MOD/ESS/021-Rev03

Measuring the transformation ratio and checking the connection

symbol
1. General
The following information is taken from standards IEC 60076-1 and IEC 60726.

1.1 Definitions:

Phase displacement of a three-phase winding: this is the angular deviation between the vectors representing HV
and LV voltages of the equivalent terminals of the same winding pair.
Vectors are assumed to rotate in an anticlockwise direction.
The HV winding vector, whose 1st phase is oriented at 12 oclock on a time dial, acts as a reference, and the
phase displacement of all the other windings is normally expressed by a time index.

Connection symbol: this is the conventional symbol indicating the connection modes for HV and LV windings and
their relative phase displacements expressed by a combination of letters and time indexes.
The transformation ratio is measured on each transformer tap.
The connection symbol of three-phase transformers and the polarity of single-phase transformers must be checked.

2. Aim of the test

- check transformer connection conformity:
- check conformity of the transformation ratio k on each tap with respect to the guaranteed values.
The part of the transformer studied during this test is encircled in the diagram below:

3. Theoretical reminders
Calculation of the rated transformation ratio for each tap:
This ratio is only valid for the same given voltage reference.

Example:
Ratio of phase-to-earth or phase-to-phase HV/LV voltages.
To obtain the rated transformation ratio, you must (for Yd and Dy connections) multiply or divide, respectively,
by 3, according to whether the measured voltages are phase-to-earth or phase-to-phase.
For Yy and Dd connections, the rated transformation ratio is not modified.

france transfo
Copy printed for information purposes. Check for the latest version in the electronic database before use.
Page 1 of 3

MOD/ESS/021-Rev03

Measuring the transformation ratio and checking the connection

symbol
Common connections
The most common connections in France are:

Capital letters correspond to the highest voltage.

4. Test methodology
The connection symbol is checked and the transformation ratio is measured on each tap, at a voltage less than or
equal to 110V applied on the HV side.
Measurement consists of comparing for each phase the HV voltage in phase with the LV voltage.
This operation is carried out using a transformation measurement bridge that places in opposition the voltages in
phases in order to compare their modules.
The transformation ratio value is displayed on the measurement device.
Connection is correct when the ratio value is the same on each phase.
Example: connection Dyn11
- the HV side of the transformer is delta-connected.
- the LV side is star-connected with extended neutral.

Measurement of the transformation ratio

france transfo
Copy printed for information purposes. Check for the latest version in the electronic database before use.
Page 2 of 3

MOD/ESS/021-Rev03

Measuring the transformation ratio and checking the connection

symbol
The connection and the transformation ratio are the same as those defined by the manufacturer when:

I IA BI I I IB CI I I IC AI I
= = =
I Ia nI I
I Ib nI I
I Ic nI I

[HV voltage module

on the same column]

[LV voltage module

on the same column]

5. Test precautions
Do not use voltages greater than 50 V in the workshop.
Confine the test zone.
Supply the transformer by the winding with the highest voltage.
Breaking means by pedal or pushbutton compulsory in the workshop.

6. Personnel safety

7. Test results
The tolerances:
No-load transformation ratio:

For the main tap:

The lowest of the 2 values below:
a) 0.5% of the ratio specified by the manufacturer.
b) 10% of the real percentage of short-circuit voltage.

On the other taps

Must be dealt with in an agreement between the supplier and the client, but the ratio must be greater than the
smallest value of a and b).

france transfo
Copy printed for information purposes. Check for the latest version in the electronic database before use.
Page 3 of 3

MOD/ESS/022-Rev03

No-load loss and current measurement test on transformers

1. General
The measurement procedure is described in standard IEC 60076-1 and 60076-11.
All the values measured during this test are brought back to the main tap, unless otherwise specified by the client, with
the transformer initially at ambient temperature.

2. Aim of the test

characterise transformer no-load losses and current.

check that these characteristics comply with the prevailing standard.

In concrete terms, no-load losses are generated by the part encircled by a dotted line below:

2.1 Theoretical reminders

no-load losses: the power drawn up by the transformer when rated voltage at rated frequency is applied to the
terminals of one of the windings, when the other winding is at open circuit.

no-load current: the RMS value of the current necessary for magnetization of the magnetic circuit.

