0 = L 0ω = 1.2 Ω / km
2 × 0.5 + 1.2 2.2
XL = = Ω / km (2)
3 3
From (1) and (2)
0.393 3 × 0.393
λ = = = 0.536 km
2.2 / 3 2.2

A.5.3.2 Calculation of the fault current corresponding to a length of faulted line

In this second example of calculation we consider the case of a fault 1.5 km away on the line.
2.2
XL λ = 1.5 = 1.1 Ω
3
245 / 3 141.45
I= = = 30.5 kA
1.1 + 3.536 4.636
30.5
It follows that M = = 0.76
40
The fault current is 76% of the maximum (single-phase) short-circuit current.

A.5.3.3 Calculation of the first peak of TRV

In the example A.5.3.1, M is 0.9 i.e. the fault current is 90% of the rated short-circuit current.

a) Contribution of the line side voltage (e)

e = d (1 − M ) 2 / 3 Ur
e = 1.6 × 0.1 × 2 / 3 × 245 = 32 kV

b) Contribution of source side voltage (e S )

As explained in sub-clause 6.2, the contribution of the source side voltage is:

e S = 2 x M (TL − 2) (3)

The time to peak TL is determined as follows:

RL = 2 ω Z M I
R L = 2 × 377 × 450 × 0.9 × 40 10 − 6 kV / µs
= 0.24 × 0.9 × 40 = 8.64 kV / µs
e 32
TL = = = 3.7 µs (4)
RL 8.64

38
From (3) and (4):

e S = 2 × 0.9 × (3.7 − 2) = 3 kV
The first peak of TRV across the circuit breaker terminals is then

eT = e + eS = 32 + 3 = 35 kV

eT 35
The rate-of-rise of recovery voltage is RRRV = = = 9.46 kV / µs
TL 3 .7
Figure A.8 shows the time variations of the TRV and supply voltage obtained by calculation of
transients in a 245 kV network with the same short-line fault conditions. These curves have been
determined by a digital transient program. It can be verified the same values of peak TRV (eT ) and
RRRV are obtained by the computer program and the simplified method of calculation.

Figure A.8 TRV and supply voltage during SLF 90% 40 kA 60 Hz 245 kV

Some engineers confuse the voltage peak that can be calculated from the peak factor defined in the
standards, and the TRV peak (as shown in figure A.8) that is by definition the peak value of voltage
seen across the circuit-breaker terminals. For this particular case, we have

- contribution of the line side voltage : e = 1.6 × 0.1 × 2 / 3 × 245 = 32 kV

with peak factor = 1.6 p.u.

- inherent TRV seen by the circuit breaker eT = e + eS = 32 + 3 = 35 kV

- ratio of TRV to the crest value of the steady state voltage at the circuit breaker before interruption
35
= 1.75 p.u.
0.1 × 2 / 3 × 245

The example A.5.3.3 shows that in many cases simple calculations, using the equations in this
document, can be done to determine the main characteristics of a TRV, at least approximate values of
parameters, and to check the validity of complex digital simulations.

39
Figure A.9 shows the TRV calculated by digital simulation in the case of a short-line fault at 75% of
40kA 60 Hz in a 245 kV network. Following the procedure presented previously, the reader is
encouraged to calculate the peak value of TRV and the RRRV, and to validate his calculation with the
curve shown in figure A.9 (peak TRV = 93.6 kV, RRRV = 8.4 kV/µs).

Figure A.9 TRV during SLF 75% 40 kA 60 Hz 245 kV

A.6 Effect of asymmetry on transient recovery voltage

The TRVs that occur when interrupting asymmetrical current values are generally less severe than
when interrupting the related symmetrical current because the instantaneous value of the supply
voltage at the time of interruption is less than the peak value (see Figure A.10).
When the circuit breaker interrupts, at current zero, the recovery voltage is lower than for the
symmetrical case, due to the DC component of the asymmetrical current, as the TRV oscillates around
a lower instantaneous power frequency voltage value.
Circuit breakers have the capability of interrupting these asymmetrical currents provided that they are
applied within their rating.
Note: IEEE Std C37.081a-1997 gives the reduction factors of TRV peak and rate of rise of recovery
voltage (RRRV) when interrupting asymmetrical currents.

40
 SUPPLY VOLTAGE

CURRENT

TIME

Figure A.10 Supply voltage and asymmetrical current

A.7 Effect of circuit breaker on transient recovery voltage

The circuit TRV can be modified or changed by the circuit breaker’s design or by the circuit breaker’s
action. Therefore, the transient recovery voltage measured across the terminals of two different types
of circuit breakers under identical conditions can be different. Recognizing the modifying abilities of
each of the various circuit breakers would be an immense task when either calculating a TRV or
specifying a related value for the circuit breaker.

To simplify both rating and application, the power system electrical characteristics are defined or
calculated ignoring the effect of the circuit breaker. Thus, the TRV, which results when an ideal circuit
breaker interrupts, is used as the reference for both rating and application. This TRV is called the
inherent TRV. An ideal circuit breaker has no modifying effects on the electrical characteristics of a
system and, when conducting, its terminal impedance is zero; at current zero its terminal impedance
changes from zero to infinity.

When a circuit breaker is fitted with voltage-distribution capacitors or with line-to-ground capacitors,
these capacitors can reduce significantly the rate-of-rise of TRV during short-line faults.

Also the remaining arc conductivity, which exists during a few microseconds after current zero, can
reduce the RRRV. It is especially the case with circuit breakers with large blast pressures, such as air
blast circuit breakers.

On the other hand, as has been explained in 7.3, current chopping can increase the load side voltage
and the TRV in the case of shunt reactor switching.

41