Wee/ern 1lec/rici/x CoorJino/ing Council

Relaying Current Transformer

Application Guide


Relay Work Group

November
2007

TABLE OF CONTENTS


Section Title Page No.

Introduction
I. CURRENT TRANSFORMER CHARACTERISTICS .......................................................... 1
A. Core Construction ........................................................................................................... 1
B. Exciting Current .............................................................................................................. 2
C. Remanence ...................................................................................................................... 3
D. Thermal Ratings .............................................................................................................. 4
II.CURRENT TRANSFORMER BURDEN ................................................................................ 7
A. Internal Resistance oI CT ................................................ Error! Bookmark not defined.
B. EIIect oI CT Connections ................................................................................................ 7
C. EIIect oI Auxiliary CT..................................................................................................... 8
III.ESTIMATES OF CT TRANSIENT PERFORMANCE .......................................................... 9
IV. PERFORMANCE REQUIREMENTS OF CURRENT TRANSFORMERS FOR
RELAYING PURPOSES ..............................................................................................................
14
A. Distribution Feeders, Phase and Ground Overcurrent..................................................... 14
B. DiIIerentially Connected CT's Ior Buses, TransIormers and Generators .............................
...................................................................................................................................... 15
C. Directional Overcurrent and Distance ............................................................................ 21
D. EHV Transmission Line Systems .................................................................................. 24
V. EXAMPLES OF CALCULATIONS ................................................................................... 25
A. Ratio and Phase Angle Error oI CT: .............................................................................. 25
B. Estimation oI Current TransIormer PerIormance: ............................................................ 30
C. ANSI CT Relaying Accuracy Classes: ........................................................................... 33
D. EIIect oI Current TransIormer Connections on Burden .................................................. 34
E. Estimates oI Transient PerIormance .............................................................................. 42
VI. REFERENCES ................................................................................................................... 44
VII. LIST OF OPERATIONS COMMITTEE PUBLICATIONS............................................... 45
RELAYING CT APPLICATION GUIDES

Introduction

This guide was prepared by the WECC Relay Work Group Ior utility engineers who apply
current transIormers Ior protective relaying. It thus addresses the needs and concerns oI the
user. It should prove to be a ready reIerence, particularly among WECC members who are
trying to solve a mutual protection problem or operation. It is also suitable as a tutorial Ior
engineers entering the protection Iield.

David H. Colwell, since retired Irom PG&E, initiated the writing oI this guide and provided
leadership and coordination in its production. He as well contributed to the body oI the guide.
The current work group acknowledges his and the other contributors' eIIorts.
1
I. CURRENT TRANSFORMER CHARACTERISTICS
A. Core Construction and Accuracy Classes
Although there exist many current transIormer designs Ior various purposes, the basic
types oI core construction can be grouped into two categories, toroidal or wound.

The bushing, window, or bar type current transIormers are oI the toroidal core
construction. The primary winding is the main conductor passing through the center oI
the core. The secondary winding is uniIormly distributed around the toroidal core.
Essentially, all the Ilux which links the primary conductor also links the secondary
winding. The leakage Ilux, and thus the leakage reactance, is negligible. This is a
common construction Ior HV and EHV current transIormers.

Since the leakage Ilux is negligible, the excitation characteristics (Section I(b)) can be
used directly to determine perIormance. Current transIormers oI this type have an
accuracy class designation oI 'C per ANSI C57.13, indicating that ratio correction at
any current can be Calculated adequately iI the burden, secondary winding resistance,
and the excitation characteristics are known. The C or K classiIication applies to all tap
sections oI the current transIormer winding. However, the previous ANSI classiIication
(L) applied only to the Iull winding. Tap sections oI current transIormers with an L
classiIication may not be uniIormly distributed.

K class current transIormers are identical to C class devices except that K class current
transIormers shall have a knee point voltage at least 70 oI the secondary terminal
voltage rating. The secondary terminal voltage rating is the voltage which the
transIormer will deliver to a standard burden at 20 times normal secondary current, i.e.
100 amps Ior a 5 amp nominal CT, without exceeding 10 ratio error.

The presence oI leakage Ilux has a signiIicant eIIect on current transIormer
perIormance. When this Ilux is appreciable*, it is not possible to calculate ratio
correction. Current transIormers oI this type have an accuracy class designation oI 'T,
indicating that ratio correction must be determined by Test.

Wound type current transIormers, T type classiIication, are usually constructed with
more than one primary turn and undistributed windings. Because oI the physical space
required Ior insulation and bracing oI the primary winding and Iringing eIIects oI non
uniIormly distributed windings, Ilux is present which does not link both primary and
secondary windings. Figure 1 is included to clearly illustrate the eIIect but does not
reIlect usual construction practice. An auxiliary current transIormer is an example oI a
wound type current transIormer.



* As stated in ANSI C57.13, an appreciable eIIect is deIined as one percent diIIerence
between the actual ratio correction and the ratio correction calculated.


Page 2 oI 48
B. Exciting Current
In an ideal current transIormer, the primary ampereturns are equal to the secondary
ampereturns. However, every core material requires some energy to produce the
magnetic Ilux which induces the secondary voltage necessary to deliver the secondary
current. Thus, in an actual current transIormer, the secondary ampereturns are equal to
the primary ampereturns minus the exciting ampereturns. For a C or K class, or a
toroidal core constructed current transIormer, the simpliIied equivalent circuit is shown
in Figure 2. Figure 3, extracted Irom ANSI C57.13, shows the typical excitation curves
Ior a multiratio C or K class current transIormer. The maximum tolerance oI excitation
values above and below the knee are also speciIied. These curves deIine the relationship
oI the secondary exciting current (I
e
) to the secondary voltage (E
e
). The unsaturated
slope is determined by the magnetic core material. The saturated region is the aircore
reactance.

When the current transIormer core is unsaturated, the error due to exciting current is
normally negligible. When the voltage is above the knee oI the excitation curve, the
current transIormer is said to be operating in its saturated region where the exciting
current is no longer negligible. ThereIore, the ratio error oI the current transIormer
becomes much greater beyond the knee.

For T class current transIormers, the leakage Ilux can be appreciable. The exciting Ilux
should be considered, along with the leakage Ilux, in determining current transIormer
accuracy. Although a test should be done, Figure 4, also extracted Irom ANSI C57.13,
shows typical overcurrent ratio curves Ior a T class current transIormer
Page 3 oI 48
C. Remanence
Remanent Ilux can be set up in the core oI a current transIormer under operating or test
conditions. During operating conditions, remanent Ilux can be leIt in the core when the
primary current is interrupted while the Ilux density in the core oI the transIormer is
high. This may occur when clearing Iault current. Testing, such as resistance or
continuity measurements, may also leave remanence.

The remanent Ilux in the core depends on many Iactors. The most important ones are the
magnitude oI primary current, the impedance oI the secondary circuit and the amplitude
and time constant oI any oIIset transient. Since the impedance oI the secondary circuit is
generally Iixed, the magnitude oI remanent Ilux is governed by the magnitude oI the
symmetrical component oI the primary current and the magnitude oI the oIIset transient
prior to the primary current interruption. Maximum remanent Ilux can be obtained under
conditions whereby the primary current is interrupted while the transIormer is in a
saturated state.

When the current transIormer is next energized, the Ilux changes required will start Irom
the remanent value. II the required change is in the direction to add to the remanent Ilux,
a large part oI the cycle may Iind the current transIormer saturated. When this occurs,
much oI the primary current is required Ior excitation and secondary output is
signiIicantly reduced and distorted on alternate halI cycles. This phenomenon is
illustrated in Figure 5. The perIormance oI both C and T class transIormers is inIluenced
by this remanence or residual magnetism. Relay action could be slow or even incorrect.