2.2 Test methodology

No-load losses and current must be measured on one of the windings (the other winding(s) are at open circuit):
- at rated frequency and at a voltage equal to rated voltage, if the test is carried out on the main tap,
- at rated frequency and at a voltage equal to the voltage of the appropriate tap, if the test is carried out on
another tap.
As a rule, the Low Voltage (LV) winding is supplied at rated voltage and at rated frequency, while the High Voltage
(HV) winding is open.
A three-phase precision wattmeter is used for measurement and gives directly:
- the applied voltage (true rms),
- the 3 currents (true RMS),
- the average current,
- the no-load losses.
NB:
The values measured on transformers with several LV windings are obtained on the winding with the highest voltage,
i.e. at the point where precision is best.

france transfo
Ejemplar imprimido en calidad de consulta. Verificar el nivel de puesta al da en la base informtica antes de toda utilizacin
Pgina 1 de 2

MOD/ESS/022-Rev03

No-load loss and current measurement test on transformers

Wiring diagram

Supply voltages must be monitored and controlled to give or take 0.125 % (corresponding to a maximum difference of
0.5 V for a voltage of 400 V).

3. Test precautions
The voltage measurement is taken directly to the terminals of the transformer. The measurement cables and power
supply must be connected so as to avoid priming with high voltage.

4. Personnel safety
In event of resonance, discharge the tested transformer prior to operation.

5. Test results
Application of reduced tolerances on no-load losses is negotiable at the time of the invitation to tender.
For this reason, to guarantee transformer performance, the International Electrotechnical Committee has stipulated
in standard IEC 60076-1 9, tolerances to be complied with as a function of the measurement taken:
Articles
1. No-load losses

2. No-load current
3. Total losses
(Po + Psc)

Tolerances
+ 15% of the declared values
provided that tolerances on total
losses are complied with.
+ 30% of the value declared by the
manufacturer.
+ 10% of the declared values

france transfo
Ejemplar imprimido en calidad de consulta. Verificar el nivel de puesta al da en la base informtica antes de toda utilizacin
Pgina 2 de 2

MOD/ESS/023-Rev02

Load loss and short circuit voltage measurement test on

transformers
1. General
The measurement procedure is described in standards IEC 60076-1 and IEC 60076-11.

2. Aim of the test

measure losses due to transformer load (Psc);

measure short-circuit voltage (Usc).

Equivalent diagram:

2.1 Theoretical reminders

short-circuit voltage: the voltage that must be applied between the line terminals of a winding to ensure flow of
nominal current when the terminals of the other winding are short-circuited.
It is always expressed as a percentage of nominal voltage.

load losses: these losses correspond to the active power that is drawn up (at rated frequency and at reference
temperature) when the rated current of the tap flows through the line terminals of one of the windings, when the
terminals of the other winding are short-circuited and the other windings, if any, are at open circuit.
These losses are also known as short-circuit losses (= Joule losses + Special losses).

Calculating load losses

- at ambient temperature (20 C on average);

Psc = PjoulesHV + PjoulesLV + Pspecial

= (3/2) x RHVx I 2

] + [(3/2) x RLV x I ]+ Pspecial

2

where RHV, RLV = phase-to-phase resistances.

For a delta connection:

Pjoules = 3 r (I/
3) 2
= (3/2) RT I 2 where RT = (2/3) x r
For a star connection:

Pjoules = 3 x r x I 2
= (3/2) x RT x I 2

where RT = 2 x r
Special losses are mainly made up of eddy current losses.
- At reference temperature:
Joule losses vary according to temperature whereas special losses are inversely proportional to temperature.

