IEEE Std 436™-1991 (R2007)

(Revision of IEEE Std 436-1977)

IEEE Guide for Making Corona (Partial

Discharge) Measurements on Electronics
Transformers

Sponsor

Electronics Transformer Technical Committee

of the IEEE Power Engineering Society

Reaffirmed 5 December 2007

Approved 11 February 1991

IEEE-SA Standards Board

Abstract: A uniform procedure for making corona (partial discharge) measurements by electrical means on
electronic tranformers is presented. Methods of applying voltage stress, the use of a sine-wave voltage to
simulated dc and ac combinations, the types and limitations of voltage stresses encountered, and the accept-
able limits of discharge pulse energy are included. Recommended test conditions and the need for negotiation
of special tests are discussed. Test apparatus and calibration are described. The aim is to establish a common
ground of understanding between transformer and systems design engineers and transformer manufacturers
in the development of service and performance requirements.
Keywords: corona, partial discharge, insulation breakdown, transformer insulation

The Institute of Electrical and Electronics Engineers, Inc.

345 East 47th Street, New York, NY 10017-2394, USA

Copyright © 1991 by the

Institute of Electrical and Electronics Engineers, Inc.
All rights reserved. Published 1991
Printed in the United States of America.

ISBN 1-55937-126-9

No part of this publication may be reproduced in any form,

in an electronic retrieval system or otherwise,
without the prior written permission of the publisher.

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 IEEE Standards documents are developed within the Technical
Committees of the IEEE Societies and the Standards Coordinating
Committees of the IEEE Standards Board. Members of the committees
serve voluntarily and without compensation. They are not necessar-
ily members of the Institute. The standards developed within IEEE
represent a consensus of the broad expertise on the subject within the
Institute a s well as those activities outside of IEEE which have
expressed an interest in participating in the development of the
standard.
Use of an IEEE Standard is wholly voluntary. The existence of an
IEEE Standard does not imply that there are no other ways t o produce,
test, measure, purchase, market, or provide other goods and services
related to the scope of the IEEE Standard. Furthermore, the viewpoint
expressed at the time a standard is approved and issued is subject to
change brought about through developments in the state of the art and
comments received from users of the standard. Every IEEE Standard
is subjected to review at least every five years for revision or reaffir-
mation. When a document is more than five years old, and has not
been reaffirmed, it is reasonable to conclude that its contents, al-
though still of some value, do not wholly reflect the present state of the
art. Users are cautioned to check to determine that they have the latest
edition of any IEEE Standard.
Comments for revision of IEEE Standards are welcome from any
interested party, regardless of membership affiliation with IEEE.
Suggestions for changes in documents should be in the form of a pro-
posed change of text, together with appropriate supporting comments.
Interpretations: Occasionally questions may arise regarding the
meaning of portions of standards as they relate t o specific applica-
tions. When the need for interpretations is brought to the attention of
IEEE, the Institute will initiate action to prepare appropriate re-
sponses. Since IEEE Standards represent a consensus of all con-
cerned interests, it is important to ensure that any interpretation has
also received the concurrence of a balance of interests. For this reason
IEEE and the members of its technical committees are not able t o
provide an instant response to interpretation requests except in those
cases where the matter has previously received formal consideration.
Comments on standards and requests for interpretations should be
addressed to:
Secretary, IEEE Standards Board
445 Hoes Lane
P.O.Box 1331
Piscataway, NJ 08555-1331
USA

IEEE Standards documents are adopted by the Institute of Electrical

and Electronics Engineers without regard to whether their adoption
may involve patents on articles, materials, or processes. Such adop-
tion does not assume any liability to any patent owner, nor does it
assume any obligation whatever to parties adopting the standards
documents.

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 (This Foreword is not a part of IEEE Std 436-1991,IEEE Guide for Making Corona (Partial Discharge) Measurements on
Electronics Transformers.)

