1

Substation Testing and Commissioning: Power Transformer Through Fault Test

M. Talebi∗ , Member, IEEE,
Power Grid Engineering
Y. Unludag†
Electric Power System

Abstract

This paper reviews the advantage of performing transformer through-fault test in substation testing and commissioning prior to the
actual energization in addition of theory of testing and numerical procedure. This test allows the engineer and relay technicians to
validate all the CTs and PTs secondaries, transformer winding configuration, protective relays and wiring correctness and circuit.
The value and benefit of performing this test is also shown in real world project.

Index Terms
Transformer, Through-Fault, Substation, Commission and Testing.

I. I NTRODUCTION

Substation commissioning and testing is a critical task prior to substation energization to ensure all equipment will function
safely in the intended manner. Medium and large power transformers in distribution or transmission power system are often
protected by differential relay because of its high sensitivity and fast operation. The principal and design of transformer
differential relaying has been covered and described in many literature and manuals and it is not the scope of this paper
[1]–[4]. After design and installation of transformer differential relay, it is essential to prove and validate its circuit prior to
substation energization. An incorrect wiring of the differential relaying will cause false operation or no operation of protective
relays during normal or fault event. Several incidents have been reported by the U.S power utility system regarding improper
wiring and circuit of CTs in power transformer protective relaying which has resulted in transformer protection mis-operations
with short and long-term impact and consequences [2]. It is still a common practice in the U.S. utility system to defeat the
transformer differential relay function before connecting it to the load and performing the in-service test to verify its circuit
after the actual substation energization. It is clear that validating all protection circuits and discovering all potential errors prior
to substation energization, without disabling the relay function, is highly desirable. As such, in this paper, a detailed procedure
of the through-fault test for power transformer differential circuit prior to substation energization using the practical examples
is covered. This practical commissioning test was developed decades ago and the authors are simply presenting the theory and
procedure of how the test performs in the substation.

The reminder of the paper is organized as follows: Section II reviews the through-fault test application and advantages. The test
and numerical calculation procedure is presented in section III. Section IV discusses performance and results of through-fault
test with real world example and section V concludes the paper.

II. P RELIMINARY

In the power transformer protection, through-fault is the system fault that are external to the transformer protection zone. It is
a well known concept that the zone of the protection in transformer differential relay will be defined by the location of the
CTs secondaries. Through-fault test apply the balance three phase voltage, equal voltage magnitude and displaced 120 degree
among phases, on one side of the transformer or circuit breaker. Since the circuit is balance, transformer positive sequence
impedance will be in series with positive sequence voltage and current of test source and circuit will complete its current
loop by shorting and grounding all bushings outside the differential relay zone of protection. Through-fault test will prove
the CTs polarity are correct and the expected direction of secondary current flow is correct for a given direction of primary
current flow. It will also simply confirm the transformer CTs ratio by proving that the ratio, as installed, is as specified and
listed in the drawing and transformer’s nameplate, and if taps are available, that they also have the correct ratio and have been
wired to the correct terminals. Furthermore, it is a useful test for verifying power transformer winding connections and phase
displacement corresponding to its vector group on the nameplate in the field. Depending on phasing connection, this phase
shift may be either plus or minus 30 degree from primary to secondary for positive sequence voltages and currents. It will
also confirm protective relaying wiring correctness. Potential transformers secondaries can also be checked during this test.
M. Talebi is a Filed Service Engineer with Power Grid Engineering, Orlando, FL.
Email: [email protected] (corresponding author)
 2

It should be noted that all other protective relays in transformer protection design scheme such as 50/51 can also be tested
during through-fault testing.

