Ieee STD 62 - Factor de Disipacion

IEEE Std 62-1978
(Revision of IEEE Std 62-1958)
IEEE Guide for Field Testing

Power Apparatus Insulation
ublished by The Institute of Electrical and Electronics Engineers, Inc 345 East 47th Street, New York, N Y 10017, USA
cember 1 4 , 1978 SH07203
Authorized licensed use limited to: Interconexin Elctrica ISA. Downloaded on November 27,2017 at 14:45:36 UTC from IEEE Xplore. Restrictions apply.
IEEE
Std 62-1978
(Revision of
IEEE Std 62-1958)

Sponsor
Power System Instrumentation and Measurement Committee
of the
IEEE Power Engineering Society
Copyright 1978 by
The Institute of Electrical and Electronics Engineers, Inc
N o part o f this publication may be reproduced in any form,

in an electronic retrieval system or otherwise,
without the prior written permission of the publisher.
IEEE Standards documents are developed within the Technical Com-
mittees of the IEEE Societies and the Standards Coordinating Commit-
tees of the IEEE Standards Board. Members of the committees serve
voluntarily and without compensation, They are not necessarily mem-
bers of the Institute. The standards developed within IEEE represent
a consensus of the broad expertise on the subject within the Institute
as well as those activities outside of IEEE which have expressed an in-
terest 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 to pro-
duce, test, measure, purchase, market, or provide other goods and ser-
vices related t o the scope of the IEEE Standard. Furthermore, the view-
point expressed a t the time a standard is approved and issued is subject
t o change brought about through developments in the state of the art
and comments received from users of the standard. Every IEEE Stan-
dard is subjected t o review at least once every five years for revision
or reaffirmation. When a document is more than five years old, and has
not been reaffirmed, it is reasonable t o conclude that its contents,
although still of some value, d o not wholly reflect the present state of
the art. Users are cautioned t o check t o 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. Sug-
gestions for changes in documents should be in the form of a proposed
change of text, together with appropriate supporting comments.
Interpretations: Occasionally questions may arise regarding the mean-
ing of portions of standards as they relate t o specific applications. When
the need for interpretations is brought t o the attention of IEEE, the
Institute will initiate action t o prepare appropriate responses. Since
IEEE Standards represent a consensus of all concerned interests, it is
important t o ensure that any interpretation has also received the con-
currence of a balance of interests. For this reason IEEE and the mem-
bers of its technical committees are not able t o provide an instant re-
sponse to interpretation requests except in those cases where the matter
has previously received formal consideration.
Comments on standards and requests for interpretations should be ad-
dressed to:
Secretary, IEEE Standards Board
345 East 47th Street
New York, NY 10017
USA
Foreword
(This Foreword is n o t a part of IEEE Std 62-1978, Guide for Field Testing Power Apparatus Insulation,)
Factory tests assure the initial quality and condition of power apparatus insulation. During ship-
ment, installation, and service, the apparatus is subject to influences that may affect the insulation
and shorten its useful life. As a result, apparatus insulation is tested in the field from time to time
t o determine its suitability for continued service and to detect deterioration which often can be
checked or corrected by suitable maintenance procedures.
The variety of test methods and equipment used for the assessment of insulation quality neces-
sitates the restriction of this guide to a general description of the more commonly used methods.
An attempt has been made to supply sufficient information to identify those tests more applicable
to a given situation and to provide references to a more detailed coverage.
This guide was first published in April 1958 as AIEE Std 62, Recommended Guide for Making
Dielectric Measurements in the Field. This revision is the first since that time. Withstand tests have
been given greater coverage and the material on measurement of insulation characteristics reflects
experience gained in the intervening years.
This revision has been prepared by a Working Group under the sponsorship of the High Voltage
Testing Techniques Subcommittee of the IEEE Power Systems Instrumentation and Measurements
Committee. The assistance of A. F. Rohlfs, Chairman, and other members of the High Voltage Test-
ing Techniques Subcommittee, is gratefully acknowledged.
At the time of approval of this standard the membership of the Working Group was as follows:
E. H. Povey, Chairman J. R . Johnston, Secretary
L. Bucklew 0. Petersens
E. B. Curdts E. F. Rodia
H. E. Foelker B. F. Slingduff
J . T. LaForte G . M. L. Sommerman
A. L. McKean L. R. Sharpe
Approved September 8, 1977

IEEE Standards Board
William R. Kruesi, Chairman Irvin N. Howell, Jr., Vice Chairman
Ivan G. Easton, Secretary
William E. Andrus R. 0. Duncan Donald T. Michael

Jean Jacques Archambault Charles W. Flint Voss A, Moore
Mark Barber Jay Forster William S. Morgan
Edward J. Cohen Ralph I. Hauser William J. Neiswender
Warren H. Cook Joseph L. Koepfinger Ralph M. Showers
Louis Costrell Irving Kolodny Robert A. Soderman
R. L. Curtis Benjamin J. Leon Leonard W. Thomas, Sr.
David B. Dobson Thomas J. Martin B. W. Whittington
Contents
SECTION PAGE
1. Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 . Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1 Personnel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3. Withstand Voltage Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.2 Significance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.3 Elementary Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.4 Comparison of Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.5 Environmental Influences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.6 TestEquipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4 . Potential-Distribution Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.2 Significance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.3 Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.4 Comparison of Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5 . Partial-Discharge (Corona) Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.2 Significance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
5.3 Measuring Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
6 . Measurement of Dielectric Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
6.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
6.2 Significance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
6.3 Insulation Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
6.4 Polarization Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
6.5 Capacitance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
6.6 Dielectric Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
6.7 Power Factor, Dissipation Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
6.9 TestEquipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
7 . Tests on Specific Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
7.1 Rotating Machinery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
7.2 Transformers, Regulators, and Reactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
7.3 Insulated Conductors (Cables). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
7.4 Switchge ar. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
7.5 Lightning Arresters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
7.6 Capacitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
7.7 Insulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
7.8 Bushings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
7.9 Insulating Fluids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
8. Applicable Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
8.1 Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
8-.2 Withstand Voltage Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
8.3 Partial-Discharge (Corona) Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
8.4 Dielectric Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
8.5 Rotating Machinery Insulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
SECTION PAGE
8.6 Transformer Insulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
8.7 Conductor (Cable) Insulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
8.8 Capacitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
8.9 Liquid Insulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
FIGURES
Fig 1 Effect of Time on Withstand Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Fig 2 Effect of Absorption on Measured Insulation Resistance .......................... 12
Fig 3 Effect at Frequency on Loss Factor and Dielectric Constant ....................... 12
APPENDIXES
Appendix A . Shielding of Equipment for the Measurement of Insulation Characteristics. . . . . . . .24
A 1 . General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
A2 . Basic Shielding Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Appendix B. Measurement Techniques for the Analysis of Complex Insulation Systems . . . . . . . . 28
B1 . Simple and Complex Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
B2 . Determination of the Characteristics of an Individual Component . . . . . . . . . . . . . . . . . . . . 28
B3 . MethodA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
B4 . MethodB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
B5 . MethodC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
APPENDIX FIGURES
Fig A1 Elementary Measuring Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Fig A2 Grounded-Guard Shield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Fig A 3 Hot-Guard Shield. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Fig A4 Cold-Guard Shield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Fig A5 Double-Shielded Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Fig A6 Shielded Bridge Circuit with Guard Balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
APPENDIX TABLES
TableB1 . MethodA .......................................................... 29
TableB2 . MethodB .......................................................... 30
TableB3. MethodC .......................................................... 31
1. Scope rect test voltage, provisions for grounding the

