IEEE STD C57.21

IEEE Standard Requirements,
Terminology, and Test Code for

Shunt Reactors Rated Over 500 kVA
TM
IEEE Power Engineering Society

C57.21
Sponsored by the
Transformers Committee
IEEE
3 Park Avenue IEEE Std C57.21™-2008
New York, NY 10016-5997, USA (Revision of
IEEE Std C57.21-1990)
1 August 2008
Authorized licensed use limited to: Stanford University. Downloaded on October 11,2012 at 00:09:19 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C57.21™-2008
(Revision of
IEEE Std C57.21-1990)

Sponsor
Transformers Committee
of the
IEEE Power Engineering Society
Approved 27 March 2008

IEEE-SA Standards Board
Abstract: All oil-immersed or dry-type, single-phase or three-phase, outdoor or indoor shunt
reactors rated over 500 kVA are covered. Terminology and general requirements are stated, and
the basis for rating shunt reactors is set forth. Routine, design, and other tests are described, and
methods for performing them are given. Losses and impedance, temperature rise, dielectric tests,
and insulation levels are covered. Construction requirements for oil-immersed reactors and
construction and installation requirements for dry-type reactors are presented. This standard also
covers thyristor-controlled shunt reactors used in static var compensators.
Keywords: dielectric tests, dry-type shunt reactor, oil-immersed shunt reactor
The Institute of Electrical and Electronics Engineers, Inc.

3 Park Avenue, New York, NY 10016-5997, USA
Copyright © 2008 by the Institute of Electrical and Electronics Engineers, Inc.

All rights reserved. Published 1 August 2008. Printed in the United States of America.
IEEE is a registered trademark in the U.S. Patent & Trademark Office, owned by the Institute of Electrical and Electronics Engineers,
Incorporated.
National Electrical Safety Code and NESC are both registered trademarks and service marks of the Institute of Electrical and Electronics
Engineers, Inc.
PDF: ISBN 978-0-7381-5412-1 STD95779

Print: ISBN 978-0-7381-5413-8 STDPD95779
No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written permission
of the publisher.
IEEE Standards documents are developed within the IEEE Societies and the Standards Coordinating Committees of
the IEEE Standards Association (IEEE-SA) Standards Board. The IEEE develops its standards through a consensus
development process, approved by the American National Standards Institute, which brings together volunteers
representing varied viewpoints and interests to achieve the final product. Volunteers are not necessarily members of the
Institute and serve without compensation. While the IEEE administers the process and establishes rules to promote
fairness in the consensus development process, the IEEE does not independently evaluate, test, or verify the accuracy
of any of the information or the soundness of any judgments contained in its standards.
Use of an IEEE Standard is wholly voluntary. The IEEE disclaims liability for any personal injury, property or other
damage, of any nature whatsoever, whether special, indirect, consequential, or compensatory, directly or indirectly
resulting from the publication, use of, or reliance upon this, or any other IEEE Standard document.
The IEEE does not warrant or represent the accuracy or content of the material contained herein, and expressly
disclaims any express or implied warranty, including any implied warranty of merchantability or fitness for a specific
purpose, or that the use of the material contained herein is free from patent infringement. IEEE Standards documents
are supplied “AS IS.”
The existence of an IEEE Standard does not imply that there are no other ways to produce, test, measure, purchase,
market, or provide other goods and services related to the scope of the IEEE Standard. Furthermore, the viewpoint
expressed at the time a standard is approved and issued is subject to change brought about through developments in the
state of the art and comments received from users of the standard. Every IEEE Standard is subjected to review at least
every five years for revision or reaffirmation. When a document is more than five years old and has not been
reaffirmed, it is reasonable to conclude that its contents, although still of some value, do not wholly reflect the present
state of the art. Users are cautioned to check to determine that they have the latest edition of any IEEE Standard.
In publishing and making this document available, the IEEE is not suggesting or rendering professional or other
services for, or on behalf of, any person or entity. Nor is the IEEE undertaking to perform any duty owed by any other
person or entity to another. Any person utilizing this, and any other IEEE Standards document, should rely upon his or
her independent judgment in the exercise of reasonable care in any given circumstances or, as appropriate, seek the
advice of a competent professional in determining the appropriateness of a given IEEE standard.
Interpretations: Occasionally questions may arise regarding the meaning of portions of standards as they relate to
specific applications. When the need for interpretations is brought to the attention of IEEE, the Institute will initiate
action to prepare appropriate responses. Since IEEE Standards represent a consensus of concerned interests, it is
important to ensure that any interpretation has also received the concurrence of a balance of interests. For this reason,
IEEE and the members of its societies and Standards Coordinating Committees are not able to provide an instant
response to interpretation requests except in those cases where the matter has previously received formal consideration.
A statement, written or oral, that is not processed in accordance with the IEEE-SA Standards Board Operations Manual
shall not be considered the official position of IEEE or any of its committees and shall not be considered to be, nor be
relied upon as, a formal interpretation of the IEEE. At lectures, symposia, seminars, or educational courses, an
individual presenting information on IEEE standards shall make it clear that his or her views should be considered the
personal views of that individual rather than the formal position, explanation, or interpretation of the IEEE.
Comments for revision of IEEE Standards are welcome from any interested party, regardless of membership affiliation
with IEEE. Suggestions for changes in documents should be in the form of a proposed change of text, together with
appropriate supporting comments. Comments on standards and requests for interpretations should be submitted to the
following address:
Secretary, IEEE-SA Standards Board