The remanence can be corrected by demagnetizing the current transIormer. This is
accomplished by applying a suitable variable alternating voltage to the secondary, with
initial magnitude suIIicient to Iorce the Ilux density above the saturation point, and then
decreasing the applied voltage slowly and continuously to zero. II there is any reason to
suspect that a current transIormer has been subiected recently to heavy currents, possibly
involving a large DC component, it should be demagnetized beIore being used Ior any
test requiring accurate current measurement.


igure 1
LEAKAGE LUX ASSOCIATED WITH CLASS T CURRENT TRANSORMERS


Page 4 oI 48

D. Thermal Ratings
Current transIormer continuous ratings can be increased beyond nominal by use oI a
continuous thermal current rating Iactor. This Iactor is deIined in ANSI/IEEE C57.13
1993R as "The speciIied Iactor by which the rated primary current oI a current
transIormer can be multiplied to obtain the maximum primary current that can be carried
continuously without exceeding the limiting temperature rise Irom 30C ambient air
temperature. (When current transIormers are incorporated internally as parts oI larger
transIormers or power circuit breakers, they shall meet allowable average winding and
hot spot temperatures under the speciIic conditions and requirements oI the larger
apparatus". Standard rating Iactors are 1.0, 1.33, 1.5, 2.0, 3.0, and 4.0.

As an application example, a power circuit breaker with a 1600 amp continuous rating
could use 1200/5 (maximum ratio) current transIormers with a thermal rating Iactor oI
1.33. In this way the current transIormer could continuously carry 1600 amps primary
and would thereIore not limit the breaker capability.

Auxiliary current transIormer thermal ratings do not conIorm to this standard, and thus
are handled diIIerently among manuIacturers. For instance, one manuIacturer supplies
all auxiliary CTs rated below 20 amps primary with thermal rating Iactors oI 1.0, while
all CTs with primaries rated greater than 20 amps have thermal rating Iactors oI 1.5.
Another manuIacturer supplies auxiliary CTs with thermal rating Iactors varying
between 1.25 and 2.0, depending on the ratio being used.


igure 2
SIMPLIIED EQUIVALENT CIRCUIT O CT ON SECONDARY N TURN BASE

Page 5 oI 48

igure 3
TYPICAL EXCITATION CURVES OR MULTI-RATIO C or K CLASS CURRENT TRANSORMERS
WITH NON-GAPPED CORES





igure 4
TYPICAL OVERCURRENT RATIO CURVES


Page 6 oI 48




igure 5a
NORMAL HYSTERESIS CURVE








igure 5b
HYSTERESIS CURVE WITH REMANENCE


Page 7 oI 48
II. CURRENT TRANSFORMER BURDEN

A. General

The perIormance oI a current transIormer used in protective relaying is largely dependent
on the total burden or impedance in the secondary circuit oI the current transIormer. The
current transIormer core Ilux density (and thus the amount oI saturation) is directly
proportional to the voltage that the current transIormer or secondary must produce. So
Ior a given amount oI secondary current, the larger the burden impedance becomes, the
greater is the tendency oI the current transIormer to saturate.

Ideally, protective relay systems would ignore current transIormer saturation. However,
that is usually not possible; so it behooves the relay engineer to minimize current
transIormer burden impedance. ManuIacturers' publications give the burdens oI
individual relays, meters, and other equipment. Adding the resistance oI interconnecting
leads and internal resistance oI the current transIormer gives the total current transIormer
burden.
In modern solid state relays with very small burdens, the total relay burden is oIten
dominated by the lead impedance.

SuIIicient accuracy results iI series burden impedances are added arithmetically. The
results will be slightly pessimistic, indicating slightly greater than actual CT ratio
inaccuracy. But, iI a given application is so borderline that vector addition oI impedance
is necessary to prove that the CTs will be suitable, such an application should be
avoided.

The current transIormer burden impedance oI most electromechanical relays decreases as
the secondary current increases because oI saturation in the magnetic circuits oI the
devices. ThereIore, a given burden may apply only Ior a particular value oI current. II a
publication does not clearly state Ior what value oI current the burden applies, this
inIormation should be requested.

At high saturation, the burden impedance approaches its DC resistance. This eIIect is
exploited in high impedance bus diIIerential relays. Neglecting the reduction in
impedance with saturation provides a quick conservative analysis, but an accurate
calculation may be necessary iI the initial calculation indicates marginal perIormance.


B. EIIect oI CT Connections

The interconnection oI two or more current transIormers supplying a common burden
inIluences the burden seen by each individual current transIormer.

When current transIormer primaries and secondaries are connected in series, the burden
on each individual transIormer is decreased in proportion to the number oI current
transIormers in use.

When current transIormers are connected in parallel the eIIect is to increase the burden
on each individual transIormer. The amount oI increase is dependent upon the number oI
Page 8 oI 48
transIormers and the distribution oI current between the transIormers. This is the case in
breaker and a halI, ring, and double bus arrangements.

In three phase current transIormer connections, the burden on an individual current
transIormer can vary with the type oI connection (wye or delta) and the type oI Iault on
the system (1 or 2line to ground or multiphase). These topics will be discussed Iurther
in Section V.D.


C. EIIect oI Auxiliary CTs

Beware oI attempts to "stepup" current Irom the main CT to the relay with the use oI an
auxiliary CT. The auxiliary CT may be adequate, but remember the main CT will see
the burden impedance multiplied by the auxiliary ratio squared.

Page 9 oI 48
III. ESTIMATES OF CT TRANSIENT PERFORMANCE

The primary Iault current in the power system is symmetrical only aIter the transients oI the
predominantly RL circuit have decayed to zero. Considering the worst case conditions oI
switching angle and power Iactor that produce the highest oIIset, the Iault current is oI the
Iorm:
|
t
e t) ( cos | I' ) (
) /L (R
1 1
t I

where I' is the peak value oI the symmetrical wave, R
1
the resistance and L
1
the inductance
oI the entire primary circuit. A time response plot is shown in Figure 6.

The exponential portion oI this wave is a high peak nondirectional or DC component, and
is responsible Ior saturation oI CTs when it is present. This is because the DC component
causes the Ilux in the CT core to exceed the saturation level very easily. The expected Ilux
and eIIects on CT perIormance are illustrated in Iigures 8, 9A and 9B.

In most applications, the saturation oI the CT by the AC component is avoided by properly
selecting the turns ratio, burden, and CT accuracy class. As long as the product oI expected
secondary current and burden impedance does not exceed the saturation or knee point
voltage oI the CT, then the CT perIormance will be satisIactory on the C component.
Figure 9 deIines saturation voltage on a typical CT secondary excitation curve.

II saturation is to be avoided on the DC component as well as the AC component, then very
severe requirements are imposed on the CT that many times are impractical or impossible to
satisIy. A simpliIied analysis oI the secondary voltage requirements shows that the
available voltage must be ) / 1 ( R L times the voltage required Ior the AC component. In
the worst case this requirement may not be attainable.

Obviously the possibility oI saturation should be known and avoided iI possible by design
and operation oI the CT/relay combination.

The IEEE Power Systems Relaying Committee produced a report on transient perIormance
oI current transIormers in 1976 titled "Transient Response oI Current TransIormers".
Publication 76CH1134 PWR contains very useIul discussion and curves Irom which the
time to saturation can be estimated. In this section we present a method that permits direct
calculation oI the time to saturation.
Page 10 oI 48
Certain system, CT, and relay parameters must be determined beIore the curves or
equations can be used. These are as Iollows:

A. K
s
Saturation factor.
This is the ratio oI the saturation voltage (Figure 9) to the voltage determined by the
product oI the expected symmetrical secondary current and the total secondary burden
impedance.