circuit until all stored charges are dissipated.
This guide discusses the significance of various 2.1.4 Warning Signs and Barriers. The test
types of tests commonly employed t o evaluate area may be marked off with signs and easily
the insulation of power apparatus in the field, visible tape. Warning signs should conform to
where environmental conditions cannot readily the requirements of governing bodies such as
be controlled; the equipment and techniques the Occupational Safety and Health Adminis-
required for each type of test; problems peculiar tration (OSHA) in the United States.
t o field testing, and methods of dealing with 2.2 Apparatus
these problems; and the application of field 2.2.1 Consequences of Failure. When the test
tests t o specific types of power apparatus. voltage to be applied t o apparatus insulation
exceeds the normal operating value, there
exists the possibility of failure under test. Be-
2. Safety fore applying the test, consideration should be
given to the time, material, and labor required
2.1 Personnel for a possible repair. If failure could result in
2.1.1 Hazards. Insulation tests in the field fire, fire-fighting equipment should be available.
can present a hazard t o personnel unless suit- 2.2.2 Overvoltage. In the conduct of high-
able precautions are taken. Apparatus or cir- voltage tests the voltage may accidently exceed
cuits t o be made available t o test personnel the desired maximum. A sphere gap, adjusted
must be disconnected from the power system. t o spark over a t a voltage slightly above the de-
Typical safety procedures call for a visual sired maximum, may be connected across the
check of the disconnection or, when this is not voltage source. By selecting the proper value of
possible, a check with a voltage indicator. series resistor, the gap may be used t o provide a
Grounds are then applied. Personnel are usually warning signal, to inhibit further rise in the test
instructed to treat all ungrounded apparatus as voltage, or to activate an overcurrent circuit
energized. breaker in the power supply circuit.
2.1.1 Ground Connection. Grounds may be 2.2.3 Graded Insulation. When an alternating
removed to permit application of test voltage. test voltage is applied t o a short-circuited
When feasible, it is preferable t o retain the winding, or when a direct test voltage is used,
ground connection during the tests and to pro- the insulation of the entire winding is subject
vide between the ground and the apparatus t o this test voltage. When the insulation level of
under test a second disconnection of sufficient the winding is graded from one end to the
clearance for the test voltage (see 6.8.3.2). other, the magnitude of the applied test volt-
2.1.3 Precautions. When the test voltage is age should correspond to the lowest insulation
over a few volts, precautions should be taken level.
t o prevent personnel from contacting the ener-
gized circuit. An observer may be stationed to
warn approaching personnel and may be sup- 3. Withstand Voltage Tests
plied with means t o deenergize the circuit. The
means may include a switch t o shut off the 3.1 Description
power source and, particularly in case of a di- 3.1.1 General. A withstand voltage test sub-
IEEE
POWER APPARATUS INSULATION Std 62-1978
jects an insulation for a restricted period of ing atom of molecule. When the gradient is
time to a voltage stress greater than that en- strengthened beyond a critical value, called the
countered under normal service conditions. intrinsic dielelectric strength of the material,
The magnitude of the test voltage used in the collisions occur with sufficient impact t o free
field is generally based on the factory accept- more electrons than are captured, and disrup-
ance test voltage, reduced by a factor t o allow tive breakdown follows.
for the effects of transportation, installation, 3.3.2 Practical Dielectric Strength. Most in-
and in-service degradation. sulating materials are not entirely homogeneous
3.1.2 Type of Voltage. The voltage specified and so contain some regions having less than
for a withstand test may be alternating at average dielectric strengths. Under test there
power frequency or some other frequency, will be a tendency for breakdown t o occur in
may be direct (having a specified polarity in a path which includes regions of lower dielec-
respect t o ground), may be a pulse having a tric strength. The actual dielectric strength
specified polarity and wave shape, or may be a of an insulating material in practice is usually
pulse-generated transient. substantially lower than its intrinsic strength.
3.1.3 Duration of Voltage Application. For 3.3.3 Time and Temperature Effects. Con-
alternating or direct test voltages, the duration tinued application of a voltage gradient below
of application is specified, commonly 1 min the value required to cause immediate break-
after the desired test value has been reached. down will cause temperature rises in local weak
Voltage applications are made at a prescribed regions. The temperature rise further reduces
rate or in scheduled increments. the dielectric strength of these regions, and
3.1.4 Impulse Voltage. For an impulse, the consequently the overall dielectric strength of
voltage-time relation (wave shape) is specified. the material, and may result in eventual failure
A standard lightning impulse wave is denoted without further increase in the applied voltage.
by two figures: 1.2/50. The first figure is the As an example of the time dependence in with-
virtual rise time in microseconds of the pulse; stand testing, Fig 1 illustrates the general be-
the second is the virtual time t o half value in havior of oil-impregnated pressboard in terms
microseconds. of withstand voltage versus time of voltage
3.1.5 Pulse-Generated Voltage. For pulse- application.
generated transients, the peak voltage and os- Materials which are normally lossy under
cillation frequency are specified. If repetitive, voltage stress are subject t o destruction from
the repetition rat.e is also specified.
3.2 Significance. The withstand voltage test is Fig 1
a demonstration that an insulation can with- Effect of Time o n
stand a specified overvoltage for a specified Withstand Voltage
length of time and is essentially a go-no-go
type of test. Successful completion of the test
I,u sec I sec I hour I year
gives some assurance that no gross defect is I I I I
present in the insulation structure. However, I
the ability of a withstand voltage test to expose I I
I 1
incipient faults is less certain. For this reason
the test is often supplemented by measure-
ments of insulation characteristics, made dur-
ing or at the completion of the withstand test.
3.3 Elementary Theory
3.3.1 Intrinsic Dielectric Strength. As op-
posed to a conductor, in which electrons move
with little restriction, an insulator is a material I I
in which t h e electrons are tightly bound t o
atoms or molecules. When an insulator is sub-
jected to a moderate potential gradient, some 10- IO0 lo4 IO8
electrons may be pulled free from their bonds, TIME, SECONDS
but are recaptured in collision with a neighbor-
JEEE
Std 62-1978 IEEE GUIDE FOR FIELD TESTING
thermal runaway. Runaway occurs when the voltage of very low frequency (for example,
applied stress is high enough t o generate more 0.1 Hz) t o obtain the advantage of reduced
heat in the material than can be dissipated into weight and reduced partial discharge damage
the environment. The voltage at which termal while providing an approximately normal stress
runaway occurs is therefore not only a func- pattern. The weight of very low frequency test
tion of the material but also of the effective- equipment is less than the usual power-fre-
ness of provisions for dissipating the generated quency test equipment and approximately the
heat. same as power-frequency equipment when
resonant methods are employed.
3.4 Comparison of Tests
3.4.4 High-Frequency Withstand Test. With-
3.4.1 Power-Frequency Withstand Test. A
stand test voltages at higher than normal fre-
power-frequency withstand test is generally
quency are used for testing the insulation of
considered t o subject the insulation under test
t o a stress pattern most closely resembling that windings on magnetic cores, as in transformers
occurring in service. The disadvantages of the and reactors. The higher frequency permits an
power-frequency test are possible damage t o overvoltage to be applied across the winding
corona-sensitive insulation and the massive without the problem of core saturation. This
equipment necessary t o energize high-capac- type of test applies overvoltage on turn-to-
itance specimens. The latter disadvantage turn insulation as well as on the insulation
may be partially alleviated by the use of between winding and ground.
resonant methods. 3.4.5 Impulse Test. Impulse tests may be
3.4.2 Direct-Voltage Withstand Test. The made on insulation systems subject t o voltage
direct-voltage withstand method requires rela- surges in service. Such systems include turn-
tively light-weight equipment. Partial discharge to-turn insulation of coils or windings. Im-
damage is less than with power-frequency tests. pulse testing of apparatus insulation in the
There is the possibility, under certain condi- field is usually restricted t o low-voltage ap-
tions, of predicting and thus avoiding break- paratus because high-voltage impulse gen-
down of weak insulation. The prediction pro- erators are not readily portable.
cedure involves the application of the test volt- 3.5 Environmental Influences. The voltage
age in steps in accordance with a predetermined withstand capability of insulation struc-
schedule, with a measurement of current or tures designed for field service is affected to
resistance made at each step. some extent by normal variations in the field
A disadvantage of the direct-voltage with- environment. Low barometric pressure, surface
stand test for a composite insulation system dirt, and moisture are conditions which tend t o
designed t o operate under alternating-voltage lower surface flashover voltages and thus inter-
stress is that the test may not produce a fere with the conduct of a withstand test. In
normal stress pattern in the insulation. With general, however, the reduction in test voltage
time of voltage application, the direct-voltage levels allowed for field testing is usually suf-
stress pattern tends t o conform more to the ficient to permit tests t o be succesfully con-
resistive characteristics of a composite system ducted.
then t o the combined resistive and capacitive
3.6 Test Equipment. Commercial equipment is
characteristics which determine the alternating-
available for all types of withstand tests. Im-
voltage stress pattern. This disadvantage is
portant specifications which must be con-
particularly apparent for systems where resis-
sidered include:
tive-capacitive stress grading is employed, since
with an applied direct voltage the stress will (1)Type of voltage output - impulse, direct,
not be graded in the designed manner. Pro- alternating (include frequency)
longed direct-voltage withstand tests are not (2) Voltage range
recommended for oil-immersed equipment be- ( 3 ) Voltage control facilities including rate-
cause contaminating particles in the oil tend of-rise capability
t o line up and form a breakdown path. (4)Power output - real and reactive power
3.4.3 Very Low Frequency Withstand Tests. capability for ac tests, output current range for
Withstand tests on high-capacitance speci- dc tests, energy discharge capability for im-
mens may be made using an alternating test pulse tests
IEEE
(5) Power input requirements results should be made by comparison of read-

(6) Weight and portability ings on similar units with the device similarly
positioned, rather than on the absolute mea-
sured values.
4. Potential-Distribution Tests 4.3.5 Two-Terminal Devices. Two-terminal
devices permit measuring the potential between
4.1 Description. The potential-distribution test two conducting parts of the insulating unit,
method requires the measurement of potential one of which may be at ground potential. The
at or between various locations over an insulat- measuring circuit between the two terminals
ing unit. The method is most often applied t o includes the meter or gap-telephone system
the testing of ac transmission line insulators (see 4.3.3) and a high-voltage high-impedance
while in service. In the application of this element. A resistive element is usually used
method t o a string of suspension insulators, for with a meter system and a capacitive element
example, the potential across each insulation with a spark-gap system.
may be measured.
4.4 Comparison of Methods. Meter devices are
4.2 Significance. A normal potential distribu- difficult to read under adverse lighting condi-
tion is an indication that the insulating unit tions, while spark-gap devices are difficult to
is in satisfactory condition. Any section of a hear under noisy conditions. Voltage readings
unit across which the potential difference is are more easily made with the meter device.
lower than normal is considered to be in ques- The inconvenience of adjusting a gap for each
tionable condition. insulator is commonly avoided by establishing
a few fixed gap settings and using the setting
4.3 Equipment
most applicable t o the insulator under test.
4.3.1 Safety. Safety is a primary require-
The insulator passes or fails the test depending
ment for potential-distribution testing devices
on whether or not it supports sufficient voltage
t o be used on energized transmission systems.
t o cause the gap to discharge.
The device must be so designed that it can be
used without the possibility of creating a trans- 4.5 Environmental Influences. The voltage dis-
mission outage from a short circuit or acci- tribution over an insulating unit or structure
dental ground. can be influenced by surface conditions. For
4.3.2 High Impedance. A potential-distribu- comparison purposes voltage distribution tests
tion testing device must have high electrical should be made when surfaces are reasonably
impedance so that its use does not appreciably clean and dry.
change an existing potential distribution. The voltage distribution method may not be
4.3.3 Measurement of Potential. Measure- reliable for detecting cracks in porcelain in-
ments of potential may be obtained either as sulation after prolonged exposure t o low-
a meter reading or as a calibrated sphere-gap humidity atmosphere, particularly when the
setting. The sphere gap is adjusted to the maxi- operating voltage stress is low.
mum setting at which a continuing discharge
or sparking can be obtained. Discharge of the
gap is monitored by listening to a telephone
receiver in the gap circuit through a highly 5. Partial-Discharge (Corona) Tests
insulated stethoscope arrangement. A more re-
cent method of monitoring a discharge is 5.1 General. Partial discharges in an insulation
through the use of a remote detector respon- system are discharges that do not completely
sive t o electromagnetic radiations from the bridge the insulation between conductors.
discharge. One form of partial discharge is corona, a
4.3.4 One-Terminal Device. A one-terminal term which more specifically describes dis-
device has an exposed metallic tip for contact- charges which emanate from a bare conductor
ing some conducting part of the insulating unit, into the surrounding air or gas. Of more con-
that is, cap, pin, or cement layer. The measur- cern to insulation life are partial discharges
ing circuit is completed by capacitance to which occur in inadvertent gaseous inclusions
ground but is also influenced by capacitance (cavities) in otherwise solid insulation.
t o nearby energized parts. Interpretation of A discharge will occur in a cavity if the
IEEE
potential gradient exceeds a critical value. The Measurement of Energy and Integrated Charge
discharge effectively short-circuits the cavity, Transfer Due t o Partial Discharges).
dissipating stored energy into the surrounding 5.3.2.2 Power Factor. Using the preceding
insulation and causing a momentary increment assumption, the measured power factor of an
in the capacitance of the system. insulation should remain constant with increas-
ing test voltage if no discharges are present.
5.2 Significance
The observed power factor increase between a
5.2.1 Reduction of Insulation Life. Partial
voltage below the discharge inception and some
discharges in an insulation system at operating
higher test voltage has been used as a measure
voltage may result in a significant reduction in
of discharge activity at the higher voltage.
the life of the insulation. Some insulations are
5.3.3 Measurement of Individual Pulses
more susceptible t o this type of damage than
5.3.3.1 General. Measurement of individual
others.
partial-discharge pulses may be desirable for
5.2.2 Cavities. The presence of partial dis-
insulation systems which are vulnerable to
charges in an insulation system may be an
attack by partial discharges. Proper measure-
indication of the presence of cavities, particu-
ments of individual pulses can be made only
larly in composite or wound insulations.
when the pulses are sufficiently separated in
5.2.3 Electromagnetic Radiation. Partial dis-
time t o make superposition improbable. Indi-
charges in an insulation system may be a source
vidual pulse measurements may be applicable
of electromagnetic radiation which interferes
when the total discharge power loss is too
with communication systems.
small t o be accurately measured.
5.3 Measuring Methods 5.3.3.2 Test Circuit. The usual circuit
5.3.1 Test Voltage. Although some measure- for the measurement of individual discharge
ment of discharge activity is possible on ap- pulses in an insulation system includes an
paratus while connected t o the power system, impedance coupled t o the terminal of the
in general it is preferable t o disconnect the system through a filter which removes the
apparatus under test from the system and en- power frequency component of the volt-
ergize it by a test transformer. A separate test age. The pulse voltages appear across the im-
voltage supply minimizes interference from pedance where they can be observed by an
sources in the system and reduces the hazard oscilloscope or measured by a peak-reading
t o test personnel from accidental contact with voltmeter. The peak-reading voltmeter indi-
system voltage. An adjustable supply permits cates the maximum pulse.
observation of discharge activity as a function 5.3.3.3 Calibration Circuit. A calibration
of test voltage. The test transformer and its circuit including a pulse generator may be
connections must be free of partial discharge. added to the test circuit so that the detected
5.3.2 Power Loss pulses can be quantitatively evaluated. Oscillo-
5.3.2.1 General. Measurement of partial- scope deflections due t o pulses are usually
discharge power losses are usually based on the interpreted in terms of picocoulomb charges
asumption that normal insulation losses vary as at the system terminals. The peak-reading volt-
the square of the applied voltages. By measur- meter can also be calibrated in terms of charge.
ing the loss in an insulation (see 6.6) at some 5.3.3.4 RIV Method. Discharge pulses may
voltage below the discharge inception voltage be measured by the radio influence voltage
and calculating the loss at some higher test (RIV) method. The measuring circuit consists
voltage according to the voltage-squared rela- of a filter as described in 5.3.3.2, with the
tion, the partial-discharge power loss at the pulses appearing across a noninductive resistor.
higher test voltage is found as the difference A narrow band of high-frequency components
between the observed and calculated losses. of the pulses, usually centered about 1 MHz, is
The assumption is often valid, but in some amplified and measured. The measurement is
cases other nonlinear voltage effects may add commonly expressed as the peak (or quasi-
to the loss increase. peak) voltage of the highest recurrent pulse.
A special bridge circuit may be used to pre- Calibration is based on the sine-wave output
sent the discharge losses on an oscilloscope of a standard signal generator.
as the area of a pattern approximating a 5.3.4 Electromagnetic Field Methods, Probes.
parallelogram (see ASTM Method D 3382-75, Many insulation systems are not completely
10
IEEE
self-shielding, and in such systems partial dis- and materials employed.