445 Hoes Lane
Piscataway, NJ 08854
USA
Authorization to photocopy portions of any individual standard for internal or personal use is granted by The Institute
of Electrical and Electronics Engineers, Inc., provided that the appropriate fee is paid to Copyright Clearance Center.
To arrange for payment of licensing fee, please contact Copyright Clearance Center, Customer Service, 222 Rosewood
Drive, Danvers, MA 01923 USA; +1 978 750 8400. Permission to photocopy portions of any individual standard for
educational classroom use can also be obtained through the Copyright Clearance Center.
Introduction
This introduction is not part of IEEE Std C57.21-2008, IEEE Standard Requirements, Terminology, and Test Code for
Shunt Reactors Rated Over 500 kVA.
IEEE Std C57.21 was first published in 1971. It was subsequently revised in 1981 and 1990 and reaffirmed
in 1995 and 2004. The current revision process has focused on the following:
a) Annex A (normative) has been added to cover the specific requirements for dry-type air-core
thyristor-controlled shunt reactors, which are an important component of static var compensator
systems. Specific clauses, subclauses, tables, etc., in the main body of IEEE Std C57.21-2008 are
referenced regarding specification and test code. Other aspects of specification and test code are
covered in Annex A.
b) Annex B (informative) has been added to provide background information on dielectric stresses
imposed on shunt reactors during switching. The objective of Annex B is to provide guidance
regarding the application and design of shunt reactors, especially in consideration of the significant
switching duty and the associated repetitive dielectric stresses. As a consequence of the switching
stresses seen by shunt reactors, reduced BIL (basic lightning-impulse insulation levels), which is
common practice for power transformers, are no longer recommended and are not included in
Table 5. Formal input regarding the content of this annex was requested of the IEEE Switchgear
Committee and was provided. Input was also received from members of IEC MT32 on Inductive
Load Switching.
c) Table 5A and Table 5B in the previous version (IEEE Std C57.21-1990) have been combined into
one table as Table 5 (insulation levels of shunt reactors). Separate tables for dry-type and oil-
immersed shunt reactors were not deemed necessary as test levels are network-dependent and are
independent of shunt reactor technology. Maximum system voltage is introduced to be consistent
with terminology in the IEC reactor standard (see IEC 60076-6:2007 [B18]).a
Notice to users
Laws and regulations