'
+

'

2
2
1
R
N
I
J
K
X
S


Where:
V
X
saturation voltage oI CT
I
1
symmetrical primary current
N
2
secondary turns oI CT
R
2
total secondary burden oI CT

This Iactor is a measure oI the margin that exists between the available voltage and the
voltage necessary to reproduce the maximum symmetrical primary current in the
secondary burden circuit.

B. T
1
primary system time constant.
This time constant inIluences the core Iluxtime relationships and is the primary
determinant oI the time it takes the core Ilux to reach saturation.

1
1
1
R
X
T seconds

Where X
1
reactance oI the primary system to the point oI Iault,

1 1
X L

R
1
resistance oI the primary system to the point oI Iault.
377 at 60 Hz

C. T
2
Current transformer secondary time constant. This time constant also
inIluences the core Iluxtime relationship and is most important in determining the time
Ior the Ilux to return to and below saturation level as the DC component decays.



2
2 2
2
R
M L
T

seconds
Where:
L
2
burden inductance iI any
M
2
CT inductance
R
2
resistance oI total secondary burden circuit (relay leads CT winding)

Page 11 oI 48
In most cases the burden inductance is negligible and the CT inductance is the equivalent
exciting inductance, determined Irom the secondary excitation curve at the point oI
maximum permeability (see Figure 9); thus:
2 2
R I J T
e e


With these three parameters, the curves in the IEEE report can be used to determine the
time to saturation oI the CT.

As an alternative, using equations derived Irom the results oI the IEEE report and Irom
other reIerences, the time to saturation and the time to exit Irom saturation can be
calculated.

These give results in close agreement with the curves in the IEEE report.

t
S
time to saturation in seconds

t
S

'
+

'

1
1
1
1 ln
T
K
T
s


t
e
time to exit Irom saturation in seconds

t
e

'
+

'

S
K
T
T
2
1
ln





Another useIul equation can be developed by solving the Iirst equation Ior K
s
thus:

1
1
1
1
1

'
+

'

T
e
K
T
t
S
S




II we now assume t
s
goes to inIinity, that is saturation does not occur, then we can
determine the value Ior K
s
required to accommodate the symmetrical and DC
components and not saturate.

then 1
1
T K
S


Combining this equation with the deIinition oI K
s
in section A:


2 2 1
N R I J K
k S
1
1
T ; and solving Ior V
k

1
1 2 2 1
T N R I J
k


Page 12 oI 48
This equation Ior V
k
indicates that iI saturation is to be avoided when the DC component
is present, then V
k
(the knee point voltage) must be 1
1
T times the steady state
voltage requirement
2 2 1
N R I . This may be quite high and impossible to provide.



igure 5
PRIMARY CURRENT WAVES






igure 6
RISE O LUX IN THE CORE O A CURRENT TRANSORMER
Page 13 oI 48

igure 8a DISTORTION IN SECONDARY CURRENT DUE TO SATURATION








igure 7
igure 8b DISTORTION IN SECONDARY CURRENT DUE TO SATURATION


Page 14 oI 48


igure 8
SECONDARY EXCITATION CURVE 230-kV CURRENT TRANSORMER
RATIO: 1200-5 AMPERES
REQUENCY: 60 Hz


IV. PERFORMANCE REQUIREMENTS OF CURRENT TRANSFORMERS FOR
RELAYING PURPOSES

A. Distribution Feeders, Phase and Ground Overcurrent
Current transIormers which operate the time overcurrent relays used Ior Ieeder protection
will typically provide satisIactory protection iI they meet or exceed the applicable ANSI
standard. For switch gear voltage classes 8.25 through 43.8kV (where most distribution
circuits lie), the accuracy classes are as Iollows:
(1)


Multi-Ratio
Current
Transformers
Accuracy
Voltage
Classes
600:5 C100
1200:5 C200
2000:5, 3000:5 C400
4000:5, 5000:5 C800

The current transIormer primary ratio normally corresponds to the continuous current
rating oI the circuit breaker (except that 800 and 1600 amp PCBs use 1200 and 2000 amp
current transIormers).

The CT ratio is usually chosen so that approximately 5 amperes Ilow in the secondary
phase relays Ior Iull load in the primary. Where 5 amp meters are used in relay circuits, a
Page 15 oI 48
higher ratio may be chosen to permit overload readings on the Ieeder circuit. Also, the
higher ratio may be necessary or desirable to minimize the secondary current during Iault
conditions.

Since overcurrent relays have a wide range oI taps, the actual CT ratio is usually not
critical. However, using higher ratio may limit the primary sensitivity oI ground
relaying.

In installations where very high Iault currents are available, care should be exercised not
to exceed the short time (one second) rating oI devices connected to the secondary.
Another possible problem with high Iault currents is severe saturation oI the CT core,
resulting in very high voltage peaks or spikes. II the CT burden is high and the primary
current is many times the CT's continuous rating, it is possible to develop voltage spikes
oI suIIicient magnitude to damage CT insulation or switch gear secondary wiring.
(2)


Also, when the secondary current is very distorted, the induction disk element oI an
overcurrent relay will deviate Irom the published time curve.
(3)
The presence oI the DC
component oI an oIIset wave not only causes saturation oI the CT but also must be
considered when setting hinged armature instantaneous trip attachments.

Determining CT perIormance requires the excitation curve, the impedance oI the
secondary connected devices and secondary wiring and the resistances oI the secondary
CT turns. The impedance oI the meters and relays will vary with the amount oI current
because their magnetic circuits saturate at high currents. Neglecting this saturation will
yield optimistic results when checking thermal conditions but pessimistic results Ior
accuracy calculations. At very high currents, (20 times normal), most secondary burdens
become so saturated they are predominantly resistive.

On multigrounded 4wire distribution circuits, the ground relay setting is usually kept at
a relatively high value by the unbalanced load Ilowing in the neutral. Where sensitive
ground relay settings are desired, care must be exercised to assure that the high burden
introduced by a very sensitive residual relay does not drive a poor quality current
transIormer into saturation. Also, as shown in the calculation section, the impedance oI
the ground relay aIIects the distribution oI secondary current and thereby aIIects the
primary sensitivity oI the ground relaying.

DiIIerentially Connected CT's Ior Buses, TransIormers and Generators
DiIIerentially connected current transIormers Ior equipment protection produces the
most severe test oI current transIormer perIormance. This is because sensitivity and
speed oI operation oI the relay is very important.

In most relay schemes where current transIormers are diIIerentially connected, the most
important consideration is that the excitation characteristics oI all current transIormers
are well matched. This does not always ensure proper operation but it allows much
easier and dependable calculation oI perIormance.

The primary current Ilow on each oI the current transIormers that are paralleled and/or
diIIerentially connected can be vastly diIIerent and thereby the perIormance calculation
is very diIIicult. Modern bus protection can be subdivided into three main categories:
Page 16 oI 48
High Impedance Overvoltage Relays, Low Impedance Overcurrent Relays, and Medium
Impedance Percentage Restraint Relays.

1. High Impedance DiIIerentially Connected Bus Protection Relays

Since its introduction in the mid1950's high impedance relays have dominated
the bus protection practices in the U.S. When properly applied, they are
dependable and secure.

The basic setting Ior the pickup oI a high impedance relay is made so that the relay
will not operate Ior a nearby external Iault with the CT on the Iaulted circuit
completely saturated, while the remaining CTs are not saturated at all. The resulting
voltage, developed by the good CTs, must Iorce current through the impedance oI the
saturated CT and leads without exceeding the voltage pickup setting oI the relay, plus
the saIety Iactor. This calculation is only valid iI all CTs are wound on toroidal cores
and have their windings completely distributed around their cores.

The secondary wiring resistance, including the CT secondary winding, is critical in
determining the relay pickup setting. Obviously a CT with an unusually high winding
resistance will limit the sensitivity oI the relay protection Ior the whole bus.