Psc T ref = K x Pjoules Tamb. + (1/K)x Pspecial Tamb.

where K = temperature correction constant:
for copper : K = (235 + Tref)/(235 + Tamb).
for aluminium: K = (225 + Tref)/(225 + Tamb).

france transfo
Copy printed for information purposes. Check for the latest version in the electronic database before use.
Page 1 of 3

MOD/ESS/023-Rev02

Load loss and short circuit voltage measurement test on

transformers
The short-circuit voltage Usc
This short-circuit voltage has a reactive component Ux and a component Ur, which unlike Ux depends on temperature.
In reduced values with respect to nominal quantities, we can write:

(Usc%) 2 = (Ux%) 2 + (Ur%) 2

where

Ur% = (Psc x 100) / Sn

Usc% = (Usc x 100) / Un

Example:
With an ambient temperature of 20C and at a refere nce temperature of 120C, we still have as a percen tage:

The expression of Usc at 120C as a % is then deter mined:

2.2 Test methodology

Short-circuit voltage and losses must be measured with a supply current at least equal to 50% of tap rated current.
The measured value of these losses must be multiplied by the square of the ratio of tap rated current over current
used for the test.
For transformers with a tap winding whose size is larger than 5 %, losses and short-circuit voltage must be
measured on the main tap and the two end taps.

As a rule, the MV winding is supplied at a nominal frequency of 50Hz, at a voltage giving a current as close as
possible to nominal current, with the LV winding short-circuited.
A three-phase precision wattmeter is used for measurement and gives directly:
- the applied voltage,
- the 3 currents,
- the average current,
- the load losses
The short-circuit voltage is expressed as a percentage of the nominal voltage:

Usc (%) = (Usc measured / Unominal) x 100

france transfo
Copy printed for information purposes. Check for the latest version in the electronic database before use.
Page 2 of 3

MOD/ESS/023-Rev02

Load loss and short circuit voltage measurement test on

transformers
Wiring diagram:

The measurement principle and the equipment used are the same as those for the no-load loss measurement test.

3. Test precautions
Measurements must be taken quickly to prevent winding temperature rises from introducing significant errors (winding
temperature must be controlled).
Measurement quality (in accuracy and precision) lies in the quality of practical production of the three-phase ShortCircuit.
To avoid introducing additional special losses (due to a bad Short-Circuit), the Short-Circuit bar cross-section must be
sufficient:
the Short-Circuit bar cross-section must be greater than or equal to the coil connection cross-section.
(In event of coil double connections, 2 Short-Circuit bars must be taken to comply with the cross-section stipulation).
Care must be taken with contact surface finish.

4. Personnel safety
5. Test results
For very special transformers, consult the Technical Department.

The tolerances
Tolerances are taken from the standard, unless otherwise specified by the client.
Articles
1.b Partial losses :
(No-load or load losses)
3. Short-circuit voltage
a) on the main tap

b) on the other taps

Tolerances
+ 15% of each partial loss,
provided that the tolerance is complied with
on total losses (+10% of values declared)
+ 7.5 % of the value declared by the manufacturer if the value of short-circuit voltage
(Usc) is > 10%.
+ 10 % of the value declared by the manufacturer if Usc is < than 10%.
+ 10 % of the value declared by the manufacturer if Usc is > 10%
+ 15 % of the value declared by the manufacturer if Usc is < than 10%

france transfo
Copy printed for information purposes. Check for the latest version in the electronic database before use.
Page 3 of 3

MOD/ESS/025-Rev03

Measuring resistance of HV and LV windings on transformers

1. General
Standard IEC 60076-11 (Dry type transformers) refers to the information in standard IEC 60076-1 (Oil-immersed
transformers).

2. Aim of the test

HV and LV resistances are the resistances internal to the transformer windings, seen from the HV and LV side that
generate Joule losses proportionally to the through current square.
Ohmic resistance for each of the transformer windings is measured.
These resistances are encircled by a dotted line in the equivalent diagram below:

2.1 Theoretical reminders

R = (U / I)
For the double bridge, we have:

if R1 = R4 and R2 = R5

2.2 Test methodology

The test is carried out in DC.
During this test, resistance is measured between the line terminals of each winding, and the ambient temperature is
read in the channel between the HV and LV windings.
There are a number of possible methods:
The resistance of HV and LV windings is normally measured using a digital micro-ohmmeter. This device is
connected between each phase
It is also possible to use the voltamperemetric method, for high resistance windings, or a double bridge for low
resistance windings.

france transfo
Copy printed for information purposes. Check for the latest version in the electronic database before use.
Page 1 of 2

MOD/ESS/025-Rev03

Measuring resistance of HV and LV windings on transformers

Wiring diagram:

Measurement of MV and LV resistances by metric voltampere

method

Measurement of LV resistance with double bridge

where
RT = (2/3) x r
In the case of Delta connection of windings.
RT = 2 x r
In the case of Star connection of windings.
Where RT is the equivalent resistance between phases, and r the resistance of a transformer winding.