Users of these documents should consult all applicable laws and regulations. Compliance with the
provisions of this standard does not imply compliance to any applicable regulatory requirements.
Implementers of the standard are responsible for observing or referring to the applicable regulatory
requirements. IEEE does not, by the publication of its standards, intend to urge action that is not in
compliance with applicable laws, and these documents may not be construed as doing so.
Copyrights
This document is copyrighted by the IEEE. It is made available for a wide variety of both public and
private uses. These include both use, by reference, in laws and regulations, and use in private self-
regulation, standardization, and the promotion of engineering practices and methods. By making this
document available for use and adoption by public authorities and private users, the IEEE does not waive
any rights in copyright to this document.
a
The numbers in brackets correspond to those of the bibliography in Annex D.
iv
Copyright © 2008 IEEE. All rights reserved.
Updating of IEEE documents
Users of IEEE standards should be aware that these documents may be superseded at any time by the
issuance of new editions or may be amended from time to time through the issuance of amendments,
corrigenda, or errata. An official IEEE document at any point in time consists of the current edition
of the document together with any amendments, corrigenda, or errata then in effect. In order to determine
whether a given document is the current edition and whether it has been amended through the
issuance of amendments, corrigenda, or errata, visit the IEEE Standards Association Web site at
http://ieeexplore.ieee.org/xpl/standards.jsp, or contact the IEEE at the address listed previously.
For more information about the IEEE Standards Association or the IEEE standards development process,
visit the IEEE-SA Web site at http://standards.ieee.org.
Errata
Errata, if any, for this and all other standards can be accessed at the following URL:
http://standards.ieee.org/reading/ieee/updates/errata/index.html. Users are encouraged to check this URL
for errata periodically.
Interpretations
Current interpretations can be accessed at the following URL: http://standards.ieee.org/reading/ieee/interp/
index.html.
Patents
Attention is called to the possibility that implementation of this standard may require use of subject matter
covered by patent rights. By publication of this standard, no position is taken with respect to the existence
or validity of any patent rights in connection therewith. The IEEE is not responsible for identifying
Essential Patent Claims for which a license may be required, for conducting inquiries into the legal validity
or scope of Patents Claims or determining whether any licensing terms or conditions provided in
connection with submission of a Letter of Assurance, if any, or in any licensing agreements are reasonable
or non-discriminatory. Users of this standard are expressly advised that determination of the validity of any
patent rights, and the risk of infringement of such rights, is entirely their own responsibility. Further
information may be obtained from the IEEE Standards Association.
Participants
At the time this standard was submitted to the IEEE-SA Standards Board for approval, the Revision of
C57.21 Working Group had the following membership:
Richard F. Dudley, Chair
Sten Andersson Derek Foster Les Recksiedler

Carlo Arpino Ramon Garcia Pierre Riffon
Peter Balma Ramsis Girgis Michael Sharp
Gene Blackburn Lars-Erik Juhlin Subhash Tuli
Alvazo Cancino Klaus Papp Tony Weekes
Joseph Cultrera Christoph Ploetner Peter Zhao
Fred E. Elliott Paulette A. Payne Powell Waldemar Ziomek
Ulf Radbrandt
v
The following members of the individual balloting committee voted on this standard. Balloters may have
voted for approval, disapproval, or abstention.
William J. Ackerman Kenneth Hanus Jeffrey Merryman

Carlo Arpino Gary Heuston Jeffrey Nelson
Ali Al Awazi Philip Hopkinson Michael S. Newman
Peter Balma John Horak Klaus Papp
Wallace Binder John Houdek CarlosA. Peixoto
Thomas Blackburn James Huddleston Alvaro Portillo
William Bloethe Charles Johnson Bertrand Poulin
Steven Brockschink Lars-Erik Juhlin Iulian Profir
Carl Bush Gael Kennedy Ryland Revelle
Donald Cash Sheldon Kennedy Pierre Riffon
Jerry Corkran J. Koepfinger Michael Roberts
Alan Darwin Jim Kulchisky Oleg Roizman
Dieter Dohnal Saumen Kundu Thomas Rozek
Richard F. Dudley John Lackey Bartien Sayogo
Gearold O. H. Eidhin Chung-Yiu Lam Devki Sharma
Fred E. Elliott Stephen Lambert Hyeong Sim
Gary Engmann Gerald Lee H. Smith
Joseph Foldi Debra Longtin James E. Smith
Marcel Fortin G. Luri John Vergis
Edwin Goodwin J. Dennis Marlow Loren Wagenaar
Randall Groves John W. Matthews James Wilson
Norman Haggerty Susan McNelly Roland Youngberg
Joseph Melanson
When the IEEE-SA Standards Board approved this standard on 27 March 2008, it had the following
membership:
Robert M. Grow, Chair

Thomas Prevost, Vice Chair
Steve M. Mills, Past Chair
Judith Gorman, Secretary
Victor Berman Jim Hughes Ron Petersen

Richard DeBlasio Richard Hulett Chuck Powers
Andy Drozd Young Kyun Kim Narayanan Ramachandran
Mark Epstein Joseph L. Koepfinger* Jon Walter Rosdahl
Alexander Gelman John Kulick Anne-Marie Sahazizian
William Goldbach David J. Law Malcolm Thaden
Arnie Greenspan Glenn Parsons Howard Wolfman
Ken Hanus Don Wright
*Member Emeritus
Also included are the following nonvoting IEEE-SA Standards Board liaisons:
Satish K. Aggarwal, NRC Representative