This guide was written to provide electronics transformer engineers with a standard method for
detecting and measuring corona (partial discharge) in electronics transformers. Since there is no
simple definition or unit for corona, it is felt that a guide of this nature will be useful in establish-
ing a common ground of understanding between transformer and systems design engineers and
transformer manufacturers in the development of service and performance requirements.
This publication was prepared by the Insulation Systems Subcommittee of the Electronics
Transformers Technical Committee of the IEEE Power Electronics Society. The Subcommittee
membership was as follows:

R, A. F'rantz, Chair
J. S.Andresen P. K. Goethe R. L. Sell
C. J. Elliott R. R. Grant B. D. Thackwray
H. Fickenscher H. E. Lee R. M. Wozniak
D. N. Ratliff

When the Electronics Transformer Technical Committee approved this guide, the membership
was as follows:

E Fickenscher, Chair
E. D. Belanger J. Silgailis H. E. Lee
C. J. Elliott J. Tardy W. E. L u c a n
R. A. Frantz B. D. Thackwray D. N. Ratliff
P. K. Goethe R. M. Wozniak R. L. Sell

The following persons were on the balloting committee that approved this standard for submis-
sion to the IEEE Standards Board:
J. S.Andresen R. R. Grant R. L. Sell
E. D. Belanger 0. Kiltie J. Silgailis
R. P. Carey L. W. Kirkwood J. Tardy
E.J. Elliott H. E. Lee B. D. Thackwray
H. Fickenscher R. Lee H. I. Tillinger
R. A. Frantz H. W. Lord R. G. Wolpert
P. K.Goethe W. Lucarz R. M. Wozniak
D. Ratliff

The final conditions for approval of this guide were met on February 11, 1991. This guide was
conditionally approved by the IEEE Standards Board on September 28, 1990 with the following
membership:

Marc0 W. Migliaro, Chair J a m e s M. Daly, Vice Chair

Andrew G. Salem, Secretary
Dennis Bodson Kenneth D.Hendrix Lawrence V. McCall
Paul L. Borrill John W. Horch L. Bruce McClung
Fletcher J. Buckley Joseph L. Koepfingerl Donald T. Michael*
Allen L. Clapp Irving Kolodny Stig Nilsson
Stephen R. Dillon Michael A. Lawler Roy T. Oishi
Donald C. Fleckenstein Donald J. Loughry Gary S. Robinson
J a y Forstefl John E. May, Jr. Terrance R. Whittemore
Thomas L. Hannan Donald W.Zipse

*Member Emeritus

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 caltata
SECTION PAGE
1 . Scope ....................................................................................................... 6

2. Purpose .................................................................................................... 6

.
3 References ................................................................................................ 6

4. D e f i n i t i o n s ................................................................................................ 6

5. Test Conditions .......................................................................................... 7

5.1 Application of This Guide ......................................................................... 7
5.2 Conditions for Acceptable Corona Tests ......................................................... 7
5.2.1 Frequencies at Which Corona Measurements May Be Performed ................... 7
5.2.2 Effects of Voids in Solid Insulation ....................................................... 7
5.2.3 Corona Discharge Pulse Rise Time ...................................................... 7
5.2.4 Circuit Response ............................................................................ 7
5.3 Signal-to-Noise Ratio .............................................................................. 8
5.4 Discrimination Device Sensitivity .............................................................. 8

6. Apparatus ................................................................................................. 8
6.1 Test-Voltage Supply ................................................................................ 8
6.1.1 Self-Excitation .............................................................................. 8
6.2 Coupling Capacitor ................................................................................. 8
6.3 Coupling Impedance ............................................................................... 8
6.4 Corona Detector ..................................................................................... 8
6.4.1 Amplifier ................................................................................... 10
6.4.2 Signal Readout Instrument(s1.......................................................... 10
6.5 Calibration Pulse Generator ..................................................................... 11
6.6 Calibration Coupling Capacitor ................................................................ 11
6.7 Control Metering .................................................................................. 11

7 . C a l i b r a t i o n ............................................................................................... 11
7.1 Deflection Sensitivity ............................................................................. 11
7.2 When Calibration Is Performed ................................................................. 11
7.3 Calibration Technique ........................................................................... 11
7.4 Calculating Detection Sensitivity ............................................................... 12

8. Test Method .............................................................................................. 12

8.1 Group I .............................................................................................. 12
8.2 Group I1 ............................................................................................. 12
8.3 Group I11 ............................................................................................ 12

9. Test Requirements ..................................................................................... 12

9.1 Ambient Conditions............................................................................... 12
9.2 Calibration Sensitivity ........................................................................... 12
9.3 Test Voltage ........................................................................................ 13
9.3.1 DC Plus AC Test Voltage .................................................................. 13
9.4 Rate of Application ................................................................................ 13
9.5 Duration of Tests .................................................................................. 13

10. Bibliography ............................................................................................. 13

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 FIGURES PAGE
Fig 1 Typical Circuit for Corona Measurement. Circuit 1 ............................................ 9
Fig 2 Typical Circuit for Corona Measurement, Circuit 2 ............................................ 9
Fig 3 Circuit Modifications for Self-Excitation ...................................................... 10
Fig 4 Parallel-T RC Filter Network ................................................................... 10
Fig 5 Peak-to-Peak Voltmeter Circuit ................................................................. 11

