g

GE Multilin
Technical Notes

Dimensioning of Current
Transformers for Protection
Applications
GE Publication No.: GER-3973A

Copyright 2003 GE Multilin

Transients on Current Transformers

INTRODUCTION When system protection installed on CT secondaries must correctly respond to short cir-
cuit transients, especially during the first few cycles, fault conditions (even more than
steady state under load conditions) must be taken into account when using current trans-
formers. As such, it is necessary to define how much a CT must be oversized to avoid
saturation due to the asymmetrical component of the fault current (the DC offset or expo-
nential component).

The initial value of this DC offset is dependent on the voltage incidence angle (the volt-
age value when the fault occurs). The line parameters may be between 0 and 2I sc ,
where I sc represents the RMS value of the short-circuit symmetrical current.

Considering this maximum value, the transient short-circuit current is defined by the fol-
lowing equation:

t T 1
i ( t ) = I sin ( t + ) + I sin ( ) e (EQ 1)

Ep
where: I = peak value of the current = ------------------------------
2 2 2
R + L
Ep = peak system E.M.F.
R = system resistance
L = system inductance
= 2f
= angle on the voltage wave at which fault occurs
= arctan ( X R )
T1 = X R of the power system

1
 GER-3973A: Dimensioning of current transformers for protection applications

Assuming that the secondary load is essentially resistive, the necessary flux in the CT to
avoid saturation is defined as follows:

ts ts
T 1 T 2 ----- T
- ------
T
T = A ------------------ e 1 e 1 sin t (EQ 2)
T1 T2

where: T1 = line time constant or primary time constant = L R

T2 = CT time constant or secondary time constant
A = peak value of symmetrical AC flux
ts = any given time during which maximum transient flux will remain without
CT saturation, or the time after which saturation is permitted.

For T2 >> T1 (the case of TPY and TPX class CTs with and without air gaps), Equation
2 is reduced to:

ts
------
T
T = A T 1 e 1 sin t (EQ 3)

As the load and wiring are mainly resistive, we can consider sin t = 1 ; as such, Equa-
tion 3 is reduced to:

ts
------
T1
T = A T 1 e + 1 (EQ 4)

Finally, since ts (relay response time + circuit breaker operating time) is normally much
greater than T1, the above expression can be reduced to:

T = A ( T 1 + 1 ) (EQ 5)

During faults, the CTs are forced to develop a flux sufficient to feed fault current to the
secondary with two components: the exponential component (DC offset asymmetrical
component) and the AC component (symmetrical component). The resultant voltage
must be greater than that necessary to feed the load connected in the secondary side of
CTs without distortions caused by saturation. Hence, the necessary oversize factor Ks is
defined by:

transient = DC + AC = K s A (EQ 6)

where the overdimensioning or transient factor is:

ts
-----
-
T1
K s = T 1 e sin t (EQ 7)

2 GE Multilin
 GER-3973A: Dimensioning of current transformers for protection applications

Voltage on CT Secondaries During Faults

DESCRIPTION Testing and experience have shown that the performance of many relays can be
adversely affected by moderate degrees of CT saturation. However, since it is not eco-
nomically feasible to test and determine the performance of all relays with different
degrees of saturation, it is common practice to specify CT requirements for various pro-
tective schemes. The requirement generally specified is that the CTs should not saturate
before the relays operate for some specified fault location.

To meet this criterion, the required transient performance for a current transformer can
be specified by calculating the minimum required saturation voltage. Generally, different
standards as IEC 185, BS3938, or ANSI/IEEE C5713 fix this voltage through the general
expression:

Vs = K0 Ks KR I2 R2 (EQ 8)

where: Vs = saturation voltage defined by the intersection of the extensions of straight

line portions (unsaturated and saturated regions) of the excitation curve
I2 = symmetrical fault current in secondary amps.
R2 = total secondary resistance burden including CT secondary, wiring loop
resistance, lead resistance, and load resistance.
K0 = the effect of the offset present during the fault (see details below)
KR = remanent flux factor (see details below)
Ks = saturation or transient factor (see details below)

The offset present during the fault (K0) is a function of the time when the fault occurs,
being maximum at zero voltage (0 or 180). Incidence angles of the faulted voltage near
90 generally produce a lower offset effect. Therefore, this factor applies in those cases
where offset exceeds 0.5 p.u.