Michael H. Kelley, NIST Representative
Lisa Perry
IEEE Standards Project Editor
Matthew J. Ceglia
IEEE Standards Program Manager, Technical Program Development
vi
Contents
1. Overview .................................................................................................................................................... 1
1.1 Scope ................................................................................................................................................... 1

1.2 Purpose ................................................................................................................................................ 2
2. Normative references.................................................................................................................................. 2
3. Definitions .................................................................................................................................................. 3
3.1 Shunt reactor ........................................................................................................................................ 3

3.2 Rating of a shunt reactor ...................................................................................................................... 4
3.3 Insulation ............................................................................................................................................. 4
3.4 Losses and impedance ......................................................................................................................... 4
3.5 Construction ........................................................................................................................................ 4
3.6 Methods of cooling .............................................................................................................................. 5
4. General requirements.................................................................................................................................. 6
4.1 Service conditions................................................................................................................................ 6

4.2 Effect of air density on flashover voltage ............................................................................................ 7
4.3 Frequency ............................................................................................................................................ 7
4.4 Effect of altitude on temperature rise .................................................................................................. 8
4.5 Electrical properties of insulating oil ................................................................................................... 9
4.6 Magnetic characteristics ...................................................................................................................... 9
5. Basis for rating shunt reactors .................................................................................................................... 9
5.1 kVA ratings ......................................................................................................................................... 9

5.2 Terms in which rating is expressed.................................................................................................... 10
6. Tests ......................................................................................................................................................... 11
6.1 General .............................................................................................................................................. 11

6.2 Types of tests ..................................................................................................................................... 11
7. Losses and impedance .............................................................................................................................. 11
7.1 Total losses ........................................................................................................................................ 11

7.2 Impedance.......................................................................................................................................... 14
8. Temperature rise ....................................................................................................................................... 14
8.1 Life of insulating materials ................................................................................................................ 14

8.2 Conditions under which temperature limits apply ............................................................................. 15
8.3 Limits of temperature rise for continuous ratings .............................................................................. 15
9. Dielectric tests and insulation levels......................................................................................................... 15
9.1 Dielectric tests ................................................................................................................................... 15

9.2 Dielectric tests for line terminals ....................................................................................................... 17
9.3 Basic lightning-impulse insulation levels and insulation test levels for neutral terminals................. 21
vii
10. Test code................................................................................................................................................. 21
10.1 General ............................................................................................................................................ 21

10.2 Resistance measurements ................................................................................................................ 22
10.3 General dielectric tests ..................................................................................................................... 24
10.4 Losses and impedance ..................................................................................................................... 37
10.5 Temperature-rise tests...................................................................................................................... 42
10.6 Audible-sound-level test .................................................................................................................. 48
10.7 Vibration tests on oil-immersed shunt reactors ............................................................................... 55
10.8 Vibration tests on dry-type shunt reactors ....................................................................................... 57
10.9 Magnetic characteristic measurements ............................................................................................ 57
10.10 Seismic performance verification on oil-immersed and dry-type shunt reactors ........................... 58
11. Construction for oil-immersed shunt reactors ........................................................................................ 58
11.1 Bushings .......................................................................................................................................... 58

11.2 Bushing-type current transformers .................................................................................................. 58
11.3 Accessories ...................................................................................................................................... 59
11.4 Terminal markings ........................................................................................................................... 65
11.5 Oil preservation ............................................................................................................................... 65
11.6 Oil-preservation systems ................................................................................................................. 65
11.7 Tanks ............................................................................................................................................... 66
11.8 Shunt reactor finish .......................................................................................................................... 66
11.9 Other equipment accessories ........................................................................................................... 67
12. Construction and installation of dry-type shunt reactors ........................................................................ 67
12.1 General description .......................................................................................................................... 67