APPENDIXES

Appendix A. Preconditioning for Determination of CIV............................................... 14

Appendix B . The Magnitude and Energy of Discharges ............................................... 15
Appendix C. Suggested Specification Requirements ................................................... 16

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 IEEE Guide for Making Corona
(Partial Discharge) Measurements on
Electronics Transformera

This guide covers the detection of corona T)ris guide shall be used in conjunction with
(partial discharge) and the measurement of its the following publications.
magnitude in electronics transformers. Test
conditions, test apparatus, calibration, and test i l l ASTM D1868-1981 (R19901, Standard
requirements are included. Method for Detection and Measurement of
Partial Discharge (Corona) Pulses in
Evaluation of Insulation Systems.'

[21 IEEE Std 100-1988, IEEE Standard

Corona (partial discharge) is an ionization Dictionary of Electrical and Electronics
phenomenon that can cause deterioration in T e r m s A t h ed. (ANSI).2
certain insulation systems. Its presence in
electronic equipment can be harmful through [3] MIL-T-27E, Military Specification
its manifestation a s circuit noise, spurious Transformers and Inductors (Audio, Power,
voltage or current pulses, or other undesirable and High-Power Pulse), General Speci-
effects t h a t lead t o circuit malfunctions. fic:&ion for (4 April 1985).3
Transformers and inductors used in elec-
tronic circuits at high voltages can be loca-
tions where this harmful phenomenon origi- 4. Definitions
nates. Even in instances where discharges
are not of concem in circuit performance, they Electrical terms used in this publication are
may be of concem because of their adverse ef- in accordance with those given in IEEE Std
fect on component life. It is, therefore, essen- 100-1988121.4
tial that the insulation system be free of corona
discharges at operating voltage stresses. The
magnitude or intensity of these discharges is
usually of such a low level that very sensitive
electronic measuring apparatus and tech-
niques are required to detect them.
'ASTM publications are available from the American
This guide presents a uniform procedure for Society for Testing and Materials, 1916 Race Street,
making corona (partial discharge) measure- Philadelphia,PA 19103, (215)299-5400.
ments by electrical means on electronics
21EEE publications are available from the Institute of
transformers. Methods of applying voltage Electrical and Electronics Engineers, Inc., Service
stress, the use of a sine-wave voltage to simu- Center, 445 Hoes Lane, P.O. Box 1331, Piscataway, NJ
late dc and ac combinations, the types and 08855-133l,1800878-4333.
limitations of voltage stresses encountered, 3This Military Specification is available from the
and the acceptable limits of discharge pulse Commanding General, US Army Communications-
energy are included. Recommended test con- Electronics Command, Ft. Monmouth, NJ 07703.
ditions and the need for negotiation of special 4The numbers in brackets refer to those of the references
tests are discussed. in Section 3.