II one or more CTs have an overall ratio that diIIers Irom the rest oI the CTs on the
bus, there is a temptation to merely connect the common ratios in parallel. On
external Iaults, the high burden oI the relay will result in the design voltage across the
CT tap used. However, by autotransIormer action, a higher voltage will appear
across the Iull CT winding. This higher voltage may exceed the capability oI the
circuit insulation.

Several possible solutions Ior the problem where the overall ratios diIIer are shown
below:

(The reader is cautioned that most oI these solutions (except Solution #1) are quite
complex in their application. The General Electric publication entitled, "Bus DiIIerential
Protection, GET6455"
(13)
should be reIerred to Ior special curves and Iormulae needed
to make the proper settings.)

Solution #1
The best solution is to make all CT ratios the same by retroIitting the oIIending breakers
with slipon CTs oI the proper ratio.

Solution #2
The relays most commonly used have Iour thyrite (or varistor) nonlinear voltage
dependent resistor disks connected in series, then in parallel with the relay. These
resistors limit the voltage which appears across the CT circuits Ior the Iirst cycle (or until
the auxiliary relay has shorted the CT leads). Since the amount oI voltage which can
appear is a Iunction oI the total relay current and the resistance oI the voltage dependent
resistor, reducing the amount oI resistance will limit the peak voltage. Using two disks
in series instead oI the usual Iour will cut the peak voltage in halI.

Page 17 oI 48
However, the energy dissipated in each disk is a Iunction oI the current through the disk
and the voltage drop across it so it is usually necessary to add a second pair oI two disks
in series in parallel with the Iirst pair to handle the extra energy.

Since the instantaneous or high current tripping relay is in series with the voltage
dependent resistor disk, its pickup increases dramatically. A typical change in pickup
setting might be Irom 3A to 50A.


Solution #3




This approach is to match the higher CT ratio with the lower one with a special
auxiliary CT. This auxiliary CT must have distributed windings on a toroidal core (a
bushing CT). (Never try to use an ordinary auxiliary CT in a current or voltage bus
diIIerential circuit.)

A variation oI this method is to use one oI the higher ratio CTs on one oI the power
circuit breakers Ior double duty (auxiliary CT and bus diIIerential CT). The
disadvantages oI this scheme is that when the PCB with the higher ratio is removed
Irom service, the bus protection must be removed Irom service.



Page 18 oI 48
Solution #4




This solution removes the concern about removing the critical matching CT Irom service,
but the diIIiculty oI determining the CT secondary current distribution in the CT
connections oIthe 1200:5 CTs limits its application.




Solution #5




This solution presents a hazard to equipment and personnel because oI the high voltage
that may be generated. It requires the calculation oI the voltage developed by highest
ratio CT when connected as described above. The peak voltage developed must be less
than the circuit insulation rating (1500 volts RMS) by an appropriate saIety Iactor, and
the tapped voltage must be enough to satisIy the relay manuIacturer's setting instruction.
Extra thyristors/varistors may be required in the relay, or across the unused portion oI the
CT(s).


2. Low Impedance DiIIerentially Connected Bus Protection Relays

Overcurrent relays have been connected in bus diIIerential circuits with varying
degrees oI success Ior over 50 years. A Iew "rules oI thumb" Ior their application
are:

a. Use the highest available CT ratios. (The maximum through Iault current should
not exceed 20 times the CT rating.)
(2)


Page 19 oI 48
b. Never use an ordinary auxiliary CT. II an auxiliary CT must be used, use a
toroidal type (bushing CT) with a distributed winding.

c. Never use an overcurrent diIIerentiallyconnected relay near a generating station
or where high X/R ratios exist since the time constant at these locations will
produce very severe CT saturation problems.

d. Never use plunger type or hinged armature instantaneous relays without time
delay in this type oI circuit unless they are set very high. They operate too Iast
and operate equally well on DC. (They may be useIul, however, when used in
coniunction with induction disk overcurrent relays. II the instantaneous unit is
set Ior the same pickup as the time unit with the contacts connected in series, by
dropping out Iaster, it prevents the time unit Irom Ialse tripping by coasting
closed aIter the error current is gone. This assumes that the drop out oI the
instantaneous device has not been extended by the DC component oI an oIIset
wave.)

e. Never use on buses that have more than three or Iour sources oI Iault current.
The current transIormer on the Iaulted line will probably saturate so severely that
the error current will cause the relay to trip Ior through Iaults.

I. A stabilizing resistor can be very useIul in improving the security oI an
overcurrent relay in a diIIerential circuit.
(See example in calculation section)
(2) (6)



3. Medium High Impedance Percentage Restraint DiIIerential Relays

The above relay was introduced in the U.S. about 1970. It violates most oI the
application rules, noted above, Ior other bus diIIerential relays in that it does not
require current transIormers with similar characteristics or even the same ratio. The
relay uses rectiIying diodes to sum the total oI the secondary currents Ior use as
restraint. The diIIerential current which Ilows is matched against this restraint and
used to operate a 1 to 2 millisecond tripping relay. The relay is designed to respond
to the output oI the current transIormers beIore they saturate and to reiect Ialse
inIormation aIter saturation. Thus, the relay does not require matched CT
characteristics or ratios, low leakage reactances or low secondary circuit resistance.
The high limits oI maximum internal or external Iault currents, and the high
sensitivity Ior internal Iaults even with an extreme umber oI sources to the bus, make
this relay easy to apply.

However it is troublesome and expensive to bring the secondaries oI all oI the PCB
CTs into the relay house instead oI paralleling them in the Iield. Since the relay
operates on 1 ampere, auxiliary CTs are usually required Ior all oI the CTs. This is
also expensive and consumes much space.


Page 20 oI 48
4. TransIormer DiIIerential Relaying

Current transIormers used Ior transIormer diIIerential relaying are subiect to several
Iactors that are not ordinarily present with other Iorms oI diIIerential protection.
Following are application principles pertaining to electromechanical relays, with
comments appropriate to solid state relays incorporated.

a. Because oI the current transIormation by the power bank, the CT ratios must be
diIIerent to compensate Ior the diIIerent primary currents. While many CT taps
and relay taps are available, they seldom make a perIect match. This, oI course,
results in error current in the relay. II the transIormer has a load tap changer, this
error will change with the tap position.

b. Since the power circuit breakers on the high and low side oI the power
transIormer are seldom oI the same voltage class, the CTs associated with them
have diIIerent characteristics and oIten diIIerent accuracy classes.

c. The power transIormer has a 30 phase shiIt iI it is connected wyedelta or delta
wye. This requires the CT to be connected deltawye (or wyedelta) to shiIt the
secondary currents into phase so that they may be compared in the relay.

d. A power transIormer connected deltawye grounded becomes a source oI zero
sequence currents Ior external Iaults on the wye side, so these currents must be
eliminated Irom the relay secondary circuits.

e. When a power transIormer is energized, magnetic inrush currents appear in the
primary circuit. These currents are oIten many times the Iullload rating and are
seen by the relay as internal Iaults.

As with other Iorms oI diIIerential protection, transIormers were originally protected
with ordinary overcurrent relays connected diIIerentially. They had to be made quite
slow and insensitive to overcome the problems mentioned above. Modern relays use
percentage restraint to take care oI the Iirst three problems (noted above), and
harmonic restraint or harmonic blocking Ior the inrush current problems.

5. Internal Faults

Faults which occur within the protected zone oI the diIIerential relay will oIten result
in very severe saturation oI at least some oI the CTs. This is oI little consequence
unless the high harmonic content oI the CT secondary current blocks the operation oI
the diIIerential relay.