The remanent flux can remain in the core due to the following:
The excitation current leads the load current by 90 and thereby under normal control
open commands, the load current is cut near or at zero crosses, but the excitation
current in the CT has significant value.
DC tests performed on the CTs.
The effect of the DC component on offset fault currents (exponential component)
which is interrupted when tripping the circuit breaker.

The saturation or transient factor Ks is expanded as follows (as per Equation 2):

ts ts
T 1 T 2 ----- -
T1
------
T1
K s = ------------------ e e sin t (EQ 9)
T1 T2

where: T2 = secondary time constant

T1 = time constant of the DC component of the fault component; this is propor-
tional to the X / R ratio of the system
= system angular frequency
ts = time to saturation; this is equal to or greater than the relay operating time

Equation 2 is valid for CTs with air-gapped cores because of their low magnetizing
impedance and then with low secondary time constant T2. The air-gaps used in CTs

GE Multilin 3
 GER-3973A: Dimensioning of current transformers for protection applications

tends to drastically reduce the effect of the remanent flux left in the core as a result of its
lower magnetizing impedance and much lower secondary time constant. The effect of
the remanent flux is also to reduce the time to saturation. This factor may vary from 1.4 to
2.6 times the rated flux in the core.

For a closed-core CTs (normal CTs), if the secondary time constant T2 is too high
( L magnetizing before saturation), Equation 7 does not include it. As such, a conserva-
tive value for time to saturation will result.

Time to Maximum Flux Time to Saturation

DESCRIPTION After the appearance of the short circuit, the flux 0 and the corresponding magnetizing
current I0 will reach a maximum at a time defined by:

T1 T2 T
t ( max ) = ------------------ ln -----1- (EQ 10)
T1 T2 T2

The time to saturation is given by the following expression:

XR Ks 1
t s = ------------ ln --------------- (EQ 11)
2f XR

where: Ks = V saturation I fault R 2

Vsaturation = saturation voltage as defined by Equation 8
Ifault = secondary fault current
R2 = total loop resistance
X / R= reactance to resistance ratio of any given circuit, generator, etc. See
the following tables and curves.

The rate of decay of the DC component is proportional to the ratio of reactance to resis-
tance of the complete circuit from the generator (source) to the short circuit.

If the ratio of reactance to resistance is infinite (i.e. zero resistance), the DC component
never decays. On the other hand, if the ratio is zero (all resistance, no reactance), it
decays immediately. For any ratio of reactance to resistance in between these limits, the
DC component takes a definite time to decrease to zero.

In generators, the ratio of subtransient reactance to resistance may be as high as 70:1;

as such, it may take several cycles for the DC component to disappear. In circuits
remotely located from generators, the ratio of reactance to resistance is lower, and the
DC component decays more rapidly. The higher the resistance in proportion to the reac-
tance, the more I2R loss from the DC component, and the energy of the direct current is
dissipated faster.

Generators, motors, and circuits all have a certain DC time constant that refers to the
rate of decay of the DC component. The DC time constant is the time, in seconds,
required by the DC component to drop to approximately 37% of its original value at the
instant of short circuit. It is expressed as the ratio of inductance in Henrys (V s / A) to
resistance in Ohms. This is merely a guide to illustrate how quickly the DC component
decays.

4 GE Multilin
 GER-3973A: Dimensioning of current transformers for protection applications

TYPICAL X/R RATIOS Typical values of X / R ratios of distribution and transmission lines, depending on their
rated voltages and geometrical configuration, are shown in the following table.

TABLE 1. X/R Ratios for Distribution and Transmission Lines

115 kV 138 kV 230 kV 380 kV 500 kV

Sequence 69 kV (avg.) (avg.) (avg.) (avg.) (line type) (line type)
X1 / R1 2.30 3.40 3.98 7.36 9.8 (horiz.) 24.3 (horiz)
9.6 (delta) 18.5 (vertical)
X0 / R0 1.95 3.05 4.23 4.08 3.2 (horiz.) 3.5 (horiz.)
3.3 (delta) 5.0 (vertical)

The table below shows the X / R ratios for selected power system elements as a function
of their rated power.
TABLE 2. X/R Ratios for Other Power System Elements