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 5. Test Conditions 5.2.2 Effects of Voids in Solid Insulation.
Corona discharges across voids in solid insu-
lation transfer some of the voltage stress to the
S.1 Application of This Guide. The test condi- solid insulation, thus reducing the stress on
tions applicable t o electronics transformers the void. This removes the condition neces-
include all the configuration and circuit ar- sary for additional discharges until some of
rangements used in power transformers and, the charge dissipates or until the applied volt-
in addition, include a large number of induc-
age changes enough to reestablish the stress
tive devices with special circuit characteris-
required for corona discharge.
tics and with electrical stress on insulation
6.2.2.1 Testing With DC Voltage. When
covering the entire range of frequencies from the applied voltage is dc, continuous corona
dc through microwave and all combinations of
(recurring discharges) usually requires a
these frequencies. At the present time, this
higher voltage than when the stress is ac. The
guide will be limited t o detection and mea- overvoltage necessary to establish continuous
surements with sine-wave applied voltage
corona with dc applied will depend largely on
ranging in frequency from zero to approxi-
the resistance of the insulation through which
mately 2000 Hz. The detection and measure- the charge must dissipate to reestablish the
ment of corona discharges with applied
stress. This means that in apparatus where dc
voltages at higher frequencies or with stress alone is characteristic, the corona dis-
nonsinusoidal wave shapes requires special
charges should be measured with dc voltage
techniques, since the discharge pulse itself
applied. Caution should be used when the ap-
also has a wide frequency spectrum, a portion
plied voltage exceeds the peak working voltage
of which may fall within the operating
of the transformer, since the ratio of the corona
frequency range. inception voltage t o the insdation breakdown
voltage may be very close t o unity. However,
5.2 Conditions for Acceptable Corona Tests. this ratio may be as low as 0.5 when the applied
The duplication of the exact voltage stresses voltage is ac.
that occur in service, including frequencies 5.2.2.2 Testing With Combined DC and
and transients, is acknowledged to result in AC Voltage. When the voltage stress is
the most accurate corona discharge measure- combined dc and ac, continuous corona
ments; however, insistence on exact condi- discharges will be due t o the peak-to-peak
tions may be impractical. The fundamental voltage (calculated by summing the dc and
conditions for acceptable corona tests are peak ac voltages) and therefore will occur as if
based on the assumptions and limitations that due to an ac voltage of the same peak-to-peak
follow. value. This means t h a t in transformers
5.2.1 Frequencies at Which Corona characterized by combined a c and dc stress,
Measurements May Be Performed. Corona measurements should be made using ac test
discharges are a function of the peak-to-peak voltage whose magnitude is determined with a
voltage, provided the voltage contains no in- peak-to-peak voltmeter (see 6.7).
termediate peaks and is applied long enough 6.25 Corona Discharge Pulse Rise Time.
for ionization to take place. Corona measure- The corona discharge pulse voltage is as-
ments may be performed at any frequency un- sumed to have a sufficiently high rate of in-
der 2000 Hz; they need not be made at the crease that its initial distribution throughout
operating frequency. However, discharges are the circuit is controlled by the circuit capaci-
affected by the test frequency; as the frequency tances.
is increased, variations occur in the response 6.2.4 Circuit Response. The response of the
times of ions within the insulation to the circuit, which includes the insulation system,
rapidly varying externally applied electric to frequencies and their harmonics may com-
field. The resulting nonuniform internal plicate corona discharge tests. Square waves,
electrical stress leads t o a decrease in the unidirectional pulses, rapid switching, or
corona inception voltage. Care must therefore high frequencies will produce stress patterns.
be taken in interpreting the results of tests The application of test voltages below rated de-
performed at other than the operating sign frequencies is generally valid, but
frequency. caution must be observed when these voltages

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 IEEE
Std4361991 JEEE GUIDE FOR MAKING CORONA (PARTIALDISCHARGE)

are applied by induction so as not t o saturate 6.1.1 Self-Excitation. As an alternative to

the core. the use of an external high-voltage source to
supply the test voltage, the high-voltage stress
5.3 Signal-to-Noise Ratio. A minimum sig- across a winding may be induced through
nal-to-noise ratio of 2 to 1 should be established self-excitation. In this method, a lower-volt-
by reducing all electromagnetic interference, age winding in the transformer is excited by
regardless of its origin, to less than one half of the use of a low-voltage power source
the amplitude of the minimum corona dis- (including a filtered ac line) whose wave-
charge level to be detected. shape characteristics remain as stated in 6.1.
The transformer ratio is used to create a high-
5.4 Discrimination Device Sensitivity. It voltage stress across the winding under test.
should be demonstrated that equipment incor- The stress distribution in this winding, and
porating discrimination devices (circuitry the resulting discharge distribution, will dif-
intended t o isolate and identify corona dis- fer from that of an externally applied voltage,
charges) does in fact discriminate between the where the entire winding is at a common po-
minimum level specified for discharges in the tential. However, the revised distribution may
unit under test and the interference otherwise better represent actual operating conditions.
present in the test system. Figure 3 shows the modified connections to the
transformer under test required for the circuit
6. Apparatus of Fig 2; the circuit of Fig 1 is not suited for
self-excitation tests. The remainder of the cir-
The equipment required for quantitative cuit in Fig 2 and the apparatus and test proce-
corona (partial discharge) measurements typ- dure described below continue to apply.
ically includes the following: a test voltage
supply, a coupling capacitor, a corona detector 6.2 Coupling Capacitor. The coupling capaci-
(frequently a single instrument combining tor isolates the detector from the test voltage.
an amplifier and an oscilloscope), a coupling Other isolation, such as a transformer or a
impedance, a calibration pulse generator, cal- parallel-T RC filter network (as shown in
ibration coupling circuitry that often includes Fig 4) may be used, subject to the requirement
a calibration capacitor, and the transformer that the initial division of both the calibration
under test. Figures 1 and 2 show two of the sev- and the corona pulse voltages through the
eral corona discharge measurement circuits circuit is determined by t h e circuit
that can be assembled using this equipment. capacitances. The capacitance value i s
The choice of test circuit involves such factors selected in relation t o other circuit components
a s the size of the test transformer, whether it to realize the desired circuit sensitivity. In
should be isolated from or connected to ground, general, increasing the capacitance improves
minimizing stray capacitance in the lead circuit sensitivity. A value not exceeding 100-
wires, and the resultant detection sensitivity. 200 pF is often satisfactory a t 60 Hz; lower
capacitance values are necessary a t higher
6.1 Test-Voltage Supply. The test-voltage sup- frequencies.
ply is normally variable and should be ca-
pable of supplying a higher voltage stress than
that required for corona inception. When ac 6.3 Coupling Impedance. The coupling
voltage is used, there should be less than 5% impedance presents the discharge pulse to the
overall harmonic distortion when supplying input of the pulse amplifier. It may be resistive
the test load. The voltage waveform need not be or inductive. Its selection will depend on the
purely sinusoidal, but it should contain no in- test frequency and the configuration of the
termediate peaks. It should be measured with a other parts of the detection system. It is con-
peak-to-peak voltmeter. The series inductor nected to the corona detector so that it appears
shown in the test-voltage supply connections of across the input terminals to the detector.
Figs 1 and 2 should have a self-resonant fre-
quency of approximately 1 0 MHz. Electro- 6.4 Corona Detector. The corona detector usu-
magnetic filtering may also be required in ally combines an amplifier and a signal
order to eliminate voltage supply noise. readout instrument such as an oscilloscope.