A saturated CT produces a highly distorted current. Second and third harmonics
predominate initially, and each may be greater than the Iundamental. Eventually the
even harmonics will disappear. The odd harmonics will persist as long as the CT
remains saturated. For these reasons, a high set instantaneous unit should be included
in the diIIerential circuit oI harmonic restrained relays that will trip in spite oI any
harmonics.

Page 21 oI 48
6. External Faults

II only one PCB, hence, one CT is used at each voltage level, the through Iault
current is limited by the power transIormer impedance and all oI the secondary
current Ilows through the restraint windings oI the diIIerential relay. II two PCBs
are used at one voltage level, such as with a ring bus, breakeranda halI, or double
busdoublebreaker, the short circuit current is not limited by the power transIormer
impedance when it Ilows through these PCBs. II the CTs are merely paralleled, they
can saturate unequally and produce error currents that may cause an incorrect
operation. II a ring bus or breakerand ahalI scheme is involved, the chances oI
this type oI through Iault occurring is high enough that each CT should be connected
to its own restraint winding. II the PCBs are part oI a doublebusscheme, the
chances oI this through current Ilowing are remote.

Two PCBs with paralleled CTs also may present problems with security Ior
through Iaults on the low side oI the transIormer because the error current oI the
existing CTs is doubled.

Current transIormers connected in delta also may cause problems with security Ior
through Iaults because the current transIormer must circulate current through two
relays and lead burdens Ior some Iaults. (See sample problem.)

C. Directional Overcurrent and Distance
The general requirements oI minimizing the burden placed upon current transIormer
secondaries applies to directional overcurrent and distance relays as well. Using
adequate CT secondary lead conductor sizes, using higher CT ratios, higher overcurrent
relay taps when possible (since the relay burden is inversely proportional to relay tap
squared), and taking care in paralleling several CT polarizing sources Ior use with
directional elements
(1), (5)
are iust a Iew examples oI the areas oI concern when applying
these relays. Even when these precautions are taken, relay circuits may be exposed to
transient (ac and dc) and harmonic current waveIorms caused by CT saturation eIIects.
The eIIects on the diIIerent types oI relays will depend upon the speciIic relay design
used.

1. Directional Elements

The directional elements used in relays oIten utilize a "polarizing" voltage or current
source to establish a reIerence phasor relationship between the "operating"
(monitored power system) secondary current and the polarizing source value. This
phasor relationship establishes an operating torque (in an electromechanical design)
or digital signal (in a solid state design) Ior Iaults in the protected zone.
Page 22 oI 48
CT saturation can cause phase shiIts and harmonics in ,the operating current source
that are diIIerent than what may be generated in the polarizing voltage or current
source. For example, induction cup electromechanical relays tend to be Irequency
dependent; operating torque is created only Ior likeIrequency operating and
polarizing waveIorms (e.g. the Iundamental Irequency waveIorms' phase angles are
compared)
(2)
.

ThereIore, the relative phase shiIt between the compared waveIorms
may lessen the operating torque Ior Iaults in the protected zone. Under extreme
cases, phase shiIts may even cause Ialse operation Ior Iaults in the reverse direction.


As in electromechanical relays, solid state relay designs also may use a polarizing
source in addition to an operating current. Some designs incorporate waveIorm zero
crossing points Ior phase reIerence and thereIore may be susceptible to phase shiIts
caused by CT saturation eIIect. Some static relays may be higher speed (less than
1/2 cycle operate time) and lower burden than their electromechanical counterparts.
ThereIore, depending upon the CT's time to saturation (see Section III "Estimates oI
CT Transient PerIormance"), solid state relays may be less likely than
electromechanical relays to operate (or not operate) incorrectly, due to the Iact that
the static relays may have completed their measurements prior to CT saturation.
Many solid state designs Iilter the input waveIorms and consequently are Irequency
dependent, as well.


2. Overcurrent Elements

The common electromechanical induction disk designs used Ior timeovercurrent
(TOC) characteristics measure the rms operating current. Under CT saturation
conditions, the actual rms value oI the relay current will be less than under non
saturated CT conditions. ThereIore, the TOC relay element will take longer to
operate than desired. Loss oI relay coordination may result Irom this delay,
especially with the applications using more inverse timeovercurrent characteristic
relays. Time overcurrent relays typically are not very susceptible to do oIIsets due to
the Iact that do oIIsets usually have died out within the delay period oI the relay.
Instantaneous overcurrent units (IOC) can be very sensitive to do oIIsets. The do
oIIsets may cause high level transient spikes in the relay current which could cause
the IOC element to trip Ior Iault current levels below its setpoint. Additionally, Ior
CT distortion oI Iault currents near the instantaneous pick up setting, the
instantaneous relay may have an undesirable trip delay (e.g., additional 2025 ms or
more).
Page 23 oI 48
Solid state relay designs using analog or digital techniques (such as microprocessor
based) will typically Iilter the input current waveIorm to eliminate dc oIIsets,
harmonics, etc. This may alleviate some oI the problems relating to the
instantaneous elements. However, the time delay elements' operating times may
experience the same type oI unpredictability, lower sensitivity, and extra delay times
as experienced with the electromechanical designs.


3. Distance Relays

In general, the reach oI distance relays will be aIIected by CT saturation. This is true
whether the relay design is an electromechanical design utilizing rms current
measurement, or a solid state design reIerencing peak instantaneous or rms current
levels or various combinations oI waveIorm values Ior measurement. The distance
relay will be desensitized Ior Iaults near the end oI its zone, sensing the Iault as being
Iurther away than it actually is, based upon the reduced currents produced Irom the
CT secondary circuits. Short line and particularly zone 1 instantaneous relay
applications with a Iault value near the relay operating decision point should be
considered careIully iI CT saturation is possible. A guideline reIerred to by G.D.
RockeIeller Irom Consolidated Edison Co. oI New York is: "To avoid delayed
tripping Ior Iaults near the zone 1 decision point, time to (CT) saturation should
exceed 1.5 cycles". Solid state relays with their high speed operation and low burden
may be preIerred in these applications, depending on the individual situations and
systems involved.

The directional characteristics oI distance relays are generally adequate and selective
Ior reverse direction Iaults even with phase shiIts in the current waveIorm
measurement. However, closein high magnitude Iaults in the reverse direction may
result in Ialse tripping by directional relays. This may occur iI the reIerence
stabilizing voltage as measured by the relay is below a design minimum value in the
presence oI zerocrossing phase shiIt error in the current waveIorm caused by
saturated CT's. Electromechanical relays using induction units produce an operating
torque based upon the product oI an integrated Iull cycle oI voltage and the
Iundamental current waveIorm. This reduces the possibility oI distortion related
Ialse tripping.

In summary, current transIormer saturation eIIects should be considered when
applying directional overcurrent and distance relays. For relay directional
elements, CT saturation may cause Ialse directional interpretation created by
phase shiIts between the relative measurement points on the distorted current
and undistorted voltage waveIorms (or possibly distorted voltage waveIorms iI
CCVT's are used). Unpredictable or added time delay and lower sensitivity may
be experienced with overcurrent elements. Inaccurate distance measurements,
maximum torque angle characteristic changes, etc., may lead to Ialse operations
in directional overcurrent and distance relays. The best way to avoid these
potential problems is to take precautionary steps through minimizing CT
secondary burden, by using larger conductor sizes in CT leads, and using lower
burden relays when possible combined with using adequate accuracy class CTs.
The transient response characteristics oI the CT's should be considered (e.g.
time to saturation) when deciding on electromechanical or solid state relay
Page 24 oI 48
applications. With proper planning measures and consideration oI the
aIorementioned Iactors, under the great maiority oI Iault conditions correct relay
operation should result.

EHV Transmission Line Systems
The eIIects on these relay systems due to CT saturation can be most serious because oI
their operating speed, the conIiguration oI current transIormer sources and the need Ior
exactly the same perIormance at the location oI the relaying systems.