charges will produce an external electromag- 6.1.2 Test Voltage. Measurements of dielec-
netic field. Crude estimates of discharge activ- tric characteristics are usually made at test
ity have been obtained from the response of voltages below the operating voltage of the
a radio receiver placed in the vicinity of such insulation system under test. The tests are
systems. therefore categorized as nondestructive. Mea-
A probe may be used to localize the par- surements at test voltages above operating
tial discharge sources in nonself-shielded sys- value are used only for special information
tems such as used t o insulate generator stator such as the prediction of failure during direct-
windings. A probe responsive to magnetic voltage withstand tests (see 3.4.2) and the eval-
fields may consist of a few turns of wire on a uation of partial-discharge activity (see 5.3.1).
ferrite core and be connected t o a meter 6.1.4 Extent of Measurement. The effect of a
through a radio-frequency amplifier. A probe local fault on the measured dielectric charac-
responsive t o electric fields may be a minia- teristics of an insulation system will be less
ture antenna, connected to a meter through pronounced when the total amount of insula-
an audio-frequency amplifier. For safety, a tion in the system is large. For this reason it is
probe should be well insulated and used on desirable t o confine a measurement t o a small
insulation systems only when energized by a amount of insulation. Test techniques have
test transformer. been established for the analysis of complex
insulation systems, by which the character-
5.4 Environmental Influences. Partial-discharge istics of individual components of the system
measurements in the field are generally made in
may be determined (see Appendix B).
the vicinity of other equipment energized at
high voltage. Partial discharges in the neigh- 6.2 Significance
boring equipment, due t o its pulse nature, may 6.2.1 Damage. Abnormal dielectric character-
cause serious interference in the measuring istics as determined from measurements on an
circuit through capacitive coupling effects. insulating structure may indicate damage to, or
Welding equipment, commutators, and radio a defect in, a component material, occurring
stations are other sources of interference. When during manufacture or from subsequent ex-
such interference is present, it may be desir- ternal causes.
able to schedule the test for a time when the 6.2.2 Deterioration. Abnormal dielectric
interfering equipment is shut down. characteristics may be an indication of de-
terioration in one or more of the component
materials in an insulation structure. Deteriora-
6. Measurement of tion may be the result of absorption of con-
Dielectric Characteristics taminants, especially moisture, or chemical
6.1 General. An insulating structure has normal reactions within or between materials, ac-
dielectric characteristics determined by its celerated by the temperatures and by electrical
and mechanical stresses which occur in service.
geometry and component materials. Dielectric
characteristics commonly measured include in- 6.2.3 Leakage. Abnormal dielectric character-
istics in an insulating structure may be caused
sulation resistance and polarization index using
by leakage over insulation surfaces which are
a direct test voltage, and capacitance, dielectric
exposed t o the deposition of moisture and
loss, power factor, or dissipation factor using
particulate matter from the environment. Test
an alternating test voltage.
results are sometimes used to indicate the
6.1.1 Normal Characteristics. The normal
necessity for cleaning the surfaces.
dielectric characteristics for a given insulating
structure may be established from measure- 6.3 Insulation Resistance
ments made on similar structures which subse- 6.3.1 Determination. Insulation resistance
quent experience has shown t o be in satisfac- may be determined as the ratio of a direct test
tory operating condition. Where sufficient data voltage applied t o an insulation, as measured
exist, statistical methods are generally used in by a voltmeter, to the resulting current through
arriving at a norm. Normal characteristics of the insulation, as measured by a current meter.
an insulating structure may often be closely Alternatively, the ratio may be determined by
approximated by a consideration of the design a single crossed-coil instrument. Bridge meth-
11
IEEE
I I I sorption characteristics are known. A PI lower

I I I than normal suggests excessive surface leakage
I I I or deteriorated insulation.
1 TRUE INSULATION RESISTANCE
6.5 Capacitance
m 6.5.1 General. Capacitance in an insulating
I
I RES I STANCE structure is present between any two conduct-
0
W
c3 ing elements separated by a dielectric (insulat-
I ing material). The capacitance value depends
on the geometry of the structure and on the
dielectric constant (permittivity) of the insula-
tion. The dielectric constant in turn depends
on the polarization processes in the insulation
I I and is usually a constant only at a given fre-
4 8 12 16
TIME, M I N U T E S quency. The effect of frequency on the value
of the dielectric constant is shown in Fig 3.
Fig 2 Even at a given frequency the constant may
Effect of Absorption o n vary with temperature and is influenced by the
Measured Insulation Resistance deterioration factors mentioned in 6.2.2.
6.5.2 Use. Capacitance measurements, while
ods may also be used t o measure insulation sometimes useful as an indication of insulation
resistance by comparison with resistors of deterioration, provide other valuable types of
known value. information. For example, abnormal capac-
6.3.2 Time Dependence. Most high-voltage itance values may indicate incorrect test con-
insulation exhibits dielectric absorption due t o nections, floating shields or other breaks in
polarization processes which require time t o conducting elements, or short-circuited por-
tions of the insulation. Capacitance measure-
complete. Because of absorption the measured
ments are thus particularly important for in-
insulation resistance appears t o increase with
sulation structures which are composed of a
the time of voltage application, approaching
series of capacitor units such as coupling capac-
a steady value, which is the true insulation
itors or condenser-type bushings.
resistance. In practice, the resistance measured
after a 1 min application of voltage is often
considered the insulation resistance. Fig 3
6.3.3 Example of Time Dependence. As an Effect of Frequency on Loss Factor
-
example of the time dependence of insulation and Dielectric Constant
resistance measurements, Fig 2 shows the
change in measured resistance with time after
applying a constant direct test voltage to an -
*
4
ATOMC
ELECTRONIC
I -7
DIPOLE---------------CI I
I
'
insulation specimen which exhibits absorption. -INTERFACIAL-
The horizontal line indicates the true insulation
resistance which the measured resistance is ap-
proaching.
6.4 Polarization Index
6.4.1 Determination. The polarization index
(PI) of an insulation structure is found from
insulation resistance measurements made at 1
and 10 min after application of the test volt-
age. The index is the ratio of the 1 0 min to L
the 1 min resistance value and is almost always DIPOLE

greater than unity. The index is convenient for 0 ATOMIC'
comparison purposes since it has no dimension. 9 I
ln
6.4.2 Use. The polarization index is useful ln
0 INFRARED VISIBLE
in the determination of the condition of in- * ELECTRICAL-&OPTICAL-