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 IEEE
MEASUREMENTS ON ELECTRONICS TRANSFORMERS 9td 436-1991

CALIBRATION
PULSE
GENERATOR
-
-
Fig 1
Typical Circuit for Corona Measurement,
c3lvuit1

TRANSFORMER
E UNDER

i--L-+- B
CORONA
DETECTOR

PULSE

Fig 2
Typical Circuit for Corona Measmment,
cilulit 2
LEGEND (FIGS 1 AND 2)
C
, = coupling capacitor
C, = calibration coupling capacitor
Z, = coupling impedance
E = high-impedance test-voltage supply

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 IEEE
Std 4361991 E E E GUIDE FOR MAKING CORONA PARTIAL DISCHARGE)

--- e
I I
I
-__----_______-__-_____---- I

TRANSFORMER UNDER TEST

B
E = LOW-IMPEDANCE TEST-
VOLTAGE SUPPLY

Fig3
Circuit Modificationsfor Self-Excitation

Reference: L. Stanton, “Theory and Application of Parallel-T Resistance-Capacitance Frequency-Selective Networks,’

Proceedings of the IRE, July 1946, p. 447.

Fig4
Parallel-TRC Filter Network

6.4.1 Amplifier. The amplifier presents the 6.4.2 Signal Readout Instrument(s).An
discharge pulse to the readout equipment at an oscilloscope may be used t o display the signals
appropriate level. The band-pass characteris- from the amplifier. It is well suited for detect-
tics should be selected t o provide maximum ing individual discharge pulses and their
attenuation of the test-voltage frequency and phase occurrence with respect to the test volt-
its harmonics as well as any other unwanted age, Frequently an oscilloscope is the primary
signals. A parallel-T RC filter network may readout instrument with one or more other de-
be used t o eliminate the test-voltage frequency vices operated in conjunction with i t when
from the amplifier input while still transmit- more data is required. Such other devices in-
ting the corona discharge pulses accurately. clude peak-to-peak reading voltmeters,

10

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 IEm
MEASUREMENTS ON EJXCTRONICS TRANSFORMERS Std 436-1991

graphic recorders, and pulse counters; the lat- v

ter are particularly useful when pulse height I=2CfV,
analysis can be used t o define the discharge where C is in Farads.
fin Hertz,
pulse spectrum in terms of pulse amplitude VW in Volts,
and repetition rate. I in Amperes.

6.5 Calibration Pulse Generator. The calibra-

tion pulse generator is the source of a pulse
that, when properly coupled to the terminals of
the apparatus under test, produces a response
in the detection equipment that approximates a
disc5arge pulse response in magnitude. The
pulse should have a rise time of no greater
than 0.1 ps and a decay to half peak of no
greater than 1ms. The output impedance of the
generator is in series with the calibration cou-
pling capacitance; it must be small, so that the
amplitude of the output pulse can be deter-
mined t o an accuracy of at least f 3%, and it (Voltage Drops in Diodes and Meter << Peak-to-Peak Voltage.)
must be known, so that the combined im-
pedance in parallel with the coupling impe- Fig 5
dance can be calculated. A typical square- Peak-to-PeakvoltmeterCircuit
wave generator with an output impedance of
50 R satisfies these various requirements.