The relay operating speed is typically 825 milliseconds. In many cases these may
have operated, beIore any saturation eIIects take place. However, on an external
Iault, although saturation is less likely, the system will be dependent upon correct
CT operation throughout the Iault period.

Most phase comparison systems are designed to accept signiIicant phase angle errors
without undue eIIects. Amore likely source oI problem is distortion oI the magnitude oI
phase quantities and thereby producing incorrect components on which proper operation
depends primarily internal Iaults.

Directional comparison systems will have perIormance problems similar to standalone
directional instantaneous elements. One diIIerence is that instead oI iust one or two
CTs aIIecting the devices perIormance the CTs at both ends are involved so there may
be as Iew as 2 or as many as 6 CTs involved.

The conIiguration oI CTs is an important consideration in the operation oI these schemes
since the primary currents in each CT and each CTs history determines what inIormation
is supplied to the relay. This exposes the relay system to possible problems Irom any oI
the connected CTs. In many cases the EHV system was designed Ior short circuit duty
that has not and may not ever be experienced. This means that it may be many years into
the liIe oI the EHV system beIore trouble is experienced with CT saturation. However,
the mismatched CTs may produce problems long beIore the problems oI saturation are
evident.
Page 25 oI 48
V. EXAMPLES OF CALCULATIONS
A. Ratio and Phase Angle Error oI CT:
The Ratio Correction Factor (RCF) is the Iactor by which the marked ratio must be
multiplied to obtain the true ratio. The true ratio equals the marked ratio times the
ratio correction Iactor or,
ratio marked
ratio true
RCF

Where:
true ratio the ratio oI rms primary current to the rms secondary current,
S P
I I
marked ratio the ratio oI the rated primary current to the rated secondary current
as given on the nameplate

The phaseangle oI a CT is designated by the Greek letter . . The PhaseAngle
Correction Factor (PACF) is the Iactor by which the reading oI a watt meter,
operated Irom the secondary oI a CT must be multiplied to correct Ior the eIIect oI
phase displacement oI current due to the measuring apparatus. It is the ratio oI the
true power Iactor to the measured power Iactor.

. 7 7 cos cos PCF

Where:
7 angle oI lag oI load current behind load voltage

7 cos power Iactor cosine oI the angle between the voltage and current.

The Iactor by which the reading oI a watt meter or the registration oI a watt hour
meter must be multiplied to correct Ior the eIIect oI ratio error and phase angle is the
TransIormer Correction Factor (TCF).

PCF RCF TCF L
Page 26 oI 48
Example 1:
II a CT has a phase angle error 15' and is used Ior measuring a load whose power
Iactor is 0.500 lagging, determine its phase angle correction Iactor, PACF.

1. The primary current lags the line voltage by an angle whose cosine equals the power
Iactor.
7 H

60 500 . 0 cos
1

or 500 . 0 60 cos cos H 7

2. The secondary current leads the primary current by 15'. ThereIore, the primary
current actually lags the primary voltage by ' 45 59H .

' 60 59 60 H H 7

' 45 59 ' 15 0 ' 60 59 H H H . 7

thus, 5038 . 0 ' 45 59 cos cos H . 7
9925 . 0
5038 . 0
500 . 0
) cos(
cos

. 7
7
PCF


II the CT has a RCF oI 1.0020, what is the TCF at the same power Iactor?

9945 . 0
9925 . 0 0020 . 1

L
L PCF RCF TCF



Example 2:
Calculation oI the ratio oI relaying CT

Consider a bushing CT with the Iollowing characteristics:

0.298 resistance secondary CT

C100 ation classiIiIc relaying
5 600 ratio


1. The rating oI C100 means that the ratio is to be calculated on the basis oI 100 volts at
the secondary terminals with 100 amp Ilowing through the burden which, in turn,
means a burden oI B1. ReIerring to Table 1 Ior a B1 burden:
inductance mH 2.3 L
resistance 0.5 R




Page 27 oI 48
2. The inductive reactance,
L
X , is then calculated:


L L L L

866 . 0
10 3 . 2 60 14 . 3 2
2
3

1L X
L
6

Where I Irequency in Hz

3. To obtain the induced or excitation voltage
SE
E , the resistance oI the secondary
winding must be added to the impedance, Z, oI the burden.

2 2
X R Z


2 2
0.866 0.5 298 . 0 Z

18 . 1

The induced voltage is thereIore:
J amps E
SE
118 18 . 1 100 L


ReIerring to Figure 10, the excitation curve Ior this transIormer indicates that an
excitation current, I
e
, oI 0.15 amps is required to produce this voltage.

Since this excitation current is shown as the equivalent secondary current, it should
be compared to the burden secondary current oI 100 amps.

The ratio oI exciting current to burden current is
0015 . 0 100 15 . 0

ThereIore the Ratio Correction Factor (RCF) is 1.0015.




Page 28 oI 48

Table 1
Standard Burdens Ior Current TransIormers
with 5 amp secondaries


Burden
Designation
Resistance
(O)
Inductance
(mH)
Impedance
(O)
Volt
Amperes
(at 5A)
Power Factor
Metering Burdens
B0.1 0.09 0.116 0.1 2.5 0.9
B0.2 0.18 0.232 0.2 5.0 0.9
B0.5 0.45 0.580 0.5 12.5 0.9
B0.9 0.81 1.04 0.9 22.5 0.9
B1.8 1.62 2.08 1.8 45.0 0.9
Relaying Burdens
B1 0.5 2.3 1.0 25 0.5
B2 1.0 4.6 2.0 50 0.5
B4 2.0 9.2 4.0 100 0.5
B8 4.0 18.4 8.0 200 0.5





Page 29 oI 48

igure 9
EXCITATION CURVES OR MULTI-RATION BUSHING CT WITH ANSI
CLASSIICATION O C100

Page 30 oI 48
B. Estimation oI Current TransIormer PerIormance:
A current transIormer's perIormance is measured by its ability to reproduce the
primary current in terms oI the secondary; in particular, by the highest secondary
voltage the transIormer can produce without saturation. CT perIormance can be
estimated by:

The CT excitation curves.
The ANSI transIormer relaying accuracy classes.

These methods require determining the secondary voltage that must be generated.

B LEAD L L S
Z Z Z I V

Where:
S
V The rms symmetrical secondary induced voltage.
L
I The maximum secondary current in amps
(this can be estimated by dividing the known maximum Iault current by the
selected CT ratio)

L
Z
The secondary winding impedance.

LEAD
Z
The connecting lead burden.

B
Z
The connected external impedance


1. Excitation Curve Method

The excitation curve oI Figure 10 can be used to determine the excitation current
required by the CT Ior a particular turns ratio, primary current and secondary
burden parameters.

Procedures:

a. determine nominal secondary current Irom primary current and desired turns
ratio:
N P L
I I I

b. determine required secondary voltage Irom

LE L L S
Z Z Z I J

c. determine secondary excitation current Irom Figure 10

d. determine approximate burden current by arithmetic subtraction oI excitation
current Irom nominal secondary current

Page 31 oI 48
Example 1:
Given a CT with excitation characteristics as shown in Figure 10

A burden oI relays and instruments oI 0.15 O instruments and overcurrent
relays with a burden oI 0.3 O on Tap 5.

Secondary lead resistance including CT winding oI 0.15 O.

For simplicity, these impedances are assumed to be at the same angle.


LE L
Z Z Z 0.15 0.15 0.3 0.6 O.

The primary Iault current expected is 12,000 amperes.
The desired CT ratio is 400:5, or 80:1.
Then I
L
12,000/80 150 amps
V
S
150 (0.6) 90V
Irom Figure 10, I
e
18.0 amp.