FREOUENCIES FREQUENCIES
sulating structures for which the normal ab-
12
IEEE
6.5.3 Determination. Capacitance is com- frequency, and for power apparatus insulation
monly determined using an alternating test it is preferable to use a frequency at or near the
voltage and bridge techniques or charging cur- power frequency.
rent measurements. 6.7.2 Definition. The power factor and dissi-
pation factor are the casine and cotangent,
6.6 Dielectric Loss respectively, of the phase angle between the
6.6.1 General. Measurable dielectric losses
applied test voltage and the resulting current
occur in all solid and liquid insulating ma-
through the insulation. Correspondence be-
terials, and in general, good insulating ma-
tween these factors may be found by reference
terials have low losses.. Increased losses are
to trigonometric tables. For most insulations
usually caused by deterioration factors such
these factors fall in the range of 0 to 0.1 (or
as mentioned in 6.2.2.
0 t o 10 percent). In this range the two factors
6.6.2 Ionization. Dielectric losses in gaseous
are often used interchangeably, since they dif-
insulations are insignificant unless the applied
fer by less than 0.005 (0.5 percent).
voltage stress reaches or exceeds the critical
6.7.3 Interpretation. Power factor, or dissipa-
value at which ionization of the gas is initiated.
tion factor, is the most widely used altemating-
A possible source of ionization losses occurs in
current characteristics for assessing the condi-
nominally solid insulation structures due t o
tion of insulation. The interpretation of insula-
inadvertent gas inclusions in the structure.
tion condition primarily depends on compari-
6.6.3 Polarization. A substantial portion of
son of a measured value with a previously es-
the normal dielectric losses in solid and liquid
tablished normal value. Exceptions are insulat-
materials under alternating voltage stress is
ing structures of very low capacitance, such as
due to polarization processes. The effect of the
rods or tubes, where surface losses can have an
frequency of the applied voltage on the loss
exorbitant effect on the measured power
factor (product of dissipation factor and dielec-
factor.
tric constant) is illustrated in the lower section
6.7.4 Measurement. Power factor or dissipa-
of Fig 3.
tion factor can be obtained directly from many
6.6.4 Alternating Stress. The loss under
types of bridges. For test instruments that read
alternating stress is always greater than under
out in terms of volts, current, and watts loss,
direct stress because of polarization losses and
power factor may be calculated using the fol-
also because deteriorated areas of insulation
lowing formula:
which may not provide the continuous path
necessary for a direct leakage current can re- watts
power factor =
flect losses into an alternating voltage mea- (volts) (amperes)
surement through capacitive coupling. 6.8 Environmental Influences
6.6.5 Measurement. Losses under alternating 6.8.1 Temperature
voltage stress may be measured by use of a 6.8.1.1 General. Temperature has a marked
wattmeter, although for low-loss materials effect on the insulation resistance of most in-
(power factor less than 0.5 percent) good sulating materials. In some cases the insulation
wattmeter accuracy is difficult to obtain. resistance decreases by a factor of 2 for each
Losses may be calculated as the product of the 10C rise in temperature. When the resistance-
applied voltage and the in-phase component of temperature characteristic of an insulating
the resulting current (component separation
material or structure is known, the measured
method). Losses may also be calculated from
resistance value can be corrected to a standard
measurements of capacitance and dissipation
temperature base for comparison purposes.
factor, or power factor, as found by bridge
6.8.1.2 Polarization. The temperature ef-
techniques.
fect on alternating-voltage loss characteristics
6.7 Power Factor, Dissipation Factor is greatly influenced by various polarization
6.7.1 General. Power factor and dissipation processes in the insulating material. Increasing
factor are dimensionless and can be used to temperature appears to decrease the relaxation
compare the loss characteristics of insulation time of a process, increasing the frequency at
structure regardless of their geometry or size. which the polarization effect peaks. Thus an
All measurements used for comparisons should increase in temperature tends t o move the loss
be made with test voltages having the same factor curve of Fig 3 to the right, in the direc-
13
IEEE
tion of increasing frequency. Thus, depending the phase of the test voltage is shifted by 180
on which side of a polarization peak the mea- (for example, by interchanging leads to the
surement is made, the losses may increase or primary of the test voltage transformer) and a
decrease due to an increase in temperature. second measured value obtained. The desired
Loss-temperature characteristics for many ma- value is the average of the two measured
terials and structures have been recorded. values.
6.8.1.3 Water. Characteristics of insulation
6.9 Equipment. Commercial equipment is avail-
structures subject t o the adsorption of water
should be tested at temperatures above 0C. At able for all types of dielectric characteristic
measurements. Some of the most important
lower temperatures any ice which forms will
features which must be considered are listed
have a far different effect on the measurements
below.
than the water existing at higher temperatures.
6.8.2 Humidity. High humidity (above 50 Features applicable to all test sets are:
percent RH) or surface moisture can have a (1)Test voltage range
significant effect on surface losses and con- (2) Ability t o measure grounded specimens
sequently on the measured insulation resistance ( 3 )Ability to measure ungrounded speci-
and loss characteristics of the insulation struc- mens
ture. Surface guarding techniques t o remove (4) Guard circuit for specimen analysis
the effect of surface losses are usually effective, (5) Ability to function in presence of inter-
especially for tests using direct test voltages. ference
6.8.3 Interference (6) Portability
6.8.3.1 ElectridMagnetic Fields. Measure-
Features particularly applicable to direct-
ments of insulation characteristics made in the
voltage test sets are:
field are generally subject t o interference from
electric and magnetic fields originating in (1)Test voltage stability
(2) Test voltage rate-of-rise capability
neighboring energized equipment. The test
( 3 ) Output current range
equipment and connecting leads can be suit-
( 4 ) Measurement accuracy (voltage and cur-
ably designed and shielded t o be immune from
rent)
such fields (see Appendix A). The specimen
itself, however, can seldom be shielded, and Features particularly applicable to alternating-
when in an electric field an interfering current voltage test sets are:
will be introduced in any measuring circuit (1)Test voltage frequency
connected t o it. (2) Capacitance or output current range
6.8.3.2 Switch. An important source of ( 3 ) Ability to make accurate measurements
interference occurs at the switch which discon- in the presence of interference
nects the test specimen from the energized
system. The open switch constitutes a coupling
capacitor through which the system voltage 7.Tests on Specific Apparatus
produces an interfering current in any measur-
ing equipment connected to the test specimen. 7.1 Rotating Machinery
Such interference can be greatly reduced by 7.1.1 Withstand Voltage Tests
grounding the specimen side of the switch and 7.1.1.1 General. Withstand voltage tests on
providing a second disconnection between the rotating machinery insulation in the field are
switch and the specimen. The second discon- usually made at the time of installation or after
nection need only withstand the test potential repairs. They may be made when there is some
to be applied to the specimen. This procedure, reason t o expect damage, as after involvement
in addition to reducing interference, enhances of a machine in a system fault, or after several
safety (see 2.1.2). years of normal operation. Withstand tests are
6.8.3.3 Power Frequency. The effect of usually confined t o the ground wall insulation
power-frequency interference on a power- of the stator and rotor windings.
frequency measurement of dielectric loss or 7.1.1.2 Alternating Versus Direct Voltage.
capacitance (or, less precisely, charging cur- N o consensus exists in the choice of alternating
rent) can be nullified by a reversal pro- or direct voltage for withstand tests on rotating
cedure. A first measured value is obtained, then machinery insulation. There are advantages and
14
IEEE
disadvantages t o either method, and the choice by many of the methods listed in 5.3.
is often a matter of convenience. The very low 7.1.2.4 Inception Voltage and Pulse Magni-
frequency test is an attempt t o combine some tude. The partial-discharge inception voltage
advantages of both methods. and pulse magnitude measurements may be
7.1.1.3 Turn-to-Turn Tests. Turn-to-turn used t o monitor the condition and the tight-
insulation tests may be made by application ness of mica-insulated stator bars in the ma-
of a high-frequency test voltage t o a coil or chine slots. When a large number of pulses per
winding. The voltage may be in the form of cycle are encountered, the accuracy of indi-
repetitively pulsed wave trains. The turn- vidual pulse magnitude measurements may be
to-turn test voltage should not exceed the questionable because of the possibility of pulse
ground wall withstand value, so that it may be superposition.
necessary t o limit a test t o one or more coils 7.1.2.5 Probe Method. Probe methods t o
of a winding. If direct connection t o coils is indicate sites of severe partial-discharge activ-
not feasible, the test voltage may be induced ity as locations of incipient failures have met
in the coils. Detection of failure is often dif- with limited success.
ficult and is usually accomplished by compar- 7.1.3 Dielectric Characteristics
ing voltage or current waveforms on similar 7.1.3.1 General. Insulation resistance and
coils or windings. polarization index measurements are often
7.1.1.4 Rotor Winding Exception. Turn- used to indicate the surface condition of ma-
to-turn withstand voltage tests of rotor wind- chine insulation. Thus, these measurements
ing insulation are seldom made. Tests are may be used t o determine when the surfaces
made, however, t o determine if there are any need cleaning o r drying before they become a
shorted turns in a rotor winding. These tests flashover hazard. The measurements are also
involve impedance or flux measurements used during the drying of machine insulation
which are not considered dielectric tests and which has been flooded o r otherwise exposed
hence are not within the scope of this guide. t o excessive water.
7.1.2 Partial-Discharge Measurements 7.1.3.2 Insulation Resistance. Insulation re-
7.1.2.1 General. Partial-discharge measure- sistance measurements may be made to ascer-
ments on rotating machine stator insulation are tain that a machine insulation is in suitable
usually made t o indicate the amount of ioniza- condition for a withstand voltage test. Re-
tion energy available t o deteriorate the stator sistance measurements may be combined with
insulation. Mica, the major ground wall insula- direct-voltage withstand tests t o provide pos-
tion for high-voltage machines, is highly re- sible warning of impending breakdown (see
sistant to all but very concentrated discharges 3.3.2).
such as slot discharges (see 7.1.2.3).Other ma- 7.1.3.3 Periodic Measurements. Periodic
terials used, however, including those for measurements of insulation resistance, using a
strand and turn insulation, may be more sus- direct test voltage, o r of capacitance and power
ceptible t o ionization damage. Failure of such factor, using an alternating test voltage, may be
insulation can develop into a ground wall made t o monitor the overall condition of the
failure. insulation. Alternating-voltage tests should be
7.1.2.2 Dissipation or Power Factor. An made at the operating voltage level t o include
indicator of partial-discharge energy is the in- ionization losses.
crement in dissipation factor or power factor 7.1.4 Analysis by Phases
(power factor tip-up) as the test voltage is 7.1.4.1 General. For a three-phase machine
raised from below the ionization inception more information on the condition of machine
voltage to the operating voltage. The indicator insulation can be obtained if the individual
may be affected by materials in the insulation phases can be tested one at a time rather than
system which have nonlinear loss character- all together. Separation of phases requires the
istics with voltage, such as some voltage-grad- opening of the neutral connection.
ing paints used in the end turn area. 7.1.4.2 Withstand Test. Dielectric withstand
7.1.2.3 Slot Discharge. Slot discharge is a tests may be made on each phase in turn, with
special type of partial discharge which occurs the other two phases grounded. This procedure
when a coil side does not make intimate con- subjects the phase-to-phase (interphase) insula-
tact with a slot. Slot discharge may be detected tion t o the test voltage.
IEEE
7.1.4.3 Dielectric Measurements. Dielectric tion is the difficulty of obtaining portable