6.6 Calibration Coupling Capacitor. The cali- 7. Calibration

bration coupling capacitor couples the calibra-
tion pulse generator to the insulation system 7.1 Deflection Sensitivity. The tesk circuit is
under test (normally to the terminals of the calibrated t o determine the deflection sensi-
transformer). The value of the calibration tivity of the corona detgctor in terms of dis-
coupling capacitance should be small with re- charge pulse voltage amplitude or terminal
spect to the internal capacitance of the appara- corona charge. The deflection sensitivity is
tus under test. Usually 200 pF maximum is influenced by the specimen capacitance, stray
suitable at 60 Hz, while lower values are capacitance, and the amplifier gain. Since
appropriate at higher frequencies. The actual these factors may vary from specimen to spec-
capacitance value of the calibration capacitor imen, the entire circuit should be calibrated as
should be known t o an accuracy equal to the part of the test procedure and also whenever
accuracy of the measurement of the calibra- any of the influencing factors are signifi-
tion signal generator pulse. When discharge cantly changed. The amplifier gain should
signal magnitude measurements are t o be not be changed between calibration and test.
made during testing, the calibration coupling
capacitor should have a voltage rating suitable 7.2 When Calibration Is Performed. Nor-
for isolation of the calibration equipment. mally, the calibration is performed with the
specimen de-energized either before or after
6.7 Control Metering. Control metering should applying test voltage. When it is desired to
have an accuracy o f f 3%. Metering for an ac make signal magnitude measurements dur-
test voltage source should be a peak-to-peak ing testing, the calibration equipment may be
voltmeter (as shown, for example, in Fig 5 ) . switched in. These calibration measurements
This measurement circuit should be cali- should be verified at zero test voltage.
brated with sine-wave rms so that the test volt-
age may be specified in rms units. When the 7.3 Calibration Technique. To calibrate, in-
test voltage is induced through self-excitation, troduce a pulse of known amplitude from the
the peak-to-peak test voltage should be mea- calibration pulse generator into the circuit a t
sured directly. the terminals of the specimen, that is, the point

11

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 IEEE
Std 4361991 IEEE GUIDE FOR MAKING CORONA PARTIAL DISCHARGE)

at which the discharge pulse is to be detected. circuit bias or t h e result of required

Observe the signal magnitude on the readout connections. Test voltage may be either ap-
device with the amplifier gain set at the level plied directly or induced across the insulation
required for observation of the actual corona t o be investigated, depending on re-
discharges. Measure the maximum deflection quir ements.
caused by the calibrating pulse. From this
measurement and from the constants of the 8.3 Group III. Group I11 includes transformers
circuit, calculate the deflection sensitivity of with induced voltage stresses within wind-
the readout device to the discharge pulses. ing(s), from portions of winding(s1 t o other
(See 9.2.) winding(s) o r circuit elements, or any combi-
nation of these. Test voltages are normally
7.4 Calculating Detection Sensitivity.To sim- induced by transformer action within the
plify the calculation of detection sensitivity, transformer under test, by autotransformer
the terminal corona pulse voltage (that is, at action, or by direct application across the
the terminals of the transformer under test) winding under test. Normally, a point in each
and the calibration pulse voltage are both con- winding is grounded or connected to a com-
sidered to be step functions. The initial divi- mon reference. When the test is made on a
sion of voltage throughout the circuit is winding which does not have restrictions as in
considered t o be determined by the circuit 8.2 above, the test should be repeated with the
capacitances. ASTM D1868-1981 [ll derives high voltage and ground connections reversed
sensitivity equations for several acceptable in order to test both ends of the windings and
calibration circuits. terminations.