12.2 Safety ............................................................................................................................................... 68
12.3 Magnetic clearances......................................................................................................................... 68
12.4 Connections ..................................................................................................................................... 69
12.5 Installed sound level ........................................................................................................................ 69
12.6 Concrete foundations ....................................................................................................................... 70
12.7 Switching—Circuit breakers ........................................................................................................... 70
12.8 Protection practices for air-core shunt reactors................................................................................ 70
Annex A (normative) Specific requirements for dry-type air-core thyristor-controlled reactors ................. 72
Annex B (informative) Dielectric stresses imposed on shunt reactors during switching ............................. 88
Annex C (informative) Dielectric tests including information on wave shape control ................................ 94
Annex D (informative) Bibliography ........................................................................................................... 95
viii
IMPORTANT NOTICE: This standard is not intended to assure safety, security, health, or
environmental protection in all circumstances. Implementers of the standard are responsible for
determining appropriate safety, security, environmental, and health practices or regulatory
requirements.
This IEEE document is made available for use subject to important notices and legal disclaimers. These
notices and disclaimers appear in all publications containing this document and may be found under the
heading “Important Notice” or “Important Notices and Disclaimers Concerning IEEE Documents.”
They can also be obtained on request from IEEE or viewed at http://standards.ieee.org/IPR/
disclaimers.html.
1. Overview
1.1 Scope
This standard applies to all oil-immersed or dry-type, single-phase or three-phase, outdoor or indoor shunt
reactors rated over 500 kVA. This standard also applies to thyristor-controlled shunt reactors used in static
var compensators (SVCs).
This standard does not apply to other devices such as the following:
a) Shunt reactors, 500 kVA and smaller