III. T EST P ROCEDURE AND N UMERICAL C ALCULATION

Prior to the test, transformer differential protection zone should be identified based on the relay functional drawing. Expected
test current value and phasor diagram should be computed before running the through-fault test. The test source voltage,
transformer and CTs winding connections, transformer impedance, transformer power capacity and its tap position under the
test, system phase rotation, phase-to-bushing connections, and CTs and PTs ratios are required for the calculation procedure.
It should be noted that the size and output voltage of temporary generator required in the filed as a test source voltage will
be determined after calculating the current flow through the circuit. Based on the microprocessor protective relays manual,
a minimum secondary current of 0.25 amps is required to pass the pickup and accuracy range of the digital relays. Hence,
sometimes transformer size and its impedance make it difficult to choose the temporary generator used for through-fault test
in the field to push the minimum required current. Practically, the common temporary generator output voltage used in the
field is at a maximum of 480-500 VAC with 125 KVA.

A. Calculation Procedure

Test source voltage will be in series with positive sequence impedance of the transformer; therefore, eq. 1 is for calculating
the p.u. value of the injecting current by test source [5], [6],
VGp.u.
IGp.u. = , (1)
Zxf mrp.u.

where IGp.u. , VGp.u. , and Zxf mrp.u. are p.u. the values of the test source current, test source voltage, and transformer impedance
respectively.
For calculating the VGp.u. , simply use the p.u. equation of,
VG
VGp.u. = (2)
VBase
,

where VG is the output voltage of the test source voltage and VBase is the voltage of the transformer side (usually the low
side) the test source is connected to.

Now, by applying the three phase power equation, we can calculate the transformer low and high side Base currents,
MV A
IHBase = √ (3)
3VHxf mr
MV A
ILBase = √ , (4)
3VLxf mr

where VHxf mr , and VLxf mr are the high and low side of the transformer voltage in KV and M V A is the power capacity of
the transformer on the tap under the test.

Finally, by applying eq. 5 and 6, we can calculate the current flow through the transformer primary and secondary side
respectively,

IHxf mr = IGp.u. × IHBase , (5)

ILxf mr = IGp.u. × ILBase . (6)

The magnitude of the current in the CTs secondaries can be calculated by dividing current magnitudes calculated in eq. 5 and
6 to CTs ratios.
 3

Fig. 1. Simplified Single-Line Diagram.

TABLE I
C URRENT M AGNITUDE AND P HASE A NGLE OF CT S FROM F IG . 1

Current P haseA P haseB P haseC

IG IG ]270 IG ]150 IGp.u. ]30
IG IG IG
ICT 1 CT1 R ]270 CT1 R ]150 CT1 R ]30
IG IG IG
ICT 2 CT2 R ]120 CT2 R ]360 CT2 R ]240
ICT 3 0]0
IG IG IG
ICT 4 CT4 R ]300 CT4 R ]180 CT4 R ]60
IG IG IG
ICT 5 CT5 R ]120 CT5 R ]360 CT5 R ]240

For measuring the phase angle in the CT secondary, a known phase voltage (usually phase A with zero degree phase angle
of test source) should be chosen as a reference. The inductive nature of power transformer in addition to phase displacement
between transformer’s winding has to be considered for CTs secondaries phase angle measurement. Fig. 1 shows a basic
substation one-line diagram. For simplicity, lets just assume that transformer differential relay zone is from CT 1 to CT 2. Test
source is connected to the low side of the transformer and as shown, bushing on the line side of the circuit breaker has been
shorted and grounded. Phase A current out of test source lags 90 degrees the reference voltage, VA ]0, and measures Ia ]270.
Looking to Fig. 1, this current will enter to the polarity side of CT 1 in primary side of power transformer and will have
same phase angle with lower current magnitude depending to CT 1 ratio. On the secondary side of the transformer, phase A
current enters to non-polarity side of CT 2 and it should measure IA ]120 in Wye connected configuration CT secondary seen
in microprocessor differential relay (transformer 30 degrees phase shift plus 180 degrees is experienced from the mirrored
polarity of both winding CTs). It is essential to know how the phase angle meter display has been programmed i.e. if it is
in leading or lagging operation state. Reading above is for the meter programmed in leading operation state which means the
value displayed is the number of degrees by which the current leads the voltage. If the meter has been programmed in lagging
operation state, the meter will display Ia ]90 and IA ]300 for CT 1 and CT 2 respectively.