Then 150 18 132 amps oI burden current, Ior an eIIective ratio oI 12,000/132
455:5.

This means the perIormance is not within the intended accuracy oI a 10C100 .
(See Section C.)

II a 500:5 or 100:1 ratio is selected then:
I
L
12,000/100 120 amps
V
S
120 (.6) 72V
I
e
0.105 amps Irom Figure 10

Burden current 120 0.1 119.9 Ior an eIIective ratio oI 12,000/119.9
500.4:5; well within the intended accuracy.


Example 2:
With the results oI Example 1, determine the primary operating current Ior a
residual relay oI burden 4.5 ohms on Tap 0.5.

See the sketch oI Figure 11.

At pick up, there will be 2.25 volts across the residual relay, (0.5A x 4.5). This
voltage also appears across the CT on the unIaulted phases. From Figure 10
each oI these CTs will require 0.017A exciting current.
The current Irom the CT on the Iaulted phase supplies 0.5 0.017 0 .017 or
0.534 amps as shown in Figure 11.

This 0.534 amps develops 2.57V across the secondary oI the CT on the Iaulted
phase. This requires an additional 0.018 amps oI exciting current.

Page 32 oI 48
The total secondary current supplied by the CT on the Iaulted phase is then
0.5 0.017 0.017 0.018 0.552 amps.







igure 11
SECONDARY CURRENT DISTRIBUTION AT PICKUP O RESIDUAL RELAY

II the CT ratio is 100:1, then the primary current must be 55.2A at pick up oI
the residual relay.

Page 33 oI 48
C. ANSI CT Relaying Accuracy Classes:

The ANSI Relaying Accuracy Class is described by two symbols letter designation
and voltage rating that deIine the capability oI the transIormer. The letter designation
code is as Iollows:

C indicates that the transIormer ratio can be calculated (as Ior the earlier lOL type
transIormers).

T indicates that the transIormer ratio must be determined by test (similar to the earlier
10H type transIormers).

The secondary terminal voltage rating is the voltage that the transIormer will deliver to
a standard burden at 20 times normal secondary current, without exceeding 10 percent
ratio correction. Furthermore, the ratio correction must be limited to 10 percent at any
current Irom 1 to 20 times rated secondary current at the standard burden. For example,
relay accuracy class C100 means that the ratio can be calculated and that the ratio
correction will not exceed 10 percent at any current Irom 1 to 20 times rated secondary
current with a standard 1.0 burden (1.0 times 5A times 20 times rated secondary current
equals 100V).

ANSI accuracy class ratings apply only to the Iull winding. Where there is a tapped
secondary, a proportionately lower voltage rating exists on the taps.

Example:
The maximum calculated Iault current Ior a particular line is 12,000 amps. The current
transIormer is rated at 1200:5 and is to be used on the 800:5 tap. Its relaying accuracy
class is C200 (Iull rated winding); secondary resistance is 0.2 ohm. The total
secondary circuit burden is 2.4 ohm at 60percent power Iactor. Excluding the eIIects
oI residual magnetism and DC oIIset, will the error exceed 10 percent? II so, what
corrective action can be taken to reduce the error to 10 percent or less.

The current transIormer secondary winding resistance may be ignored because the
C200 relaying accuracy class designation indicates that the current transIormer can
support 200 volts plus the voltage drop caused by secondary resistance at 20 times rated
current, Ior 50 percent powerIactor burden. The CT secondary voltage drop may be
ignored then iI the secondary current does not exceed 100 amps.

N 800/5 160
I
L
12,000A/160 75 amps

Page 34 oI 48
The permissible burden is given by:
Z
B
(N
P
V
CL
)/100

Where:
Z
B
permissible burden on the current transIormer
N
P
turns in use divided by total turns
V
CL
current transIormer voltage class

N
P
800/1200 0.667 (proportion oI total turns in use)

Thus, Z
B
0.667 (200)/100 1.334 O

Since the circuit burden, 2.4 ohms, is greater than the calculated permissible burden,
1.334, the error will be in excess oI 10 percent at all currents Irom 5 to 100 amps.
Consequently, it is necessary to reduce the burden, use a higher current transIormer ratio,
or use a current transIormer with a higher voltage class.

D. EIIect oI Current TransIormer Connections on Burden
Whenever two or more current transIormers have their secondary circuits
interconnected, the eIIect is to alter the secondary burden on each transIormer. The
way in which the burden is aIIected is dependent upon the particular connection.

Various types oI connections are described in the Iollowing discussion.

SeriesConnection.
When current transIormers are connected with their secondaries in series, the general
eIIect is to decrease the burden on each individual transIormer. This statement
assumes that the secondary currents are nominally in phase and oI equal magnitude.

II N transIormers with identical excitation characteristics are connected in series and
are supplying current to a burden Z, then the burden on each transIormer equals Z/N.

ParallelConnection.
When two or more current transIormers are connected in parallel, the general eIIect is
to increase the burden on each individual transIormer. The amount oI increase is
dependent upon the type oI connection, the number oI transIormers, and the
distribution oI current between transIormers.

When low ratios are required, standard CTs may not be available to supply the
required burden. It is sometimes possible to apply 2 standard higher ratio CTs having
a higher relaying accuracy classiIication voltage (with the primaries connected in
series and the secondaries connected in parallel) to supply this burden. The desired
overall low ratio is achieved with a substantially improved accuracy. In paralleling the
secondary circuits oI CTs the secondary winding shall be paralleled at the meter to
keep the common burden as low as possible. The eIIective burden on each
transIormer should not exceed its rated burden.

Page 35 oI 48
The Iollowing diagrams show some oI the more common ways in which current
transIormers are interconnected. General equations are given Ior the burden on a
typical current transIormer. In applying these equations, it should be noted that all
impedances in series, including lead resistance, have been lumped into one value to
simpliIy the equations.

Parallel connection



General case:

1
2 1
1
....
1
I
I I I
Z Z Z
N
C CT


Special cases:
1. Two CTs in parallel, equal contributions Irom each CT,
2 1
I I :

C CT
Z Z Z 2
1
1


2. Two CTs, unequal contributions:
K I I
1 2
; K Z Z Z
C CT
1
1
1





Page 36 oI 48
Delta connection



General case:

1
3
1
2
1
1 1
1
I
I
Z
I
I
Z Z Z
C CT



Special cases:

1. Symmetrical burdens:

1
3 2
1
2
1
I
I I
Z Z Z Z Z Z Z Z
L L CT L C



2. Symmetrical burdens and balanced 3 phase currents:

L CT
Z Z Z 3
1
1










Page 37 oI 48
Wye connection




General case:

1
3 2 1
1
1
I
I I I
Z Z Z
N CT



Special cases:

1. three phase Iault:

1
1
Z Z
CT


2. phasephase Iault (12):

1
1
Z Z
CT

3. lineground Iault:

N CT
Z Z Z
1
1


4. two lineground (12G):

1
2 1
1
1
I
I I
Z Z Z
N CT





38

TransIormer DiIIerential Connection

When used Ior transIormer diIIerential relaying, current transIormers can be connected in several
diIIerent way depending on the type oI transIormer being protected. As a general rule, relay
misoperation due to high burden is a problem only Ior external Iaults where Ialse tripping may
occur iI one current transIormer saturates.