3. Test precautions
For dry type transformers, the test begins after a rest period of at least 48 hours of the transformer at stabilised
ambient temperature. Resistance and temperature must be measured simultaneously, using for winding temperature,
sensors placed at significant positions, preferably in the windings.
For oil-immersed transformers equipped with oil-filled pockets, temperature is measured at the pocket.
For Distribution oil-immersed transformers (not equipped with oil-filled pockets), the temperature measured is the
ambient temperature.

4. Personnel safety
Do not disconnect the installation when the measurement device is energised.
Opening of an inductive circuit could result in an electric arc.
There is a potential risk for personnel and measurement equipment during these transient states.

5. Test results
This measurement is an intermediate phase to determine load losses and temperature rises.

france transfo
Copy printed for information purposes. Check for the latest version in the electronic database before use.
Page 2 of 2

MOD/ESS/027-Rev05

Dielectric tests on transformers

1. General
The following test methods are taken from standards NF EN 60726 (2003), IEC 60726 (1982) and IEC 60076-3
(2000), modified, if necessary, by an internal FT rule.
Useful information on HV measurements and on measurement device calibrations are available in standards IEC
60060-1 and IEC 60060-2 on high voltage measurement techniques.
To verify withstand between phases, between turns and at the transformer earth, 2 routine tests are conducted: an
induced voltage test and an applied voltage test.

2. Induced voltage test

2.1 Aim of the test

Check quality of internal insulation of transformer windings and insulation between phases under stresses
representative of voltage rise at power frequency on networks.

For Um 72.5 kV the test is normally conducted with partial discharge level measurements to check absence of
partial discharges in conditions with the transformer in operation. Refer to standard IEC 60076-3

2.2 Theoretical reminders

One of the transformer windings is supplied with AC voltage (as a rule, the lower voltage winding, for example 410V).
An AC voltage of the same frequency is induced in the other winding. Voltage per turn is the same in all coils. The total
voltage of each coil is proportional to its number of turns.
Insulation between turns, between layers and between disks of windings is stressed during the test, together with
insulation between phases. This insulation is of solid, liquid and/or gaseous nature. The capacity of this insulation to
withstand a voltage is linked to the duration of application of this voltage and, thus, in the case of voltages at power
frequency, to the number of cycles applied.

2.3 Test methodology

Three-phase transformers must undergo the test with a symmetrical three-phase power supply.
When a transformer has a neutral, it must be earthed during the test.
An AC voltage must be applied to the terminals of a transformer winding. Voltage shape must be as close as possible
to sinusoidal shape, and its frequency must be sufficiently high with respect to rated frequency to avoid excessive
magnetizing current during testing.
Unless otherwise specified, the value of the test voltage, along a winding without transformer tap, must be as
close as possible to twice the rated voltage, without however exceeding the transformer applied voltage test value.
The test must begin at a voltage that is less than or equal to a third of the value of the test voltage, and voltage must
be increased as quickly as possible. On completion of testing, voltage must be quickly reduced to a voltage less than
one third of test value before breaking.
Duration of the test at full test voltage must be 60 s for all test frequencies twice the rated frequency. When test
frequency is greater than twice the rated frequency, the duration of the test expressed in seconds must be:

120 rated frequency / test frequency

with a minimum of 15 s
(IEC 60076 chapter 12.1)
The test is satisfactory if there is no collapse in test voltage or sudden fluctuation in current.

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 1 of 4

MOD/ESS/027-Rev05

Dielectric tests on transformers

200 Hz

Current transformer

LV

MV
2*Un

2*Un

U1-2
U2-3
U1-3

Generator
Motor+alternator

I1
I2
I3

Measurement
device

Tested transformer

2.4 Test precautions

room temperature > 10C

unless otherwise specified, the transformer is at ambient temperature

It is not necessary for the transformer to be in its enclosure

Specific instructions can be given concerning:

Test duration

Tap position during testing

For Um 72.5 kV the test is normally conducted with measurement of partial discharge level. Refer to standard
IEC 60076-3

For transformers with HV windings with non-even insulation, refer to standard IEC 60076-3.