b) Current-limiting reactors (see IEEE Std C57.16™-19961)
c) Arc-suppression coils
d) Neutral-grounding devices (see IEEE Std 32™-1972)
e) Saturable reactors
f) Line-resonant reactors
g) Filter reactors
1
Information on references can be found in Clause 2.
1
IEEE Std C57.21-2008
IEEE Standard Requirements, Terminology, and Test Code for Shunt Reactors Rated Over 500 kVA
1.2 Purpose
This standard specifies the requirements, terminology, and test code for oil-immersed and dry-type shunt
reactors rated over 500 kVA and for thyristor-controlled shunt reactors used in SVCs.
2. Normative references
The following referenced documents are indispensable for the application of this document (i.e., they must
be understood and used, so each referenced document is cited in text and its relationship to this document is
explained). For dated references, only the edition cited applies. For undated references, the latest edition of
the referenced document (including any amendments or corrigenda) applies.
ANSI S1.4-1983 (Reaff 2006), American National Standard Specification for Sound Level Meters.2
ANSI S1.11-2004, Specification for Octave-Band and Fractional-Octave-Band Analog and Digital Filters.
ANSI/ASME B1.20.1-1983 (Reaff 2001), Pipe Threads, General Purpose (Inch).
ANSI/CGA V-1-2005, Compressed Gas Cylinder Valve Outlet and Inlet Connections.
ASTM A167-1999 (Reaff 2004), Standard Specification for Stainless and Heat-Resisting Chromium-
Nickel Steel Plate, Sheet, and Strip.3
IEEE Std 1™-2000 (Reaff 2005), IEEE Recommended Practice—General Principles for Temperature
Limits in the Rating of Electric Equipment and for the Evaluation of Electrical Insulation.4, 5
IEEE Std 4™-1995, IEEE Standard Techniques for High-Voltage Testing.
IEEE Std 32™-1972 (Reaff 1990), IEEE Standard Requirements, Terminology, and Test Procedures for
Neutral Grounding Devices.
IEEE Std 315™-1975 (Reaff 1993), IEEE Standard Graphic Symbols for Electrical and Electronics
Diagrams (Including Reference Designation Letters).
IEEE Std 693™-2005, IEEE Recommended Practice for Seismic Design of Substations.
IEEE Std 1277™-2002, IEEE Standard General Requirements and Test Code for Dry-Type and Oil-
Immersed Smoothing Reactors for DC Power Transmission.
IEEE Std C37.015™-1993 (Reaff 2006), IEEE Application Guide for Shunt Reactor Switching.
IEEE Std C37.109™-2006, IEEE Guide for the Protection of Shunt Reactors.
IEEE Std C57.12.00™-2006, IEEE Standard for Standard General Requirements for Liquid-Immersed
Distribution, Power, and Regulating Transformers.
2
ANSI publications are available from the Sales Department, American National Standards Institute, 25 West 43rd Street, 4th Floor,
New York, NY 10036, USA (http://www.ansi.org/).
3
ASTM publications are available from the American Society for Testing and Materials, 100 Barr Harbor Drive, West
Conshohocken, PA 19428-2959, USA (http://www.astm.org/).
4
IEEE publications are available from the Institute of Electrical and Electronics Engineers, 445 Hoes Lane, Piscataway, NJ 08854,
USA (http://standards.ieee.org/).
5
The IEEE standards or products referred to in Clause 2 are trademarks owned by the Institute of Electrical and Electronics
Engineers, Incorporated.
2
IEEE Std C57.12.80™-2002, IEEE Standard Terminology for Power and Distribution Transformers.
IEEE Std C57.12.90™-2006, IEEE Standard Test Code for Liquid-Immersed Distribution, Power, and
Regulating Transformers.
IEEE Std C57.13™-1993 (Reaff 2003), IEEE Standard Requirements for Instrument Transformers.
IEEE Std C57.16™-1996 (Reaff 2001), IEEE Standard Requirements, Terminology and Test Code for Dry-
Type Air-Core Series-Connected Reactors.
IEEE Std C57.19.00™-2004, IEEE Standard General Requirements and Test Procedures for Power
Apparatus Bushings.
IEEE Std C57.19.01™-2000 (Reaff 2005), IEEE Standard Performance Characteristics and Dimensions for
Outdoor Apparatus Bushings.
IEEE Std C57.98™-1993 (Reaff 1999), IEEE Guide for Transformer Impulse Tests.
IEEE Std C57.106™-2006, IEEE Guide for Acceptance and Maintenance of Insulating Oil in Equipment.
IEEE Std C57.113™-1991 (Reaff 2002), IEEE Guide for Partial Discharge Measurement in Liquid-Filled
Power Transformers and Shunt Reactors.
IEEE Std C62.11™-2005, IEEE Standard for Metal-Oxide Surge Arresters for AC Power Circuits (>1 kV).
IEEE Std C62.22™-1997, IEEE Guide for the Application of Metal-Oxide Surge Arresters for Alternating-
Current Systems.
3. Definitions
For the purposes of this document, the following terms and definitions apply. The Authoritative Dictionary
of IEEE Standards Terms [B19] 6 should be referenced for terms not defined in this clause.
3.1 Shunt reactor
3.1.1 reactor: A device used for introducing impedance into an electric circuit, the principal element of
which is inductive reactance.
3.1.2 shunt reactor: A reactor intended for connection in shunt to an electric system for the purpose of
drawing inductive current.
NOTE—The normal use for shunt reactors is to compensate for capacitive currents from transmission lines, cables, or
shunt capacitors. The need for shunt reactors is most apparent during light loads. 7
6
The numbers in brackets correspond to those of the bibliography in Annex D.
7
Notes in text, tables, and figures of a standard are given for information only and do not contain requirements needed to implement
this standard.
3
3.2 Rating of a shunt reactor
3.2.1 maximum operating voltage: The maximum line-to-ground voltage or phase-to-phase voltage
associated with maximum system voltage. Unless otherwise specified, the maximum operating voltage
shall be taken to be 105% of rated voltage.
3.2.2 maximum system voltage: The highest rms phase-to-phase voltage that occurs on the system under
normal operating conditions and also corresponds to the highest rms phase-to-phase or phase-to-ground
voltage for which the shunt reactor is to be designed and operated without exceeding temperature limits.
Dielectric test levels are based on this voltage.
3.2.3 rated current: Derived from the rated voltage and rated kVA (kilovoltamperes).
3.2.4 rated kVA: The apparent power at rated voltage for which the shunt reactor is designed.
3.2.5 rated voltage: The voltage to which operating and performance characteristics of shunt reactors are
referred; the associated phase-to-phase or phase-to-ground voltage is used for calculation depending on the
shunt reactor system connection.
3.3 Insulation
3.3.1 dry-type shunt reactor: A shunt reactor in which the coils are neither impregnated with an insulating
fluid nor immersed in an insulating oil. The insulating medium for the windings is a solid dielectric and/or
air. The insulation to ground is provided by porcelain or polymeric insulators.
3.3.2 oil-immersed shunt reactor: A shunt reactor in which the coils and magnetic circuit are immersed in
or impregnated with an insulating oil.
3.4 Losses and impedance
3.4.1 impedance: The phasor sum of the reactance and effective resistance, expressed in ohms per phase.
The impedance may be derived from the rated kVA and rated voltage.
3.4.2 total losses: For an oil-immersed shunt reactor the total losses shall be taken as the sum of the
conductor I2R loss, the conductor eddy loss, magnetic circuit loss, shielding loss, and all other stray losses
in the shunt reactor.
For dry-type shunt reactors, the total losses shall be taken as the sum of the conductor I2R loss, the
conductor eddy loss, and the stray loss in the windings and manufacturer supplied framework, but not
including stray (eddy) losses in the user supplied support structure, surrounding structures (bus supports
etc.), or mounting pad (rebar).
3.5 Construction
3.5.1 indoor shunt reactor: A reactor that, because of its design and construction, must be protected from
the weather.
3.5.2 outdoor shunt reactor: A reactor of weather-resistant design and construction.
4
3.6 Methods of cooling
Shunt reactors shall be identified according to the cooling method employed. For liquid-immersed shunt
reactors, this identification is expressed by a four-letter code as described below. For dry-type reactors, the
identification is expressed similarly to that for oil-immersed shunt reactors but by considering only the third
and fourth letters of the four-letter code as used for liquid-immersed shunt reactors. These designations are
consistent with IEEE Std C57.12.00-2006 and IEC 60076-2:1997 [B14].
First letter: Internal cooling medium in contact with the windings (applicable only to liquid-immersed
shunt reactors):
⎯ O mineral oil or synthetic insulating liquid with fire point ≤ 300 °C