WYE-WYE CT connection (delta-delta transformer)



General case:

I
I
Z Z Z
X
CT
1
1
0
1
1 1 1



Special Cases:

1. External Iault with ideally matched CT ratios:

1 1
CT
Z Z

2. Internal Iault with equal perunit contributions Irom each side oI the transIormer:

1 1 1
0
2Z Z Z
CT




Page 39 oI 48

WYE-DELTA connection




General case:

X
I I
I I
Z Z Z Z
X
CT
3 1
1 3
0 0
1 1 1 1

X
X
X
X
X
X
X
X
I
I
Z
I
I I I
Z
I
I I I
Z
I
I
Z Z Z Z Z Z Z
X

X X X CT
1
2
1
2 1 2
0
1
3 3 1
0
1
3
0 0 4
2 2 1 1 2 2 1 1 4


Special Cases assuming matched CTs and symmetrical burdens

1. External three phase Iault:

1 4 1 1
3
4 X CT CT
Z Z Z Z Z

2. Internal three phase Iault with equal perunit contributions Irom each side oI the transIormer:

1 1 4 1 1 1
0 4 0
6 3 2 Z Z Z Z Z Z Z
X CT CT












Page 40 oI 48
DELTA-DELTA CT connection (WYE-WYE transformer)







General case:
) ) )
3 1 3 1 3 3 1 1 3 1 3 1 1
0
1
3
0
1
2
0 0
1
1
0
1
3
0
1
2
0 0 1
Z
I
I
Z
I
I
Z Z
I
I
Z Z
I
I
Z Z
I
I
Z Z Z Z Z Z

CT



Special cases assuming matched CTs and symmetrical burdens:

1. External lineground Iault on phase 1

1 1
2
1 CT
Z Z Z

2. External three phase or line to line (12) Iault

1 1
3
1 CT
Z Z Z

3. Internal line to ground on phase 1 with equal perunit contributions Irom each side oI the
transIormer

1 1 1
0 1
4 2 Z Z Z Z
CT





Page 41 oI 48
WYE-DELTA-WYE connection

General case:
)
2 2 2 1 1 1 2 1 1
2 0 0 3 0 0
1
1
1
CT
Z I Z I Z I Z I
I
Z Z Z Z

)
X X

I I I I I I
I I
Z
Z Z
CT 1 3 1 3 3 1
1 3
0
1
1 4

Special case:
External three phase Iault with symmetrical burden and ideally matched CTs


1 1
1 CT
Z Z Z

1 4
CT
Z Z

Page 42 oI 48
E. Estimates oI Transient PerIormance
Example 1
CT turns N
2
240
CT saturation voltage V
X
850 volts
CT primary current I
1
15,000 amps
CT sec. burden R
2
1 O
Primary system X
1
/R
1
15

Ks Then
2 2 1
N R I
Jx
6 . 13
240 000 , 15
850

X
J or

'
+

'

1
377
) 15 ( 377
) 1 (
240
000 , 15 ) 1 (
2
1 2 1
N
T R I


;olts 000 , 1


Comparing the calculated V
X
(1000) with the available V
X
(850) means we can expect
some saturation under Iully oIIset conditions.

The time to reach saturation is given by
S
t :

'
+

'

1
1
1
1 ln
T
K
T t
s
S


04 . 0 377 1 15
1 1 1
R X T

'
+

'

377
1 6 . 13 04 . 0 1
1 ln 04 . 0
S
t Then
.0722 seconds
4.33 cycles

Note that the time to saturation is sensitive to the primary system time constant and the
saturation Iactor K
S
. II the oIIset is less than Iull, then K
S
is greater and the time to
saturation is long or saturation may not occur.

The X/R ratio determines the possible dc oIIset and the system time constant; thereIore,
K
S
can be increased Irom the value calculated to that corresponding to the X/R ratio.


Page 43 oI 48
In this case the X/R ratio is 15 and Irom the Iollowing table:

X/R I`/I
2 1.18
4 1.38
7 1.52
10 1.54
20 1.68
50 1.73
100 1.75

The oIIset could be 1.64/1.75 or .94 per unit oI the maximum possible.

and K Then
S
47 . 14 94 . 0 6 . 13
0845 . 0
S
t seconds instead oI 0.0722 seconds

The time to leave saturation is given by:

S e
K T T t
2 1
ln L
sec 0 . 2
2
T
6 . 13
S
K
sec 161 . 0 6 . 13 ) 2 ( 377 ln 04 . 0
e
t
cvcles 66 . 9






Page 44 oI 48

VI. REFERENCES

1. NEMA SG4, Table 34.

2. The Art & Science oI Protective Relaying, C. Russell Mason, John Wiley & Sons.

3. Industrial and Commercial Power System Applications Series Relay Current
TransIormer Application Guide (a publication by Westinghouse RelayInstrument
Division PRSC6, May 1982).

4. American National Standard Guide Ior Protective Relay Applications to Power
System Buses.

5. Applied Protective Relaying, Westinghouse Electric Corp.

6. Protective Relays, Their Theory & Practices, A. R. Van C. Warrington, Chapman &
Hall, London.

7. A HalICycle Bus DiIIerential Relay and Its Application, T. ForIord, J. R. Linders,
IEEE T74 0337, pp. 11101120.

8. a. IEEE Report No. 76CH11304 PWR, "Transient Response oI Current
TransIormers".

b. "Protective Current TransIormers and Circuits", P. Mathews. The MacMillan Co.,
New York, 1955.

9. ANSI C57.13.

10. "Current TransIormer Burden and Saturation", Louie J. Powell, Jr., Senior Member,
IEEE, IEEE Transactions on Industry Applications, Vol. 1A15, No. 3.

11. "Static Relaying Measuring Techniques Which Optimize the Use oI Available
InIormation", by A.T. Giuliante (ASEA), John Linders (consultant), L. Matele
(ASEA), presented at Western Protective Relay ConIerence, October 1618, 1979,
Spokane Washington.

12. "Relaying CT's A source oI Vital InIormation and MisinIormation", G.D. RockeIeller
(System Protection Engineer) Consolidated Edison Co. oI N.Y., Inc., presented to 1973
ConIerence on Protective Relaying, Georgia Institute oI Technology, Atlanta, Georgia.

13. "Bus DiIIerential Protection", General Electric Company, GET6455.

14. "Transient Response oI Current TransIormers; Publication 76CH11304 PWR (IEEE
Power Systems Relaying Committee Report).

15. ANSI/IEEE C57.13.11981, Guide Ior Field Testing oI Relaying Current TransIormers

Page 45 oI 48
16. ANSI/IEEE C57.13.31983, Guide Ior the Grounding oI Instrument TransIormer
Secondary Circuits and Cases
VII. LIST OF OPERATIONS COMMITTEE PUBLICATIONS

The Iollowing is a list oI additional Operations Committee Publications that are available
upon request Irom:

Western Systems Coordinating Council 540
Arapeen Drive, Suite 203
Salt Lake City, Utah, 84108
(801) 5820353

1. System Protection Guides and Test Procedures (July 1985)
(Consolidation oI the Iollowing publications:
Guide Ior Model Testing oI Relay Systems
System Protection Guides
UnderIrequency Load Shedding Relay Application Guide
Test Procedure Ior Power System Stabilizers)

2. Operating Procedures Compiled in accordance with Article V, Section 3 oI the
WECC Agreement (March 28, 1990)

3. Listing oI Spare TransIormers Owned by WECC Members and AIIiliate Members
(January 1, 1989)

4. DispatcherSystem Operator Training Manual (June 30, 1986)

5. Dispatcher/System Operator Handbook (July 1986)

6. Communication System Operator's Guide (November 1988)

7. Energy Management System InterUtility Communication Guidelines (October
1989)

8. Operations Committee Balloon Diagrams and Inadvertent Interchange Summaries
(Compiled twice weekly and distributed to Operations Committee members)

9. Guide Ior Development and Testing oI Remedial Action Stability Schemes (May
1986)

10. Relaying Current TransIormer Application Guide (June 1989)

11. Guide Ior SpeciIication oI a Digital Fault Recording System (June 1988)

12. Guidelines Ior Synchronization oI Oscillographs and Event Recorders
(January 1990)