3. Applied voltage test

3.1 Aim of the test
Check insulation quality:

Between transformer primary and secondary windings, on the one hand

Between these windings and the parts of the transformer intended to be earthed, on the other under stresses
representative of voltage rise at power frequency.

In the case of graduated insulation, the test is replaced by an induced voltage test.

3.2 Theoretical reminders

Insulation between windings and between winding and magnetic circuit, with earthed metal parts, are stressed during
testing. This insulation is of solid, liquid and/or gaseous nature. The capacity of this insulation to withstand a voltage is
linked to its thickness, integrity and degree of impregnation.

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 2 of 4

MOD/ESS/027-Rev05

Dielectric tests on transformers

3.3 Test methodology
The test is conducted with a single-phase AC voltage of a frequency 80% than rated frequency.
Full test voltage must be applied for 60 seconds between all the tested winding terminals connected to one another,
on the one hand, and, on the other hand, all the terminals of the other windings, the magnetic circuit, the enclosure or
the tank, all earthed.
The test must begin by applying an AC voltage less than or equal to one third of the specified test value. Voltage is
brought to the test value as quickly as possible. Full voltage is applied for 60 seconds. On completion of testing,
voltage is quickly brought down to a value less than one third of test value before breaking.
Test voltage depends on the level of insulation of the winding and is specified in the standards or the clients
specification sheet.
The test is satisfactory if there is no collapse of the test voltage.

Single-phase
step-up
transformer

Capacitor voltage
divider or VT

Tested
transformer

The test is performed by connecting the HV to earthing system and powering the BT and inversely.

3.4 Test precautions

For windings with non-uniform insulation, the test is conducted with allowance for insulation of the neutral terminal

Room temperature > 10C

Unless otherwise specified, the transformer is at ambient temperature

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 3 of 4

MOD/ESS/027-Rev05

Dielectric tests on transformers

4. Personnel safety
Recommendations concerning applied and induced voltage tests are as follows:

After all dielectric tests, use an insulating pole to earth the terminals and discharge the transformer, prior to work
on the device.

Should test duration exceed 5 minutes, the test must not be carried out in a confined area.

5. Test results
The test is conform if the if it there is no collapse in voltage.

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 4 of 4

MOD/ESS/029-Rev05

Partial discharge tests on transformers

1. General
The following test conditions are taken from standards NF EN 60726 (2003), EN 60726 (2003), IEC 60076-11 (2004),
IEC 60076-3 (2000), and EDF HN 52 S 07 (1975) and its amendment 1 (1978), modified if applicable by an FT inhouse regulation.
Useful information on partial discharge measurements can be found in standard IEC 60270.
The test checks absence of transformer partial discharges in operating conditions: in addition to normal operating
voltage, transformer insulation must be able to withstand the surges that occur during operation without generation of
partial discharges persisting on resumption of normal voltage

2. Aim of the test

The test checks absence of transformer partial discharges in operating conditions.
For cast resin transformers, partial discharge measurement is a routine test on windings where Um 3.6 kV.
For oil-immersed distribution transformers, partial discharge measurement is a special test specific to EDF, described
in specification HN 52-S-07.
For power transformers where Um > 72.5 kV, partial discharge measurements can be requested during the induced
voltage test. Refer to standard IEC 60076-3.

3. Theoretical reminders
Partial discharges are electrical discharges generated by application of an electric field, located in insulating
environments. These discharges, which take the form of individual pulses, progressively deteriorate the dielectric
properties of insulating materials.
7.

Partial discharge (PD):

a localized electrical discharge that partially short-circuits the insulating distance separating conductors and that may
or may not be adjacent to a conductor. As a rule, partial discharges are a consequence of local concentrations of
electrical stresses in the insulation or on the insulation surface. Normally these discharges appear in the form of
pulses with durations far shorter than 1 s.
8. Apparent load q:
the apparent load of a PD pulse that, if injected in a very short time between the terminals of the object being tested
placed in a specified test circuit, would give the same reading on the measurement device as the actual PD pulse.
Apparent load is normally expressed in picocoulombs (pC)
9.