Fig. 2 shows the expected phasor diagarm of voltage and current for standard Delta-Wye transformer differential circuit. By
selecting phase A voltage, VA ]0, as a reference, phasor diagram of voltage and current of test source connected to Wye side
should be expected as Fig. 2(a). Remember that the power transformer is considered as an inductive load and current lags
voltage by 90 degree. In practical power transformer there is always resistive component so there will be 3−5% difference from
the expected value. Fig. 2(b) shows the phase angle relationship between CTs secondaries on both sides of the transformer,
CT 1 and CT 2. Again, it should be noted that the angles are measured by using temporary generator phase A voltage as a
reference. The magnitude of phasors shown in the Fig. 2 is arbitrary and purpose is just showing the phase angle relationship.
Table I summarized the expected values for CTs magnitude and phase angle corresponding to Fig. 1 with phase angle meter
in leading operation state.

B. Testing Procedure
Here are the steps should be taken in the field for performing the through-fault test:
1) Prior to performing the through-fault test, it is important that all safety procedures regarding the switching, locking, and
tagging be observed and formalized to avoid the protection engineers and technicians from being exposed to substation
equipment that has not been properly de-energized.
2) Location of connecting the test source in one side and shorting and grounding of bushing on the other side of protection
zone based on the relaying functional diagram and substation layout has to be identified.
3) Primary conductors and phasing needs to be visually proven and identified to have a correct phasing prior to the test.
4) Temporary generator phase rotation needs to be verified before the test.
 4

(a) Phase angle relationship between test

source voltage and current in primary side
of Transformer

(b) Phase angle relationship between cur-

rent in CT 1 and CT 2 seen by protective
relay

Fig. 2. Expected Phasor Diagram of Fig. 1.

5) Generator current magnitudes need to be known and a phase reference needs to be established in order to analyze the
reading. Current can be measured using a clamp-on ammeter and compared with the first CT calculation in calculation
sheet. Phase reference is chosen to prove correct phasing and polarity of all CT circuits to be tested and it can be phase
A to ground voltage of the temporary generator. All the phase angle readings during the test will be based on the chosen
reference. Therefore, engineer or technician needs to have extended test leads to go from reference point to all measuring
points.

IV. P RACTICAL E XAMPLE C ASE S TUDY

A real world project with expected and measured values under the test is covered and illustrated in this section to show the
effectiveness of the testing method.

A. Solar Project

Fig. 3 shows the modified single line diagram of a 50 MW solar project. The GSU transformer nameplate information were
as follows: 3 phase, 24/32/40M V A, 115/34.5KV Delta-Wye two winding configuration. The impedance of the transformer
under 24M V A was measured as %7.33 and transformer was under 115 KV tap. The transformer differential relay zone of
protection starts from CTs on the line side of breakers 52-1 and 52-2, CT 4 (600/5 tap) and CT 3 (600/5 tap) and goes to line
side of transformer breaker, 52-21, CT 1 (300/5 tap). CT1 and CT2 are 115KV line side CTs of main circuit breaker, CT3
and CT4 are 34.5KV line side CTs of feeder breakers.

Through-fault test was performed at solar substation by connecting a 3-phase 500V generator to the line side disconnects of
each 34.5 KV feeder and applying shorts on the line side disconnects of the 115 KV circuit breaker. Prior to the test, expected
values for current calculated as follows by applying equations (1)-(6):
500
VGp.u. = = 0.0144 (7)
34.5KV
 5

Fig. 3. Modified single line diagram of solar project.