characteristic tests and partial-discharge tests test-voltage sources which meet the test re-
on individual phases permit a comparison of quirements. Normal overvoltage tests on a
results between phases. A localized fault will transformer require a source of alternating
produce a greater change in the test results on a voltage at some higher than normal frequen-
single phase than on a complete machine. cy and with sufficient capacity t o supply the
7.1.4.4 Interphase Measurements. Separa- magnetizing current and core losses of the
tion of phases permits the determination of di- transformer. Impulse tests on a transformer
electric characteristics on the interphase insula- require a high-voltage pulse generator which is
tion. A three-phase generator may be treated as not readily portable.
a four-electrode system (see Appendix B). 7.2.2.2 Oil Samples. Alternating voltage
7.1.5 Water-cooled Stators. Useful measure- withstand or breakdown test may be made
ments of dielectric characteristics cannot be on samples of oil removed from the trans-
made directly on the stator insulation of water- former. A special test cell is used for these
cooled machines because low-resistance water tests (see 7.9).
paths shunt the insulation. An approximate 7.2.3 Partial-Discharge Tests
measure of strictly alternating-voltage loss may, 7.2.3.1 General. Partial-discharge tests on
however, be obtained. Losses, including those transformers in the field usually must be made
in the water path, are first measured at a selec- a t normal system voltage, with the transformer
ted alternating voltage V (rms). Direct-voltage energized by the connection of at least one
loss, which includes the water-path loss, is then winding to the power system. While a more
calculated as V 2 / R , where R is the direct- effective test can be made using a separate
voltage resistance measured a t the same voltage source t o energize the transformer at
terminals and preferably at direct voltage V. some overvoltage, providing such a source (see
All measurements are made with normal water 7.2.2.1) is usually impractical.
flow. The difference between the two losses 7.2.3.2 Interference. A normal connection
is considered t o be the alternating-voltage loss t o the power system does not include sufficient
due t o polarization and ionization. This dif- impedance t o prevent discharges which occur
ferential loss can be used with the measured elsewhere in the power system from interfer-
alternating current in the calculation of power ing with the measurement, or t o prevent the
factor and power-factor tip-up. attentuation of those discharge pulses which
originate in the transformer. Field measure-
7.2 Transformers, Regulators, and Reactors ments of partial discharges under such condi-
7.2.1 General. To avoid unnecessary wording tions are not accurate and are chiefly used t o
in this section, only transformers are specifical- check for any intense discharge activity in a
ly mentioned. In general, the same material is transformer in which trouble is suspected.
also applicable t o regulators and reactors. A 7.2.3.3 Capacitance Taps. Connection of
reactor may be treated as a single-winding partial-discharge measuring equipment t o the
transformer. A regulator usually has shunt and transformer under test is facilitated when the
series windings. When these windings can be transformer bushings are equipped with capac-
separated, the regulator may be treated as a itance taps. The measuring equipment can be
two-winding or multiwinding transformer; connected t o a tap, thereby eliminating the
otherwise, it is treated as a single-winding necessity for a high-voltage coupling capacitor
transformer. and any high-voltage connections.
Transformers and regulators sometimes are
equipped with internal surge arresters or other 7.2.4 Dielectric Characteristics
protective devices which may interfere with a 7.2.4.1 General. Insulation resistance using
dielectric test. The internal wiring diagram, a direct test voltage, and power factor or dissi-
usually found on the name plate, should be pation factor using an alternating test voltage,
examined for this possibility before proceed- are the characteristics most commonly used for
ing with the test. the routine testing of transformer insulation.
7.2.2 Withstand Voltage Tests These tests are particularly sensitive t o mois-
7.2.2.1 General. I t is not general practice ture absorption by the insulation and in oil-
t o subject transformers in the field t o with- filled transformers t o contaminants in the oil
stand voltage tests. One reason for this situa- and t o deterioration of the oil itself.
IEEE
7.2.4.2 Bushings. The terminal bushings are ity of the condition causing the decomposition.
an important but often small part of the total Continuous or periodic checking for decompo-
insulation system of a transformer. A separate sition products in oil or gas is becoming a com-
test may be made on each bushing t o deter- mon practice for very large transformers.
mine its characteristics. Bushing tests are facili- 7.2.5.2 Core Grounds. Core-to-ground in-
tated where the bushings are equipped with sulation on core-type transformers may be
capacitance taps, test taps, or insulated draw checked for accidental grounds, particularly
leads (see 7.8). after shipment.
7.2.4.3 Oil. The oil in an oil-filled trans- 7.2.5.3 Others. A description of tests
former is often sampled on a routine basis which indicate situations leading t o insulation
and its dielectric characteristics measured in failure, but which are not tests on the insula-
a special test cell (see 7.9). tion itself, is outside the scope of this guide.
7.2.4.4 Temperature. The measured dielec- Some of these tests are listed by name only:
tric characteristics of transformer insulation winding resistance, tums ratio, and low-
can be greatly influenced by the temperature voltage impulse tests ( t o detect displacement
of the insulation at the time of measurement. of windings due t o short circuits).
The effect of temperature varies greatly with
different insulation systems; in general, the 7.3 Insulated Conductors (Cables)
older systems are more greatly affected than 7.3.1 General. Power cable is subject t o bend-
modem systems. For comparison purposes ing, tensile, and compression stresses during
the measurements should be converted t o a installation. Improper handling may result in
common temperature base. damage t o the cable insulation and its shield-
7.2.4.5 Components. The insulation system ing systems. Insulation tests are often made on
of a multiwinding transformer can be separated a cable after installation t o reveal any such
into components and the characteristics of damage.
A complete cable assembly includes joints,
each component found as described in Ap-
when necessary, and terminations. Termina-
pendix B. The tank and core usually form the
tions and joints are generally assembled on
ground electrode. Each separate winding forms
site. Tests t o check the insulation quality of
an additional electrode. A tapped winding, as
termination or joint assemblies must therefore
in an autotransformer, must be treated as a
usually be made in the field.
single electrode. In the measurement or cal-
Periodic tests may be made on the insula-
culation of the components, it is sometimes
found that a shield or winding placement is tion of the cable during its service life. Cable
such as t o practically eliminate one or more insulation is exposed t o several deteriorating
of the interwinding components. When the influences in service including the effect of
results indicate that a component has a near- transient overvoltages and circuit overloads.
zero capacitance, any calculated character- Cable systems are subject t o physical stresses
istic values should be discounted because of due t o load (temperature) cycling, and t o
accuracy limitations (see B2.2, Appendix B). damage from hostile environmental conditions
7.2.4.6 Interpretation. The interpreta- including erosion by stray earth currents, as
tion of the results of individual component well as chemical, bacteriological, and rodent
characteristics may sometimes be facilitated attacks.
by noting that the components involving the 7.3.2 Withstand Voltage Tests. Withstand
ground electrode include bushing insulation voltage tests are generally considered the most
while the interwinding components include reliable tests for exposing defects which origi-
only the internal insulation of the trans- nate during the installation of the cable, in-
former. cluding those in joint and termination assem-
blies.
7.2.5 Special Tests 7.3.2.1 Alternating Versus Direct Voltage.
7.2.5.1 Chemical. A chemical analysis may Withstand tests on cable insulation may be
be made on a sample of oil or the overlying made with either alternating or direct test
gas in the oil-filled transformer t o detect prod- voltages. Because of the high capacitance of
ucts of insulation deterioration. The composi- the average cable run, direct test voltages are
tion and concentration of these products furn- often used t o avoid the high reactive power
ish information regarding the nature and sever- required by an altemating-voltage test. Reso-
17
IEEE
nating inductors in power-frequency alternat- on short lengths of low-voltage cable. In

ing-voltage supply systems, or a very low fre- general, the ability of such measurements t o
quency (for example, 0.1 Hz) may be used t o detect local weak spots decreases with cable
reduce reactive power requirements. length.
7.3.2.2 Withstand/Partial-Discharge Tests. 7.3.4.2 Time-Voltage Characteristics. A
Withstand tests made with alternating test volt- series of insulation resistance measurements
ages are sometimes combined with partial- may be made while a cable is energized at
discharge (corona) tests (see 7.3.3.2),especially various voltages to determine resistance as a
when testing extruded insulation systems. function of voltage and time. These tests are
7.2.2.5 Determination of Test Voltage. sometimes combined with a withstand test. To
The rate of rise, the maximum value, and the facilitate comparison of test results, the test
duration of application of the withstand test procedure should follow an established time-
voltage depend on the rating and on the type voltage schedule.
of cable insulation. Manufacturers recom- 7.3.4.3 Dielectric Loss, Power Factor, Dis-
mendations and published standards should be sipation Factor. Measurements of dielectric loss
consulted in determining the test voltage t o be or an associated dimensionless factor (power
used for a particular cable. Consideration factor or dissipation factor) are sensitive indi-
should be given t o the insulation level of cators of the general condition of cable insula-
terminations, joints, and any apparatus con- tions. Measurements are preferably made at or
nected t o the cable. near operating voltage and frequency, especial-
7.3.3 Partial-Discharge (Corona) Tests ly if ionization is suspected. Measurements to
7.3.3.1 General. Partial-discharge tests on show changes in power factor or dissipation
cable insulation are preferably made with an factor with variations in the applied test volt-
alternating test voltage. A test voltage level age may be used t o evaluate nonlinear pro-
somewhat higher than the normal operating cesses, including ionization, taking place in the
value will activate any source of incipient cable insulation.
ionization. Because of the difficulty of provid- 7.3.4.4 Oil Samples. Oil from pipe-type
ing a suitable test voltage and controlling cables may be sampled and tested for support-
extraneous interference in the field, partial- ing information on the effect of service condi-
discharge tests on installed cable are gen- tions on the cable insulation.
erally limited t o short runs of cable and t o 7.3.4.5 Terminations. Dielectric character-
cable systems with suspected ionization prob- istic measurements on installed terminations
lems. (potheads) may be made separate from those
7.3.3.2 During Alternating-Voltage With- of the cable itself if the terminations are pro-
stand Test. Partial-discharge tests may be vided with test taps. For terminations not
made during an alternating-voltage withstand equipped with test taps, some information on
test on the cable. In this case, the test voltage the insulation characteristics may be obtained
supply and test connections must be free of by applying one or more temporary external
significant partial discharges, a requirement not electrodes (collars) t o the outer insulating
necessary for a simple withstand test. surface of the termination and measuring be-
7.3.3.3 Equipment Sensitivity. Most cable tween the electorde or electrodes and the
insulation is susceptible t o damage from low- conductor being terminated.
level partial discharges. Discharge measuring 7.3.4.6 Component Insulation. Measure-
equipment should therefore have sufficient ments on a belted multiconductor cable, in
sensitivity t o detect charges in the pico- which the conductors are not individually
coulomb range in high-capacitance specimens. shielded, may be made t o obtain the dielectric
7.3.4 Measurement of Dielectric Character- characteristics of the insulation between any
istics conductor and the lead jacket, or between any
7.3.4.1 General. Measurements of dielec- two conductors. Methods for analyzing the
tric characteristics are most often made t o de- component insulations in a multiconductor
tect changes in the overall condition of cable cable are described in Appendix B.
insulation which occur with time as a result of 7.4 Switchgear
service and environment. Measurements made 7.4.1 General. Switches for power circuits are
t o detect local weak spots are most successful made in a wide variety of design. Only a very
18
IEEE
general discussion of their insulating structures, ages open the circuit by a series of gaps, usual-
as they affect insulation testing, can be given in ly in an environment of air of SF, a t high pres-
this guide. sure. The voltage to be interrupted is distributed
7.4.2 Tests. Tests on switch insulation are among the gaps by grading capacitor assemblies.
usually confined t o the measurement of dielec- Such switches are usually designed in a module
tric characteristics such as insulation resistance, arrangement which permits each grading capaci-
power factor, or dielectric loss. Many other tor t o be measured separately. Capacitance
types of tests are made t o check the operation measurements on these assemblies are of par-
of a switch, including mechanism-motion ticular value in locating deteriorated capaci-
analysis and contact-resistance measurements, tors, since the measured capacitances may be
but these tests are beyond the scope of this compared with each other and with the design
guide. value which is generally known.
7.4.3 Terminals. A switch must have two 7.4.7 Oil Circuit Breaker. A common type of
terminals across which a circuit is closed or switch is the oil circuit breaker in which the
opened by the switching mechanism. Each circuit is broken within a tank of oil. The tank
terminal must be insulated for system voltage. may be dead (at ground potential), in which
When the switch is open, the terminals must be case the circuit is brought into the tank by
insulated from each other for a higher voltage bushings similar t o those of a transformer. A
in the event that there are system voltages a t contact assembly, including an arc suppression
the two terminals that are not in phase or in device, is fastened t o the lower end of each
synchronism. The basic test procedure for bushing, and a conducting crosspiece is lifted
switch insulation requires a test or measure- or rotated by an insulated operating member to
ment t o be made on each terminal of the open engage the contacts and complete the circuit.
switch, with the other terminal at or near Insulation in the tank in addition t o the bush-
ground potential. Each of these tests, there- ings, arc suppressor, operating member, and
fore, involves both insulation t o ground and oil may include guides for the operating
insulation between terminals. A supplementary member and a tank liner. Information on the
test with the switch closed removes the direct condition of the tank insulation may be de-
effect of insulation between terminals and is duced from open- and closed-breaker tests.
helpful in the analysis of test results. Separate tests may be made on the bushings
7.4.4 Opening and Closing. The opening and by techniques described in 7.8. Also, oil
closing of a switch requires mechanical force, samples may be taken from the tank for tests
which is usually generated at ground potential. specifically applicable t o oil (see 7.9).
Insulation is required in transmitting this 7.5 Lightning Arresters
force t o the switching element. A common 7.5.1 General. Withstand tests, in the usual
means for transmitting this force is an insulat- meaning of the term, cannot be made on a
ing rod or a column of insulating fluid con- lightning arrester, since the protective func-
fined in an insulating tube. This insulating rod tion of the arrester inhibits overvoltages be-
or tube may be included in the test of one or yond a specified value. The application of over-
both terminals, but in some switches the in- voltages t o check the operation of an installed
sulation is not directly in the test circuit unless arrester is not generally practiced, although
the switch is closed. For such switches, a some testing has been done a t lower transmis-
closed-switch test is particularly important.
sion voltages for which impulse generators are
7.4.5 Operating Means. Switches designed
more easily transportable. The energy output
to open a circuit under load may use an insulat-
of any impulse generator used for lightning-
ing structure designed t o quench the arc. This
arrester tests should not be sufficient to dam-
structure is usually associated with a switch
age the arrester.
contact and in some cases extends t o a grounded
-
support. Arc products and moisture may affect 7.5.2 Insulation Measurements. Measure-
the insulation characteristics of these structures, ments of the insulation characteristics of an ar-
and the condition of the insulation is reflected rester may be made t o assess the ability of the
in measurements made at the switch terminals. arrester t o withstand continued normal system
7.4.6 Arc-Quenching Structure. Switches for voltage. Such measurements have also revealed
use on the higher range of power-system volt- conditions which could affect the protective
19
IEEE
function of the arrester. Measurements of 7.7 Insulators