Amplitude of the largest repetitive partial discharge

the largest amplitude recorded by a measurement system with a pulse train response complying with standard IEC
60270
10. Specified partial discharge amplitude:
the largest amplitude of a quantity characterizing PD pulses authorized in a tested object at a specified voltage, by
applying a specified conditioning and test procedure. For tests conducted at AC voltage, the amplitude specified for
apparent load q is the amplitude of the largest repetitive partial discharge
11. Background noise
the signals detected during PD tests not originating in the object being tested. Background noise can be made up of
white noise from the measurement system, radio broadcasts or other continuous or pulsed signals.

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 1 of 7

MOD/ESS/029-Rev05

Partial discharge tests on transformers

12. Partial discharge appearance voltage Ua
the applied voltage from which repetitive partial discharges are observed for the first time in the test object, when the
voltage applied to this object is increased progressively from a low value for which such partial discharges are not
observed. In practice, the appearance voltage Ua is the lowest applied voltage for which the amplitude of a quantity of
the PD pulse becomes equal to or greater than a specified low value.
13. Partial discharge extinction voltage Ue
the applied voltage from which repetitive partial discharges cease to be observed in the test object, when the voltage
applied to this object is reduced progressively from a high value for which such partial discharge pulses are
observed. In practice, the extinction voltage Ue is the lowest applied voltage at which the intensity of a quantity relating
to the PD pulses becomes equal to or less than a specified low value.

4. Test methodology
14.

Measurement circuit

A capacitor free from partial discharges, C, in series with a detection impedance, Zm, is connected to each of the HV
terminals. The 3 capacitances must be identical. The measuring device is connected to the measurement impedance
terminals and analyzes the signals generated by the partial discharges.
The partial discharges produced in the transformer cause load transfers in the measurement circuit and current pulses
in the measurement impedance Zm.
Measurement sensitivity depends on the value of capacitance C. An acceptable sensitivity is normally reached when C
is approximately 1 nF or more.
Coupling capacitances

LV

MV

3
Tested transformer
Zm

Generator
150 or 200 Hz

Measurement
device

High-pass filter

Low-pass filter

A measuring device bandwidth from 40 to 400 kHz gives significant results for transformer measurements.

15. Calibrating the measurement circuit

The measurement circuit is calibrated systematically for each transformer measured.
Partial discharges, which are localized, generate electric pulses that are attenuated both in the transformer windings
and in the measurement circuit. A calibration, carried out on the measurement circuit connected to the transformer, is
thus required to know the value of this attenuation.
As test object capacitance affects circuit characteristics, each device must be calibrated.

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 2 of 7

MOD/ESS/029-Rev05

Partial discharge tests on transformers

Devices are calibrated by injecting pulses simulated using a calibrated discharge generator at the transformer
terminals. The rate of repetition of the calibration pulses must be of the order of twice excitation voltage frequency.
Generator capacitance must be far lower than the capacitances C of the measurement circuit.
In view of the partial discharge levels measured, calibration must be carried out between 50% and 200% of specified
PD amplitude, while being at the same time 1.5 times greater than background noise.
16.

Test diagram

The measurement circuit capacitances are connected to the HV winding terminals via conductors of minimum
length. These connecting conductors must have a sufficiently high diameter to avoid the corona effect in the
voltage range concerned. Connections between terminals, conductors and capacitances must not exhibit any
stress concentrations likely to generate discharges.

The magnetic circuit, chassis, flanges, enclosure, tank and all metal parts other than the windings must be
earthed at a single point.

The earthing terminals of the measurement circuit capacitances must be earthed at a single point.

Special care must be taken when earthing. Earthing loops must be avoided.

17.

Measuring test voltage

In the case of three-phase transformers, the three phase-to-phase voltages may be unbalanced. The largest of the 3
voltages must be measured and is taken into account when carrying out the test cycle.
18.