0.0144
IGp.u. = = 0.197 (8)
0.0733
24M V A
IHBase = √ = 120.49A (9)
3115KV
24M V A
ILBase = √ = 401.64A (10)
334.5KV
IHxf mr = 0.197 × 120.49 = 23.73A (11)
ILxf mr = 0.197 × 401.64 = 79.41A. (12)

Applying CT 4 and CT 3 ratio of 600/5, current magnitude in the secondary is 0.66A. For the PT located in 34.5KV bus,
P T 1, with 19900/120 ratio, the expected voltage magnitude in PT secondary can be calculated as
288
VP T = = 1.74V (13)
165.8

After calculating expected values, voltage and current readings were taken on each CT and PT circuit to validate CT and PT
wiring and relay settings. Table II shows the expected values vs. measured values of CTs circuits. The phase angle meter was
in lagging operating state. During final walk through before the testing started, it was discovered that several CT shorting
screws were not removed and CT tap ratios did not match the single line. After these issues were corrected, the testing started.
During the test it was discovered that A and C phase wiring on all CT and PT circuits were rolled and CT connections shown
on the single line were connected to the wrong relay windings. As shown in the table II, the expected value of phase A
current was 0.66]90 but the measured value was 0.66]327 which was the expected value of C phase. It was also discovered
that the differential relay CT connection compensation settings were set wrong. Winding 2 CT, CT 4 and CT 3, connection
compensation setting was set as 1 where it should have been 11 to compensate 30 degrees phase shifting from low to high
side, meaning Wye side leading Delta side by 30 degree.

TABLE II
M AGNITUDE AND P HASE ANGLE OF C URRENTS OF S OLAR FARM P ROJECT- E XPECTED V.S. M EASURED

Expected Value Measured Value

CT S
Aφ Bφ Cφ Aφ Bφ Cφ
CT1 0.397]300 0.397]60 0.397]180 0.385]177 0.388]58 0.388]297
CT2 0.397]300 0.397]60 0.397]180 0.383]176 0.388]58 0.389]297
CT3 0.662]90 0.662]210 0.662]330 0.66]328 0.65]208 0.67]88
CT4 0.662]90 0.662]210 0.662]330 0.66]327 0.65]207 0.67]87
 6

After this discovery, wiring for CTs and PTs were corrected and tested again to confirm the correctness of the circuit.

V. C ONCLUSION

In this paper, the theory, application, and advantages of transformer through-fault test in substation commissioning was covered.
Through-fault test employs a temporary generator as a three phase test source to inject three phase balance currents into the
power transformer windings and its CTs; therefore, allowing the engineer and relay technicians to measure current and voltage
magnitude and phase angle in the CTs and PTs secondary and protective relays. The test will check the integrity and correctness
of the circuit and will discover incorrect wiring, polarity, and setting prior to actual energization. This gives enough time to
relay technicians and engineers to troubleshoot the circuit and take necessary corrective actions.

VI. ACKNOWLEDGMENT

The authors would like to thank Duke C. and M. engineering and technicians team for their inputs, specifically Rick Jonell.

VII. R EFERENCES
R EFERENCES
[1] C. L. Headley, “Differential Relay Sensitivity Adjuster,” IEEE Trans. American Inst. of Electrical Engi., vol. 69, pp. 1203-1206, Jan. 1950.
[2] D. Costello, “Lessons learned through commissioning and analyzing data from transformer differential installations,” Power Sys. Conf.: Advanced
Metering, Protection, Control, Communication, and Distributed Resources,, Clemson, SC, USA, March 2007.
[3] ABB Group, “Differential Relays for Protection of AC Generator, Transformer and Station Bus,” Application Data 41-301E, Coral spring, FL, Apr.
1991.
[4] Z. Gajic, “Differential Protection for Arbitrary Three-Phase Power Transformers,” Department of Industrial Electrical Engineering and Automation ,
Lund University, Sweden, 2007.
[5] P. Kundur, N. J. Balu, M. G. Lauby, “Power system stability and control,” McGraw-Hill, Jan 1, 1994.
[6] Duke C. and M. Engineering Group, “Relay Construction and Maintenance Process for Verifying CT Circuits using the, Through Fault Test Method,”
TECP-CNST-TRM-00117, Raleigh, NC.