8. Test Method 9. Test Requirements

The test voltage should stress the insulation Unless otherwise specified, the following
as it will be stressed in normal operation in- requirements should apply.
cluding interwinding, intrawinding, bias,
and termination stresses. Because it is often 9.1 Ambient Conditions. Corona (partial dis-
not practical to test transformers in the actual charge) tests should be performed at 25 k 5 "C,
circuit in which they are designed to operate, atmospheric pressure of 101 kPa k 20%, and
test facilities should simulate operating condi- relative humidity of 70% maximum. Trans-
tions a s nearly as possible. To accomplish formers should be conditioned by remaining
this, electronics transformers and inductors in the ambient conditions for at least 8 h prior
may be classified in three groups. Complete to testing.
evaluation will sometimes require more than
one type of voltage application. 9.2 Calibration Sensitivity. In Appendix B, it
is shown that the meaningful quantity to be
8.1 Group I. Group I includes transformers measured is the energy of discharge. W is in
that are fclly insulated for all applied or nanojoules when Vi is in kilovolts peak and Q
induced voltages normally encountered in is in picocoulombs as shown below:
operation and in which insulation between
windings or from winding(s) t o ground, or W = 1/2 Q, Vi .
both, is to be tested for corona discharges. The Test-set sensitivity is usually expressed in
test voltage is applied across the insula- picocoulombs and voltage in rms units.
tion that is stressed in normal operation. Therefore,
This test usually does not include induced
voltages. Qu = G W I V,
8.2 G r o u p 11. Group I1 includes trans- where Q, is the calibration sensitivity in
formers that operate with specified voltage picocoulombs, W is the discharge pulse energy
limits, either a c o r dc, across speci- in nanojoules, and Vu is the applied voltage in
fied insulation. The limits may be external kilovolts rms.

12

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 IEEE
MEASUREMENTS ON ELECTRONICS TRANSFORMERS Std 436-1991

9.3 Test Voltage. Any convenient frequency [B41 Goldman, M. and Sigmond, R. S.
from 50 to 2000 Hz inclusive may be used for ac “Corona and Insulation,” IEEE Transactions
tests; however, the caution in 5.2.1 should be on Electrical Insulation, vol. EI-17, no. 2, 1982,
noted. Induced voltage tests should be p. 90.
conducted at a frequency that will not result in
core saturation or other forms of wave [B5] Harrold, R. T. and Dakin, T. W. “The
distortion. The rms test voltage should be 120% Relationship Between the Picocoulomb and
of the rms working voltage. Microvolt for Corona Measurements on HV
9.3.1 DC Plus AC Test Voltage. When the Transformers and Other Apparatus.” IEEE
test voltage is dc plus ac, the dc components Transactions on Power Apparatus and
should equal the rated dc bias or 50% of the Systems, vol. PAS-92,1973, p. 187.
peak working voltage, whichever is greater.
The ac component should be such as t o cause [BSI Krueger, F. H. Discharge Detection in
120% of the peak working voltage stress across High Voltage Equipment. New York:
the insulation under test. American Elsevier Publishing Co., 1965.

9.4 Rate of Application. The test voltage should [B71 Lalli, V. R., Mueller, L. A., and Koutnik,
be raised uniformly from zero to 50% of the E. A. “System Reliability Analysis through
maximum in not less than 5 s. From 50% to the Corona Testing.” ZEEE Power Electronics
maximum, the rate of rise should not exceed Specialists Conference 1975 Record, p. 51.
500 VIS.
[B81 Mason, J. “Discharges.” I E E E
9.6 Duration of Tests. The test stress should be Transactions on Electrical Insulation, vol.
maintained for a minimum of 1 min or 10 000 EI-13, no. 4,1978, p. 211.
cycles, whichever comes first, but not for less
than 18 s. [B91 Okamoto, H., Kanazashi, M., and
Tanaka, T. “Deterioration of Insulating
Materials by Internal Discharge.” I E E E
10. Bibliography Transactions on Power Apparatus and
Systems, vol. PAS-96, no. 1,1977, p. 166.

[Bll R. Bartnikas, R. and McMahon, E. [BlOI Parker, R. D. “Corona Testing of High

Engineering Dielectrics, Volume. 1: Corona Voltage Airborne Magnetics.” ZEEE Power
Measurement and Interpretation, American Electronics Specialists Conference 1975
Society for Testing Materials, 1979. Record, p. 43.

[B21 Dakin, T. W., “Corona Discharges and [Bill Schafer, B. “A Diagnostic Journey: High
T h e i r Effects on Insulation,” I E E E Voltage Insulation Systems.” Z E E E
Proceedings of the 1962 Electrical Insulation Proceedings of the 16th Electrical /Electronics
Conference, p. 87. Insulation Conference, 1983, p. 295.

CB31 Dunbar, W. G., “Interpretation of Corona [B121 Zwass, S. “Testing of High Voltage
Measurements on High-Voltage Assemblies,” Components and Systems.” IEEE Power
Electronic Packaging and Production, Electronics Specialists Conference 1975
Cahners Publishing Co., Jan. 1981, p. 139. Record, p. 38.