Conditions for applying voltage and partial discharge limit levels

The LV winding is supplied from a filtered three-phase source such that 3 balanced voltages appear on the HV output
terminals. Voltage is as close as possible to the sinusoidal waveform and of a frequency that is sufficiently higher than
rated frequency to avoid an excessive magnetizing current during testing.

For TRIHAL transformers:

Partial discharges are measured once all the dielectric tests have been completed. They are measured on the
transformer in the minimum tapping position and in the nominal tapping position.
The test sequence is as follows, according to the routine test defined in standards EN 60726 (2003) and IEC 60076-11
(2004):
a phase-to-phase prestress voltage of 1.8 Un is induced for 30 seconds, followed without interruption by a phase-tophase voltage of 1.3 Un for 3 minutes. Partial discharges are measured throughout this cycle, together with Ua and
Ue.

1.8 Un

1.3 Un

30 sec

3 min

Un
0V

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 3 of 7

time

MOD/ESS/029-Rev05

Partial discharge tests on transformers

Test sanction:

Partial discharge limit level is 10 pC at 1.3 Un, throughout the 3 minute period

Partial discharge limit level is 50 pC at 1.8 Un, throughout the 30 second period

Note: in some cases, at the clients request formulated on ordering, the partial discharge measurement is taken with
one of the HV terminals earthed. Three tests are then required with, for each one, a different earthed line terminal.

For oil-immersed distribution transformers:

Partial discharges are measured once all the dielectric tests have been completed. They are measured on the
transformer in the nominal tapping position.
The test sequence is as follows, according to specification EDF HN 52 S 07 (1975) and its amendment 1 (1978):

1.5Um
1
Um

30 sec

1.2 Um

5 min
Um
Un
0V

time

a phase-to-phase prestress voltage of 1.5 Um is induced for 30 seconds, followed without interruption by a phase-tophase voltage of 1.2 Um for 5 minutes. Partial discharges are measured throughout this cycle, during voltage increase
and decrease and plateau periods, together with Ua and Ue.

Test sanction:

Partial discharge limit level is 50 pC at Um, on voltage raising and lowering

Partial discharge limit level is 100 pC at 1.2 Um, throughout the 5 minute period

Discharge level should not markedly increase during the voltage plateau at 1.2 Um, and the extinction voltage of
discharges observed during voltage lowering should not be too different from discharge appearance voltage. A
difference of less than 15% between discharge appearance and extinction voltages should be considered as
normal.

Note: If, in special cases, Um is greater than 1.2 Un, the voltages to be applied must be covered by specific
specifications.

For oil-immersed power transformers:

Partial discharges are measured once all the dielectric tests have been completed. They are measured on the
transformer in the nominal tapping position.
Specific case:
In some cases it may be necessary to take partial discharge measurements on the minimum tap in order to eliminate
potential resonance problems.
Precautions before conducting the tests

When topping up after vacuum filling, fully degas the transformer;

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 4 of 7

MOD/ESS/029-Rev05

Partial discharge tests on transformers

The transformer must be left to rest for one night before platform tests;

Following installation on the test platform, all the components must be purged and in particular the bushings at
their top and base and all the accessories in contact with oil (relays, valves, etc.).

Transformers with Um 72.5 kV

Normally no partial discharge level measurements are required. However a partial discharge measurement may be
taken at the request of the technical departments or the project engineering & design department on some
transformers.
Transformers with Um > 72.5 kV
Unless otherwise specified in an agreement, these transformers must all be tested with measurement of partial
discharge level. Phase-to-phase test voltages must not exceed power frequency withstand voltages.
voltage level U2 must be:
1.3 Um / 1.732 phase-to-earth and 1.3 Um phase-to-phase
The test sequence is as follows according to standard IEC 60076-3 (2000-03):

test duration

Note: During voltage increase up to U2 and decrease as from U2, the values of any discharge appearance and
extinction voltages must be noted.
Background noise level must be less than 100 pC.

The test is satisfactory if:

there is no collapse of test voltage;

the permanent level of apparent load at U2 during the second 5 minute period is 300 pC for all measurement
channels;

the partial discharge level does not show a continuous tendency to increase;

the permanent level of the apparent load does not exceed 100 pC at 1.1 Um/ 1.732.