⎯ K insulating liquid with fire point > 300 °C
⎯ L insulating liquid with no measurable fire point
Second letter: Circulation mechanism for internal cooling medium (applicable only to liquid-immersed
shunt reactors):
⎯ N natural convective flow through cooling equipment and in windings

⎯ F forced circulation through cooling equipment (i.e., coolant pumps), natural convective flow in
windings (also called nondirected flow)
⎯ D forced circulation through cooling equipment, directed from the cooling equipment into at least
the main windings
Third letter: External cooling medium:
⎯ A air
⎯ W water (not applicable to dry-type reactors)
Fourth letter: Circulation mechanism for external cooling medium:
⎯ N natural convection
⎯ F forced circulation [fans (air cooling), pumps (water cooling)]
3.6.1 dry-type shunt reactors: Dry-type shunt reactors are generally cooled by the natural circulation of
the cooling air (class AN). Forced air circulation may also be used.
3.6.2 liquid-immersed shunt reactors: Liquid-immersed shunt reactors are generally cooled by natural
circulation of the cooling mineral oil within the winding section and natural circulation of the cooling air
over the external cooling surface (radiators, class ONAN). Other insulating liquids and/or forced liquid
circulation and/or forced air or water cooling may also be employed if specified.
NOTE—IEEE Std C57.12.00-2006 can be consulted as a reference.
5
4. General requirements
4.1 Service conditions
4.1.1 Usual temperature and altitude service conditions
A shunt reactor conforming to this standard shall be suitable for operation at its rated kVA and voltage,
provided that
a) The temperature of the cooling air (ambient temperature) does not exceed 40 °C, and the average
temperature of the cooling air for any 24 h period does not exceed 30 °C. 8
b) The altitude does not exceed 1000 m (3300 ft).
c) The top liquid temperature of an oil-immersed shunt reactor (when operating) shall not be lower
than –20 °C. Starting temperatures below –20 °C are not considered as usual service conditions.
4.1.2 Usual voltage conditions
Shunt reactors conforming to this standard shall be capable of operating continuously at maximum
operating voltage (line-to-ground voltage associated with maximum system voltage), and with the
increased current due to this overvoltage, without exceeding the rated temperature rise.
4.1.3 Usual installation conditions
The installation arrangement for single-phase shunt reactors without magnetic-field shielding and
connected in three-phase banks shall be as specified by those responsible for the design and application of
the shunt reactors.
For reactors without magnetic field shielding, the spacing between reactors of a three-phase bank and
between reactor banks is one determinant of the mutual reactance between adjacent phases and banks, and
thus, also of the rated kVA.
4.1.4 Unusual temperature and altitude service conditions
A shunt reactor may be applied at higher ambient temperatures or at higher altitudes than specified in 4.1.1,
but its performance will be affected. Typically, if not taken into consideration in the design, unusual service
conditions can impact the ability of the shunt reactor to operate at nominal/rated and maximum voltage.
Special consideration should be given to these applications.
4.1.5 Other unusual service conditions
Where unusual conditions other than those discussed in 4.1.4 exist, they should be brought to the attention
of those responsible for the design and application of the shunt reactor. Examples of some of these
conditions are as follows:
a) Damaging fumes or vapors, excessive or abrasive dust, explosive mixtures of dust or gases, steam,
salt spray, excessive moisture or dripping water, etc.
b) Abnormal vibration, shocks, tilting or seismic conditions (see 10.10 regarding seismic performance
verification)
8
It is recommended that the average temperature of the cooling air be calculated by averaging 24 consecutive hourly readings. When
the outdoor air is the cooling medium, the average of the maximum and minimum daily temperature may be used.
6
c) Excessively high or low ambient temperatures