Large Power
Generators Transformers Reactors Utilities
40 to 120 see curve 40 to 120 15 to 30
typically 80 typically 80 (near generating plant)

Power transformer X / R ratios for 0.03 to 3.0 MVA and 3 to 200 MVA rated transformers
are indicated in the curves below:
10
9
8
7
6
X/R Ratio

5
4
3
2
1
0
0 0.5 1 1.5 2 2.5 3 3.5

ONAN MVA Rating (0.03 to 3 MVA)

50

40
X/R Ratio

30

20

10

0
0 50 100 150 200 250
ONAN MVA Rating (3 to 200 MVA)
842738A1.CDR

FIGURE 1. Power Transformer X/R Ratios

GE Multilin 5
 GER-3973A: Dimensioning of current transformers for protection applications

Tripping Time of Protection Devices

DESCRIPTION Instantaneous overcurrent (ANSI device 50) and distance (ANSI device 21) protection
normally operate in a 15 to 30 ms range. As such, dimensioning factors must recognize
that relay tripping times should be lower than time to saturation ts. To guarantee the cor-
rect operation of protection devices, Equation 2 must be applied with parameter t repre-
senting the instantaneous overcurrent operating time of the relay. The following table
shows typical tripping times for different GE relays and the necessary overdimensioning
factor Ks, using a class TPX CT with a secondary time constant T2 = 3 seconds.

TABLE 3. Tripping Times for Selected Relays

Primary time constant T1

Instantaneous (after which saturation is
Relay operating time permitted) Overdimensioning factor Ks
MIC / MRC 25 ms 40 ms 6.81
60 ms 7.39
70 ms 7.57
80 ms 7.72
DLP 20 ms 70 ms 6.45
ALPS 10 ms 70 ms 3.92
DGP 25 ms 70 ms 7.57
489 45 ms 70 ms 11.34
SMOR 25 ms 70 ms 7.57
DTP 45 ms 70 ms 11.34

Resultant Fault Voltages and CT Dimensioning

DESCRIPTION With results shown in the table above and neglecting the K0 and KR parameters in Equa-
tion 8, the following example illustrates how to determine the resultant precision limit
and the necessary overdimensioning of the CT core (rated power dimensioning) to avoid
saturation previous to the tripping time of relays under consideration.

If the phase-to-phase short circuit current is assumed to be the same order of magnitude
as the phase-to-ground short circuit current, then a single equation should be used. If
not, then the Ks factor must be verified for both situations: the positive sequence compo-
nent during three-phase faults as well as the zero sequence component for phase-to-
ground faults. In the present case, Equation 8 will be used for all calculations.

6 GE Multilin
 GER-3973A: Dimensioning of current transformers for protection applications

EXAMPLE Given a system with the following parameters:

Ks = 6.18, Vrated = 1.38 kV at 50 Hz, Rrelay = 0.04 ,

Pshort-circuit = 0.597 GVA (assumed), CT ratio = 600:1, CT class = 5P20,
CT secondary winding resistance = 1.5 (assumed),
Lwiring = 2 10 m (6 mm2 cross-section cable, assumed), and Rwiring = 0.059 .

K0 and KR are not considered.

The saturation voltage is given by:

P short-circuit 3V rated
V s = ----------------------------------------------------- K s ( R CT + R wiring + R relay )
CT ratio
0.597 GVA ( 3 13.8 kV ) (EQ 12)
= ----------------------------------------------------------------------- 6.18 ( 1.5 + 0.059 + 0.04 )
600
= 411 V

The equivalent power is therefore:

V 20- 1.5 ( 1 ) 2 = 19 VA
P equivalent = 411
------------------------- (EQ 13)
1A

Terms and Definitions

RATED PRIMARY SHORT The rated primary short circuit current is the RMS value of the primary symmetrical short-
CIRCUIT CURRENT circuit current on which the rated accuracy performance of the current transformer is
based.