particular value include insulation resistance 7.7.1 Line and Station. Line and station in-
using a specified direct test voltage and power sulators in general have such a good service
loss at a specified alternating test voltage. Since record that testing is usually limited t o spot
nonlinear elements may be included in light- checks or t o the few designs that have a signif-
ning arrester measurements, accurate adjust- icant failure rate. Tests are usually made by the
ment of the test voltage is necessary for com- voltage-distribution method (see Section 4)
parable results. Normal test values for a specific while the insulators are in service.
type of arrester are usually obtained by statis- 7.7.2 Single Piece. Single piece insulators
tical analysis from a large number of tests. located between an energized line and ground
7.5.3 Gaps. Some arresters include in their cannot be effectively tested by the voltage-
internal circuits a gap or series of gaps which distribution method. One method of testing
may be shunted by resistors. Corrosion prod- such insulators requires the line to be grounded,
ucts may short-circuit a gap, a condition which a temporary electrode applied around the
is reflected in test measurements by decreased center of the insulator, and an insulation mea-
resistance and increased loss. An open shunting surement made between the electrode and
resistor, on the other hand, causes an increase ground.
in resistance and a decrease in loss. Partial- 7.7.3 Insulator Stack. An insulator stack
discharge tests may be useful in identifying supporting a conductor from ground may be
these types of deterioration. tested by grounding the conductor and testing
the insulators in pairs, applying the test voltage
7.6 Capacitors
t o the connection between the insulators and
7.6.1 General. Capacitors are used in high-
grounds t o the outer connections. The test may
voltage power systems as couplers for carrier
be a measurement of insulation character-
current circuits and as components of potential
istics. If the test voltage exceeds the voltage
devices. Such capacitors are seldom subject to
normally impressed on each insulator in ser-
withstand tests in the field, but are readily
vice, the test may also be considered as a sort
tested for insulation characteristics such as
of withstand test.
capacitance, power factor, and insulation
7.7.4 Corona. While not strictly an insulation
resistance.
problem, corona may occur at tie wires or in-
7.6.2 High Voltage. High-voltage capacitors
sulator hardware and cause objectionable radio
may consist of two or more lower voltage units
or television interference. Sources of corona
stacked in series. In this case, measurements
may be located by ultrasonic detectors or by
should be made on each individual unit.
7.6.3 Capacitance Measurements. Capaci- test equipment, receivers, or detectors which
operate in some frequency band between ordi-
tance measurements are important, since an
nary radio frequencies and a frequency beyond
increase in capacitance suggests the failure of
that of visual light. Details of corona location
one or more sections of a capacitor. For capaci-
are beyond the scope of this guide.
tors used with coupling-capacitor voltage trans-
formers, a change in capacitance or power fac- 7.8 Bushings
tor can affect the accuracy of the output volt- 7.8.1 Spares. Spare bushings are usually
age. An increase in measured power factor crated and stored in confined spaces. Satisfac-
indicates a possible incipient fault. tory withstand tests on a spare bushing usually
7.6.4 Power Factor Correction. Capacitors require that the bushing be removed t o an
are also used t o correct the power factor of a open space, uncrated, and mounted on a metal
system load. Tests on this type of capacitor stand with its lower end in oil to a level repre-
t o determine incipient faults are economically senting the normal operating environment.
questionable. Internal discharge resistors which Measurements of the insulation characteristics
shunt the capacitance may interfere with the of spare bushings can be adversely affected by
interpretation of insulation resistance measure- the crate and other surrounding material. Re-
ments. Capacitors in which one o r more sec- moval of the crate and use of the ungrounded-
tions have been punctured and short-circuited specimen test mode (grounded-guard circuit;
may be found by measurements of capacitance see A2.1.3, Appendix A ) are often helpful in
or charging current made at a low test voltage. obtaining a satisfactory reading.
IEEE
7.8.2 Connection t o Other Components. A 7.8.4 Correspondence with Nameplate Values.

bushing installed in power apparatus is con- Many high-voltage bushings have capacitance
nected within the apparatus tank t o some and power factor values as measured at the
other circuit component such as a transformer time of manufacture recorded on a nameplate.
winding. Isolating the bushing by disconnec- Test measurements should show a reasonable
tion within the tank is usually impractical, so correspondence with the nameplate values.
that it is.impossible t o confine a test measure-
7.9 Insulating Fluids
ment to the bushing insulation with certain
7.9.1 General. Insulating fluids are usually
exceptions.
sampled for test and introduced into a test cell
7.8.2.1 Test Tap. Some bushings are con- having a specific electrode configuration. While
structed with an electrode positioned near the samples can be sent to a testing laboratory,
ground flange so that the main insulation of tests in the field at the time of sampling
the bushing appears between the electrode and avoid misleading results due t o changes in the
center conductor. An insulated tap brought sample from handling and transportation.
through the ground flange of the bushing per- 7.9.2 Dielectric-Strength Test. A common
mits connection t o the electrode for energiz- test for an insulating fluid is the dielectric-
ing a potential device or for a test connection. strength test, in which the test voltage applied
This construction feature is referred to as a t o the electrodes is raised at a specified rate
potential tap or test tap. An ungrounded- until rupture occurs. The dielectric strength of
specimen test made between the center con- common insulating liquids is only moderately
ductor of the bushing and the tap will confine influenced by temperature over ordinary op-
measurements of insulation characteristics t o erating ranges. The dielectric strength of
the main insulation of the bushing. gases, however, is influenced by temperature
7.8.2.2 Draw Lead Bushing. Some bushings and pressure. The temperature and pressure of
and transformers are designed so that an in- the gas in the test cell at the time of test should
sulated lead from the transformer winding is be included in the test report.
brought through the hollow center conductor 7.9.3 Conducting Contaminants. Low dielec-
of the bushing t o the top or cap of the bushing tric-strength values obtained on insulating
where connection is made. By opening this fluids may usually be traced t o conducting
connection, the center conductor of the bush- contaminants. Water in one or more of its
ing is isolated from the transformer winding. phases is a common contaminant. Gases are
A test may then be made between the center sometimes checked for moisture contamina-
conductor and the bushing flange. To con- tion by means of dew-point measurements. An
fine the measurement to the bushing insula- additional cause of low dielectric-strength
tion and t o exclude the insulation of the trans- values for liquids are gas bubbles which may
former lead, the lead must be connected t o have been introduced in pouring the sample
the guard terminal of the measuring circuit. into the test cell. This problem can be avoided
Because the clearance at the cap and the lead by permitting the sample to remain in the cell
insulation will withstand only a few volts, the for sufficient time to allow the bubbles t o sur-
test voltage using a cold-guard circuit must face and disappear before the test voltage is
be kept very low. The voltage limitation may applied.
be avoided by use of the hot-guard circuit, 7.9.4 Test Methods. Measurements of the
which is recommended for this type of test. insulation characteristics of insulating fluids
(See A2.1.4, Appendix A). are usually confined to liquids. A test cell is
used in which the electrodes form the plates
7.8.3 Test on Local Area. A test may be of a capacitor and the liquid constitutes the
made on a local area of insulation above the dielectric. For accurate tests on low-loss
flange by applying a temporary electrode liquids, the electrode system includes guard
(conducting band or collar) around the in- electrodes t o avoid measurement of edge ef-
sulating surface of the bushing at a point fects and the insulation of the electrode sup-
nearest to the area in question. The test mea- ports. The capacitance of the cell with air as
surement is usually made between the elec- the dielectric is used in conjunction with a test,
trode and the grounded center conductor of measurement to obtain the dielectric con-
the bushing. stant (permittivity) or resistivity of the sample.
21
IEEE
7.9.5 Temperature. The insulation character- ASTM D 3382-75, Measurement of Energy