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 5 of 7

MOD/ESS/029-Rev05

Partial discharge tests on transformers

For NEUTRAL EARTHING COIL
The cycle of measurement of partial discharge on Neutral earthing coil is the same than the test, on oil immersed
distribution transformer.
To induce the voltage, we used a step up transformer without partial discharge in series with the Neutral earthing coil :

Step up
transformer

Generator
Coupling
capacitance

Neutral earthing coil

Measurement
Low pass filter
Z

Fig. Circuit for measuring of partial discharge on neutral earthing coil

5. Test precautions

The power supply used for testing must be filtered to avoid generating background noise greater than guaranteed
PD level

The transformer must be clean and without condensation

Unless explicitly required, the transformer is at ambient temperature

Distances to be complied with between the connecting cables, these cables and the earth and between the
transformer terminals and capacitances:

Un (kV)

7.2

12

17.5

24

36

40

distances (mm)

90

150

220

300

450

500

Partial discharge measurement is very sensitive to the voltage sequence:

Appearance of partial discharges depends on voltage application time and voltage build-up speed. It is thus
very hard to reproduce.

The amplitude of all quantities relating to a PD pulse may vary in random fashion during successive periods
as well as exhibit a general increase or decrease with voltage application time

Extinction of partial discharges can be affected by voltage application duration and amplitude as well as by
voltage decrease speed.

Partial discharge measurement is very sensitive to disturbances:

Disturbances that occur even when the test circuit is not supplied. These stem, for example, from operations
in other circuits, machines with collector, high voltage tests conducted nearby, radioelectric emissions, etc.
including the basic noise of the actual measurement device. Disturbances may also occur when the HV supply
is connected to the test circuit, with zero voltage.

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 6 of 7

MOD/ESS/029-Rev05

Partial discharge tests on transformers

Disturbances that are present only when the circuit is supplied but that do not occur in the object being tested.
These disturbances normally increase with test voltage and may, for example, include partial discharges on
conductors at high voltage. Disturbances can also come from arcing due to imperfect earthing of nearby
objects or from defective contacts between parts brought to high voltage, for example arcings between
distribution blocks and other HV conductors connected to the distribution blocks for test needs. Disturbances
may also be caused by test voltage high number harmonics that are in or close to the bandwidth of the
measurement system. These high harmonics are often present in the LV source due to the presence of
switching devices with semi-conductors (thyristors, etc.) and are transmitted, with arcing discharge noises, via
the test transformer or other connections, to the test and measurement circuits.

Use of an oscilloscope as a guideline measurement device or evaluation of quantities relating to PDs

numerically acquired may help the operator distinguish between partial discharges occurring in the object
being tested and external disturbances, for example background noise. This may help determine the type of
disturbance or identify the type of partial discharge.

Disturbances can be reduced by suitable earthing near the test area of all the conductive structures (that must
also possess no sharp protuberances) and by filtering the test and measurement circuit supply networks.

Partial discharges generate modifications in the gases present in the defects: modification of type of gas and
pressure. They also change the surface of the defects: modification of resistivity for example. Some of these
phenomena are reversible after a certain time, others are not. Repetition of partial discharge measurements may
thus yield different results, a difference that may depend on the period of time elapsing between the 2
measurements.

6. Personnel safety
Test recommendations are as follows:
in event of an incident during testing, thoroughly discharge the capacitances prior to handling

7. Test results
The following results are logged into the computer
partial discharge level at 1.8 and 1.3 Un, or 1.2 Um and Um, as applicable
Ua and Ue
Should the guaranteed values be exceeded, a NonConformity sheet is opened and the relevant expert is informed.
19.

Using results in event of nonconformity:

Display of the output voltage of individual pulses on an oscilloscope screen may facilitate recognition of partial
discharge type and allow a difference to be made with disturbances. Voltage pulses should be displayed on a linear
time base activated by test voltage, on a sinusoidal time base synchronized on test voltage frequency or on an
ellipsoidal time base synchronized with test voltage frequency.
Example of an oscilloscope recording obtained in event of partial discharges inside an insulator:

For special transformers, contact the relevant expert.

france transfo
Copy printed for consultation. Check the update level in the computer database prior to use.
Page 7 of 7