13

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 EEE
std 436-1991 IEEE GUIDE FOR MAKING CORONA (PARTIAL DISCHARGE)

Appendixes
(These Appendixes are not part of IEEE Std 436-1991, IEEE Guide for Making Corona (Partial Discharge) Measurements
on Electronics Transformers, but are included for information only.)

AppendixA
Preconditioningfor Determinationof CIV
In some insulation systems employing high rate until continuous corona discharges are
resistivity materials, such as epoxies, the detected. This will be the CIV.
initial observation may not be representative In liquids, on the other hand, a corona
of operating experience due t o the time extinction voltage (CEV)lower than the CIV
dependency of the onset of corona discharges. may be the result of liquid breakdown forming
One way t o determine the corona inception gas bubbles that continue t o allow discharges
voltage (CIV) quickly and reliably is to raise well below the stress level at which they were
the test voltage to some predetermined higher formed. In this case, preconditioning cannot
voltage for a short time (the transformer may be used and, in fact, may be injurious to the
be well into the partial discharge range but system.
below dielectric flashover). The test voltage is When preconditioning voltage is to be used,
then reduced t o zero and immediately the magnitude and time of application should
reapplied, increasing at a specified uniform be specified.

14

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 IEEE
MEASUREMENTS ON ELECTRONICS TRANSFORMERS 9td 436-1991

AppedB
TheMagnitudeandEnergyofDischarges
When there is a discharge, it occurs as if the small capacitor C, in the model circuit was
instantaneously short-circuited. A charge transfer occurs, given by

1 1
T
va
v,
= peak applied voltage
= vi = CIV
Va C a Cb Qa = apparent discharge magnitude
detected at the terminals
T

Simultaneously, a voltage pulse, which is effectively a step voltage having a rise time of between
10 and 100 ns, is generated at the terminals of the insulation.

sva = sv, - cb
ca + c b '

The apparent discharge magnitude, observed at the terminals, is

Qa = 6Va( Ca +- c bcc
c b + cc ) = SV,
ca c b +cac c + c b c c
c b +cc

Let
G = cacb +cacc + c b c c .
Usually, a small area of the discharge site is almost completely discharged, so that

sv, = v, = cb
v,: -
c b +cc

Qc
- SVc
------- C3 cb+Cc SVa Ca+Cb
Qa sva ca+cb c3 mc c b

= 1+-.C C
c b

If it is assumed that most of the charge is liberated from the region where SV, -+ Vc , the energy
liberated will be

W = 1f2QCV, = lf2Qa ( 1 + %)Vi -- If2 &,Vi

1, + CC
(Eq B7)
L'b

where W is in nanojoules, Vi is in kilovolts peak, and Q, is in picocoulombs.

Reference
Mason, J. H., Proceedings of the Institution of Electrical Engineers, vol. 112, no. 7, July 1965.

15

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.
 AppendixC
Suggested speciscationRequ”ents
The specification of requirements other than Class I. For insulation systems with a
the basic ones listed herein may require history of satisfactory life a t the voltage
special test equipment. Requirements that in- stress used-70 n J maximum.
clude the use of dc voltages, specific frequen- Class ZI. For insulation with considerable
cies, wave shapes other than sinusoidal, mea- tolerance for discharges such as inorganic
surement of discharge pulse energy a t spec- insulation-700 n J maximum.
ified voltage($, o r special ambient conditions Class III. Pulse energy level to be specified.
should be negotiated.

C3. Other Information

CL Normal Requirement The specification of other information shall
be considered special and should be negoti-
The normal requirement is that no continu-
ated. Examples of such information are:
ous discharges shall occur a t the specified
Corona inception voltage (CIV).
pulse energy level and at the ac test voltage in
Conditioning in accordance with
9.3.
Appendix A must be defined.
Corona extinction voltage (CEV).
Pulse energy of maximum discharge
C2. Pulse Energy Level Classes pulses at specified voltage(s).
Number of discharge pulses of given
The pulse energy levels shall be selected pulse energy at specified voltage(s) per
from the following classes. In selecting the unit of time.
class, consideration must be given t o the fact Pulse height analysis techniques are particu-
that calibration sensitivity depends on test larly useful in providing the last two items of
voltage level. information.

16

Authorized licensed use limited to: Universidad de chile. Downloaded on October 16,2020 at 19:51:04 UTC from IEEE Xplore. Restrictions apply.