INSTANTANEOUS ERROR The instantaneous error current (Ie) is the difference between instantaneous values of
CURRENT the primary current and the product of the turns ratio times the instantaneous values of
the secondary current. When both alternating current and direct current components are
present, Ie must be computed as the sum of both constituent components:

I e = I e ( AC ) + I e ( DC )
(EQ 14)
= ( nI secondaryAC I primaryAC ) + ( nI secondaryDC I primaryDC )

PEAK INSTANTANEOUS The peak instantaneous error (i ) is the maximum instantaneous error current for the
ERROR specified duty cycle, expressed as a percentage of the peak instantaneous value of the
rated primary short-circuit current

PEAK INSTANTANEOUS AC The peak instantaneous AC component error is the maximum instantaneous error of the
COMPONENT ERROR alternating current component expressed as a percentage of the peak instantaneous
value of the rated primary short-circuit current.

100I e ( AC )
AC = ------------------------------------------------- % (EQ 15)
2I primary short-circuit

GE Multilin 7
 GER-3973A: Dimensioning of current transformers for protection applications

CT ACCURACY CLASS / The accuracy class is defined by the class index (see below) followed by the letter P.
CLASS INDEX
The class index represents the accuracy limit defined by composite error (c ) with the
steady state symmetrical primary current. This number indicates the upper limit of the
composite error at the maximum accuracy current feeding the accuracy load. The stan-
dard class indexes are 5 and 10.

There is no limit for remnant flux.

LIMIT FACTOR The limit factor represents the ratio between the limit accuracy current and the rated pri-
mary current. For protection applications this factor normally is 10 or 20.

CT CLASSES The various CT classes are indicated as follows:

Class P: These are protection current transformers designed specifically to feed
protection relays. The accuracy limit is defined by composite error AC with steady
state symmetrical primary current. There is no limit for remanent flux.
Class TPS: These are low-leakage flux current transformers. Their performance is
defined by the secondary excitation characteristics and turns ratio error limits. There
is no limit for remanent flux.
Class TPX: The accuracy limit for class TPX CTs is defined by the peak instanta-
neous error (i ) during the specified transient duty cycle. There is no limit for rema-
nent flux.
Class TPY: The accuracy limit for class TPY CTs is defined by the peak instanta-
neous error (i ) during the specified transient duty cycle. The remanent flux does not
exceed 10% of the saturation flux.
Class TPZ: The accuracy limit is defined by the peak instantaneous alternating cur-
rent component error (AC) during single energization with maximum DC offset at
specified secondary loop time constant. There are no requirements for DC compo-
nent error limit. Remanent flux is practically null.

PRIMARY AND SECONDARY The primary time constant T1 represents the time constant of the DC component of the
TIME CONSTANTS primary current on which CT performance is based.

The secondary time constant T2 represents the time constant of the secondary loop of
the CT obtained from the sum of the magnetizing and leakage inductance (Ls) and the
secondary loop resistance (Rs). Normally, this value is higher than T1 in TPS class cur-
rent transformers (about 10 seconds). The value of T2 depends on the specific require-
ments but normally oscillates between 0.3 and 1 second for TPY class CTs. For TPZ
class CTs, T2 is generally much more lower (approximately 0.07 seconds).

TIME TO MAXIMUM FLUX The time to maximum flux ( t (max) see Equation 10) is the elapsed time during a pre-
scribed energization period at which the transient flux in a CT core achieves maximum
value, assuming that core saturation does not occur.

SECONDARY WINDING AND The secondary winding resistance RCT represents secondary winding DC resistance in
LOOP RESISTANCE ohms, corrected to 75C (unless otherwise specified) and inclusive of all external burden.

The secondary loop or burden resistance (RB) is the total resistance of the secondary cir-
cuit, unless otherwise specified, and inclusive of all external burden.

8 GE Multilin
 GER-3973A: Dimensioning of current transformers for protection applications

FLUX PARAMETERS A low leakage flux current transformer is a CT for which a knowledge of the secondary
excitation characteristic and secondary winding resistance is sufficient for an assess-
ment of its transient performance. This is true for any combination of burden and duty
cycle at rated or lower value of primary symmetrical short-circuit current, up to the theo-
retical limit of the current transformer determined from the secondary excitation charac-
teristic.

The saturation flux (S) is the peak value of the flux that exists in a core during a transi-
tion from a non-saturated to a fully saturated condition. This corresponds to the point on
the B-H characteristic of the core at which a 10% increase in B causes H to be increased
by 50%.

The remanent flux (R) is the value of flux that remains in the core three minutes after
the interruption of an exciting current of sufficient magnitude as to induce the saturation
flux (S).

GE Multilin 9