istics of many liquids are sensitive to tempera- and Integrated Charge Transfer Due to Par-
ture, so that the temperature of the liquid at tial Discharges (Corona) Using Bridge Tech-
the time of test should be included in the test niques
report. Low-loss liquids which may be used at
8.3 Dielectric Characteristics
high temperatures (for example, 100C) are
often tested at high temperatures, since in- ASTM D 257-76, Tests for D-C Resistance or
creased losses provide more accurate test re- Conductance of Insulation Materials
sults. D 150-74, Tests for A-C Loss Characteristics
7.9.6 High Power Factors, Dissipation Fac- and Dielectric Constant (Permittivity) of
tors, Low Resistivity. Abnormally high power Solid Electrical Insulating Materials
factors or dissipation factors, or abnormally
low resistivity values, may usually be traced to 8.4 Rotating Machinery Insulation
contaminants such as resins which dissolve in ANSI/IEEE Std 43-1974 Recommended Prac-
the liquid and cause extra polarization losses. tice for Testing Insulation Resistance of
7.9.7 Other Tests. Many other tests are ap- Rotating Machinery
plied t o insulating liquids which are not primar- ANSI/IEEE Std 56-1977 Guide for Insula-
ily electrical and which are outside the scope tion Maintenance for Large AC Rotating
of this guide. Such tests include viscosity, fire Machinery
and flash points, identification and evaluation
of dissolved gases, etc, and are usually made in ANSI/IEEE Std 95-1977, Recommended
a laboratory rather than in the field. Practice for Insulation Testing of Large AC
Rotating Machinery with High Direct Voltage
IEEE Std 286-1975, Recommended Practice
for Power-Factor Tip-Up of Rotating Ma-
8. Applicable Documents chinery Stator Coil Insulation
ANSI/IEEE Std 433-1974, Recommended
8.1 Withstand Voltage Tests
Practice for Insulating Testing of Large AC
IEEE Std 4-1978 IEEE Standard for High- Rotating Machinery with High Voltage at
Voltage Testing Techniques Very Low Frequency
ASTM D 149-75, Tests for Dielectric Break-
8.5 Transformer Insulation
down Voltage and Dielectric Strength of
Electrical Insulating Materials at Commercial ANSI/IEEE Std 93-1968, Guide for Trans-
Power Frequencies1 former Impulse Tests
8.2 Partial-Discharge (Corona) Tests 8.6 Conductor (Cable) Insulation
ANSI/IEEE Std 454-1973, Recommended IEEE Std 82-1963 (Reaff 1971), Test Pro-
Practice for the Detection and Measurement cedure for Impulse Voltage Tests on Insulated
of Partial Discharges (Corona) During Dielec- Conductors
tric Tests
ASTM D 1868-73, Detection and Measure- IEEE Std 83-1963 (Reaff 1971), Test Pro-
cedure for Radial Power Factor Tests on In-
ment of Corona Pulses in Evaluation of In-
sulating Tapes in Paper-Insulated Power
sulation Systems
Cable
NEMA Std 107-1964 (R 1971), Methods of
Measurement of Radio Influence Voltage of IPCEA T-24-380 (1974), Partial-Discharge
High-Voltage Apparatus2 Test Procedure (Corona)3
8.7 Capacitors
ANSI/IEEE Std 18-1968 Shunt Power Ca-
ASTM documents are available from American Soci- pacitors
ety for Testing and Materials, 1916 Race Street, Phila-
delphia, Pa. 19103.
2NEMA documents are available from National Elec-
trical Manufacturers Association, 2101 L Street, N.W., 31PCEA documents are available from National Elec-
Suite 300, Washington, D.C. 20037. trical Manufacturers Association.
POWER APPARATUS INSULATION
8.8 Liquid Insulation IEEE Std 76-1974, Guide for Acceptance

ANSI/IEEE C57.106-1977, Guide for AC- and Maintenance of Transformer Askarel in
ceptance and Maintenance of Insulating Oil Equipment
in Equipment 9. Applicable Documents in Preparation4
4When the following documents are completed, ap- IEEE Standards Project P400 (in preparation) High
proved and published, they become a part of this Potential Direct Voltage Dielectric Tests o n Insulated
listing. Conductors
IEEE Guide for Installation of Oil-Immersed Trans- IEEE Standards Project P510 (in preparation), Recom-
formers (10 000 kVA and larger, 69-287 kV Voltage mended Safety Practices in High-Voltage
Rating) ( T o be published in ANSI C57 Series).
23
Appendixes
(These Appendixes are not a part of IEEE Std 62-1978, Guide for Field Testing Power Apparatus Insulation.)
Appendix A
Shielding of Equipment for the Measurement

of Insulation Characteristics
A l . General A1.2 Shielding Against Internal Interference

A1.2.1 Guard Shield. A shielding system is
used to prevent the measurement of the cir-
Al.1 Shielding Against External Interference cuit insulation along with the insulation of the
A1.1.1 General. Measurement of insulation test specimen. This system is often designated
characteristics must often be made on appara- the guard shield and is usually so located and
tus located near conductors operating at high connected as t o return all currents, produced in
voltage and carrying heavy currents. The elec- the insulation of the measuring circuit by the
tric and magnetic fields emanating from these test voltage source, directly back t o the source
conductors may extend t o the measuring without passing through any measuring ele-
circuit, inducing voltages and currents therein ment.
which interfere with measuring accuracy. A1.2.2 Test Specimen. The guard shielding
A1.1.2 Loops. Magnetic fields induce volt- system may be extended t o the test specimen
ages in any wiring loops in the measuring cir- t o aid in the analysis of complex insulation
cuit. Usually wiring can be arranged t o avoid systems. Details of this use are discussed in Ap-
loops. Certain components, however, such as pendix B.
inductors and some wire-wound resistors, con- A1.2.3 Surface Leakage. An extension of the
tain many loops. Magnetic interference induced guard shielding system in the form of a wire
in these components can be minimized by or band around the insulating surface of a test
enclosing them in a shield made of a material specimen may be used for the purpose of ex-
having high initial permeability. cluding surface leakage from the measurement.
A1.1.3 Electric Fields. Electric fields induce Surface guards are more efficacious for direct
currents in the measuring circuit, an effect test voltages than for alternating voltages.
which may be described as due t o capacitive
coupling between the interfering conductor
A2. Basic Shielding Systems
and the circuit. This type of interference can
be effectively eliminated by enveloping the A2.1 Guard Shields
measuring circuit in a conductive shield con- A2.1.1 General. The need for a guard shield
nected t o ground. (see A1.2.2) is illustrated by the circuit of
A1.1.4 Test Specimen. The test specimen, Fig A l , in which the specimen insulation re-
an external part of the measuring circuit, sistance R , is t o be measured by the current
should be included in the ground shield. It is which flows through it into the meter in re-
seldom feasible, however, t o apply shielding sponse t o the test voltage. The measuring cir-
t o high-voltage power apparatus for test pur- cuit has three conductors, A, B, and C, with
poses. Interference induced in the test speci- associated insulation resistances R ,b , R ,,
men by external voltage sources can be dealt and Rat. Rab shunts the voltage source but
with by special test techniques (see 6.8.3). does not affect the measurement. Rbc shunts
A1.1.5 Safety of Personnel. For protection the meter but has such a high value in com-
of personnel, a conducting shield connected parison with the meter resistance that the
t o ground may be placed over any parts of the shunting effect is insignificant. R,, shunts the
test circuit energized by the test voltage test specimen, and the current which the test
source. voltage produces in it is measured by the meter
24
IEEE
along with the test specimen current. The A2.1.5 Cold-Guard Shield. A more con-
meter then does not give a true indication of venient circuit when one terminal of the test
the current through the specimen. specimen is grounded is shown in Fig A4. The
A2.1.2 Elimination of Leakage Path. The shield is connected to conductor B as in Figure
leakage path Ra, can be eliminated by means A2, but extends in the opposite direction to
of the guard shield shown by broken lines in enclose the voltage source and conductor A.
Fig A2. With the path eliminated, the meter The shield is slightly' above ground potential
now reads only the leakage current through by the voltage drop across the meter, which is
the test specimen. Leakage from conductor A conveniently in the ground circuit. In this case
to the shield adds to that through Rab of Fig the shield may be called a cold-guard shield.
A 1 and returns to the voltage source without
passing through the meter. In Fig A2 the com- A2.2 Ground Shields
bined leakage path is represented by Zab. For A2.2.1 Safety Use. In the circuits of Fig A2
alternating-current measuring circuits, charging and A3, the addition of ground shields is
currents due t o capacitance as well as to leak- necessary only for safety. Any currents picked
age currents must be considered (see A2.3). up by the measuring circuit due t o capacitive
A2.1.3 Grounded-Guard Shield. Conductor coupling with external voltage sources will flow
C in the shielding system of Fig A2 cannot be to ground either directly or through the low-
grounded, or the shielding system would theo- impedance voltage source without passing
retically have to cover the entire earth. A through the meter.
ground can be placed on conductor B as A2.2.2 Protection of Guard Shield. In the cir-
shown, which restricts the circuit t o the mea- cuit of Fig A4, currents from external sources
surement of ungrounded test specimens. The picked up on the guard shielding system will
ground on conductor B puts the meter con- flow to ground through the meter and thus
veniently in the ground circuit, and also puts cause erroneous readings. To prevent the error,
the shield at ground potential. In this case the the guard shield must be entirely covered by a
shield may be called a grounded-guard shield. ground shield. The circuit with both shields is
A2.1.4 Hot-Guard Shield. Conductor A of shown in Fig A5. Note that the test voltage
Fig A2 can be grounded instead of conductor source is shown as a transformer and that the
B, as shown. If one terminal of the test speci- primary winding is treated as an external
men is grounded, the test specimen is placed voltage source. The ground shield is interposed
in the circuit with its grounded terminal con- between the primary winding and the guard
nected to conductor A. Fig A 3 is an inversion shield.
of Fig A2 to show the connections in a more
conventional manner. Both the meter and the A2.3 Effect of Shielding System Capacitance
shield are at test potential, and in this case the A2.3.1 Phase Shift. For measurements using
shield may be called a hot-guard shield. There an alternating test voltage, the effect which
is an obvious difficulty in reading or adjusting test circuit capacitances have on the measure-
a meter which is inside a shield at test poten- ment must be considered in addition t o the
tial. effect of the resistances discussed in A2.1. The
TEST TEST
VOLTAGE SPEC I MEN
SOURCE Roc (Rx)
Rbc
B C
v v METER V
25
IEEE
r 4 I
1- VOLTAGE
SPECIMEN
I
METER
Fig A2
Grounded-Guard Shield
Fig A3
Hot-Guard Shield
Fig A4
Cold-Guard Shield
26
IEEE
Fig A5
Double-Shielded Circuit
capacitance which shunts the meter or measur- acteristic measurement due t o the effect of
ing element is that between conductors B and capacitance shunting the measuring element
C of Fig A1 and is enhanced by the shielding is proportional t o the resistance of the mea-
system of Figs A2, A3, A4, or A5. The effect suring element. The error will be insignificant
of this capacitance on a resistive measuring if the resistance of the measuring element is
element is to shift the phase of the current sufficiently low.
through the element and thus cause an error A2.3.4 Guard Balance. In bridge circuits,
in the measured loss characteristic, for ex- when the capacitance shunting the measur-
ample, dielectric loss, or power factor or dis- ing element is that between the guard and
sipation factor. ground shields, the effect of the capacitance
A2.3.2 Shunting Capacitance. Additional on the measurement may be nullified by
shunting capacitance may be present when bringing the potential of the guard shield
measuring components of a complex insula- to ground. The procedure is referred t o as a
tion system by method B or C described in guard balance. A bridge circuit incorporating
Appendix B. The test connections required by a guard balance feature is shown in Fig A6.
these methods place the capacitance of one or In this circuit, Z2 is the measuring element,
more of the components of the insulation sys- Z3 and Z4 are the standard arms used in the
tem in shunt with the measuring element. main balance of the bridge, and Z, and Z6
A2.3.3 Loss Error. The error in a loss char- are the guard balance arms.
Fig A6
Shielded Bridge Circuit with Guard Balance
27
Appendix B
Measurement Techniques for the Analysis of

Complex Insulation Systems
B1. Simple and Complex Systems accomplishing the determinations are de-
scribed in the following subsections, and for
convenience in this Appendix they are de-
B1.1 Simple System. A Simple insulation SYS- noted methods A , B, and C. The choice of
tem consists of two electrodes separated by method is limited by the facilities provided by
insulation, and can be represented as a single the measuring equipment.
capacitor. An example of a simple system is an B2.2.1 Method A. Method A requires calcu-
apparatus bushing, with its center conductor lation involving three or more measurements
and mounting flange as the two electrodes. t o determine the characteristic of any one
component. Method A may be used by measur-
B1.2 Complex System. A complex insulation
ing equipment capable of measuring insulation
system has three or more electrodes insulated
t o ground.
from each other. Capacitance may exist be-
B2.1.2 Method B. Method B permits a direct
tween any pair of terminals. A three-electrode
measurement on each component represent-
system may therefore be represented by a
ing insulation t o ground. The characteristics
network of three capacitors, a four-electrode
of the remaining components are found by cal-
system by six capacitors, and an N-electrode
culation similar t o that of method A. Method
system by Yd(N-1)capacitors. An example of
a complex system is a multiconductor belted
B requires the measuring equipment t o have
cable, in which each conductor and the lead the additional facility of a usable guard circuit,
sheath are electrodes. that is, a guard circuit t o which some com-
ponents of the insulation system can be con-
B1.3 Grounded Electrode. In a practical sys- nected with no appreciable effect on the mea-
tem, one of the electrodes is at ground po- surement accuracy.
tential. Thus in an N-electrode system there B2.1.3 Method C. Method C permits a direct
are N-1 ungrounded electrodes, each insulated measurement of any component. In addition
from ground. Consequently, there are N-1 t o facilities required by method B, method C
component capacitors representing insulation requires the further facility of adapting the
from ground. The remainder of the Yfl(N-1) measuring circuit to measure between un-
components represent insulation between ungrounded electrodes. In the ungrounded mea-
grounded electrodes. surement mode it must have a usable guard
circuit at ground potential.
B2.2 Direct Measurement. In general, greater
B2. Determination of the Characteristics accuracy is afforded by direct measurement.
of an Individual Component The accuracy of a determination calculated
from several readings is subject t o cumulative
B2.1 Methods of Measurement. The deter- error and is poor when the result represents a
mination of the characteristics of the indi- small difference between two large quantites.
vidual components of a complex insulation
system is of value in the detection and loca-
tion of defective insulation within the sys- B3. Method A
tem. The characteristics of any component
may be determined from measurements made B3.1 Individual Components. A characteristic
at the electrode terminals. Three methods for of an individual component cannot be directly
28
IEEE
Table B1
Method A
Three-Electrode System
~ ~~
Test Connections Calculations
To Measure Energize Ground Component Formula
A B,G I K m = 0.5 ( K 1 + K2 - K 3 )
B
A,B
A,G
G
II KAG = 0.5 ( K 1 + K3 - K 2 )
K , = 0 . 5 ( K 2 + K3 - K I )
Four-Electrode System
,,KAC
I
~~ ~~
To Measure Energize Ground Component Formula
measured by this method. Measurements on ponents identified by numerical subscripts.

various combinations of components can be Thus t o be able t o calculate the characteristic
made, however, and from these measurements KAB in a three-electrode system, it is neces-
the characteristic of an individual component sary t o have measured values for combinations
can be calculated. K1, K 2 , and K 3 .
B3.2 Calculation Formulas. In the right-hand B3.3 Test Connections. Test connections for
side of Table B 1 formulas are given for the measurement of the characteristic K for a
calculation of a characteristic of each individ- particular combination of components, as
ual component of three- and four-electrode sys- identified by a numerical subscript, are shown
tems. The characteristic t o be calculated is on the left-hand side of Table 1. The individ-
denoted by the letter K , and the location of ual components involved in the combination
the component in the system is identified are shown in parentheses, for information only.
by letter subscripts and also shown in dia- Thus, to measure K 1 in a three-electrode sys-
grams. The calculation requires the measure- tem, the test voltage is applied between elec-
ment of K for certain combinations of com- trode A and ground with electrode B also
29
IEEE
Table B2
Method B
To Measure Energize Guard Ground Component Formula
KAG A B G -
B A G -
K AG
K1 (= KAG + KAB) A - B,G K, = K1 -KAB
To Measure Energize Guard Ground I Component Formula
grounded. It may be noted that the measure- P is calculated from a simultaneous measure-
ment is on the parallel combination of com- ment of capacitance and dissipation factor for
ponents KAG and KAB . a given component or combination of com-
ponents. For example, P I = C I D l .
B3.4 Measurable Characteristics. The char-
acteristic K in Table B 1 is restricted t o capaci- B3.5 Power Factor or Dissipation Factor.
tance or a quantity generally proportional t o Power factor or dissipation factor of an indi-
capacitance. A list of characteristics which K vidual component cannot be directly calculated
can represent includes: from Table B1, since these characteristics are
C capacitance independent of capacitance. I t is necessary t o
A charging current at a specified voltage use the table twice f o r - t h e same component,
I leakage current at a specified voltage once with K = C or A , and again with K = W
G insulation conductance or P.
W loss (watts) at a specified voltage B3.5.1 Power Factor. If W and A (at speci-
P product of capacitance and dissipation fied voltage V ) have each been calculated for a
factor. certain component according t o the appropri-
30
Table B3
Method C
Ungrounded Test Mode Grounded Test Mode
To Measure Energize Measure Ground Guard Ground
,,KAC
Ungrounded Test Mode Grounded Test Mode
T o Measure Energize Measure Ground Guard Ground
KAB A B C, G - -
KAC A C B,G - -
KBC B C A, G - -
K AG A - - B,C G
- -
KBG B A,C G
- -
KCG G A,B G
ate component formula, the power factor of CAB of a threeelectrode system, DAB =
that component can then be calculated as PAB/CAB. Expanding,
PF = W/VA D AB = 0.5 (Pi + P, - P 3 )/0.5 (C, + C2 - C 3 )
and
B3.5.2 Dissipation Factor. If P and C have
each been calculated for a certain component D A B =(CID1 + c 2 D 2 - - c 3 D 3 ) / c 1 +c2-c3)
according to the appropriate component form- B3.6 Insulation Resistance. The determination
ula, the dissipation factor of that component of the insulation resistance of a component
can then be calculated as requires that the calculations be made with K
representing conductance (the inverse of re-
D = P/C sistance), or the leakage current at a specified
As an example, having calculated PAB and test voltage. The insulation resistance in ohms
31
IEEE
( R ) for a certain component may be found trodes may be individually determined by

from the calculated conductance in mhos (G) calculations similar t o those described in
by the relation R = 1 / G , or from the calcu- method A. Test connections and calculation
lated leakage current in amperes ( I ) and the formulas for three- and four-electrode systems
test voltage (V) by the relation R = V/I. are listed on Table B2. The restrictions on K
described in B3.4, and the methods for find-
ing power factor, dissipation factor, or insula-
B4. Method B tion resistance described in B3.5 and B3.6 also
apply.
B4.1 General. The characteristics of a com-
ponent of a complex system which exists be-
tween any ungrounded electrode and ground B5. Method C
may be measured directly by connecting all
other ungrounded electrodes to the guard B5.1 General. The characteristics of any com-
circuit of the measuring circuit. Test connec- ponent of a complex system may be measured
tions for three- and four-electrode systems are directly by use of test equipment which pro-
listed in Table B2. vides facilities for measuring both insulation
B4.2 Methods of Calculation. The character- t o ground and insulation between ungrounded
istics of those components of a complex sys- electrodes. Test connections for three- and
tem which exist between ungrounded elec- four-electrode systems are listed in Table B3.
32

Ieee STD 62 - Factor de Disipacion

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Ieee STD 62 - Factor de Disipacion

Încărcat de

Drepturi de autor:

Formate disponibile

IEEE Std 62-1978

(Revision of IEEE Std 62-1958)

IEEE Guide for Field Testing

IEEE Guide for Field Testing

The Institute of Electrical and Electronics Engineers, Inc

N o part o f this publication may be reproduced in any form,

Approved September 8, 1977

William E. Andrus R. 0. Duncan Donald T. Michael

1. Scope rect test voltage, provisions for grounding the

(5) Power input requirements results should be made by comparison of read-

self-shielding, and in such systems partial dis- and materials employed.

I I I sorption characteristics are known. A PI lower

the 1 min resistance value and is almost always DIPOLE

in the determination of the condition of in- * ELECTRICAL-&OPTICAL-

7.1.4.3 Dielectric Measurements. Dielectric tion is the difficulty of obtaining portable

nating inductors in power-frequency alternat- on short lengths of low-voltage cable. In

function of the arrester. Measurements of 7.7 Insulators

7.8.2 Connection t o Other Components. A 7.8.4 Correspondence with Nameplate Values.

7.9.5 Temperature. The insulation character- ASTM D 3382-75, Measurement of Energy

Suite 300, Washington, D.C. 20037. trical Manufacturers Association.

8.8 Liquid Insulation IEEE Std 76-1974, Guide for Acceptance

Shielding of Equipment for the Measurement

A l . General A1.2 Shielding Against Internal Interference

Measurement Techniques for the Analysis of

Test Connections Calculations

To Measure Energize Ground Component Formula

To Measure Energize Ground Component Formula

measured by this method. Measurements on ponents identified by numerical subscripts.

To Measure Energize Guard Ground Component Formula

To Measure Energize Guard Ground I Component Formula

Ungrounded Test Mode Grounded Test Mode

To Measure Energize Measure Ground Guard Ground

Ungrounded Test Mode Grounded Test Mode

T o Measure Energize Measure Ground Guard Ground

( R ) for a certain component may be found trodes may be individually determined by

S-ar putea să vă placă și