d) Unusual transportation or storage conditions
e) Unusual space limitations
f) Unusual duty, frequency of operation, difficulty of maintenance, poor waveform, unbalanced
voltage, special insulation requirements, etc.
g) Unusual voltage conditions that may exist during periods of light system loading or unterminated or
open transmission lines during which time significant voltage rises may occur. (See 4.1.2 for usual
voltage conditions.)
h) Unusual operating requirements that might result from the absence of surge protection or the
proximity to magnetic material or structures.9
4.1.6 Environmental aspects
The environmental conditions shall be stated clearly in the inquiry, e.g. degree of salt contamination or
humidity, location of site if in the tropics or close to the sea, installation in desert areas or in heavy
pollution areas such as steel mills. The environmental aspects shall be considered in the design, for
example, in selecting materials and dimensioning of surface and turn-to-turn stresses of dry-type air-core
shunt reactors, to ensure safe operation during the expected lifetime of the reactor. IEC/TR 60815:1986
(see supporting published material [1] of A.9) is a useful reference document as it contains a table listing
ESDD (equivalent salt deposition density) levels versus site descriptors.
In addition, other environmental conditions such as actual maximum and average ambient temperatures at
the site (versus standard weather bureau data), seismic requirements, wind loads, etc., should be included in
the reactor specification. IEEE Std 693-2005 provides information regarding seismic design/qualification
requirements.
4.2 Effect of air density on flashover voltage
4.2.1 General
The effect of decreased air density due to high altitude is to decrease the flashover voltage for a given
distance. See IEEE Std 4-1995 for use of a correction factor with sphere gaps.
4.2.2 Insulation
The dielectric strength of a shunt reactor that depends in whole or in part upon air for its insulation
decreases as the altitude increases. The insulation level at 1000 m (3300 ft) multiplied by the correction
factor from Table 1 shall be not less than the required insulation level at the required altitude.
4.3 Frequency
The reference system frequency used in this standard for shunt reactors is 60 Hz.
NOTE—As this standard is used outside North America, 50 Hz may be the system operating frequency. In this case
calculations may have to be modified accordingly.
9
If a shunt reactor is not shielded magnetically, to minimize heating effects due to stray fields, consideration should be given to its
location relative to other apparatus and metallic structures, including reinforced concrete.
7
Table 1 —Dielectric strength correction factors for altitudes

greater than 1000 m (3300 ft)
Altitude
Altitude correction factor
(m) (ft) for dielectric strength
≤1000 ≤3 300 1.00

1200 4 000 0.98
1500 5 000 0.95
1800 6 000 0.92
2100 7 000 0.89
2400 8 000 0.86
2700 9 000 0.83
3000 10 000 0.80
3600 12 000 0.75
4200 14 000 0.70
4500 15 000 0.67
NOTE—An altitude of 4500 m (15 000 ft) is considered a maximum
for reactors conforming to this standard.
4.4 Effect of altitude on temperature rise
4.4.1 General
The effect of decreased air density due to high altitude is to increase the temperature rise of a shunt reactor
that is dependent upon air for the dissipation of its heat losses.
4.4.2 Operation at rated kVA
A shunt reactor can be operated at rated kVA at altitudes greater than 1000 m (3300 ft) without exceeding
temperature limits, provided that the average temperature of the cooling air does not exceed the values of
Table 2 for the respective altitudes.
Table 2 —Maximum allowable 24 h average temperature of

cooling air for carrying rated kVA
Altitude
Method of cooling
apparatus ≤1000 m 2000 m 3000 m 4000 m
(3300 ft) (6600 ft) (9900 ft) (13 200 ft)
Oil-immersed, self-cooled 30 °C 28 °C 25 °C 23 °C
Dry-type, self-cooled
55 °C rise 30 °C 27 °C 24 °C 21 °C
80 °C rise 30 °C 26 °C 22 °C 18 °C
100 °C rise 30 °C 24 °C 18 °C 12 °C
125 °C rise 30 °C 23 °C 16 °C 9 °C
150 °C rise 30 °C 22 °C 15 °C 7 °C
NOTE—Recommended calculation of average temperature is described in 4.1.1.
8
4.5 Electrical properties of insulating oil
The oil used in new shunt reactors shall meet or exceed the requirements listed in IEEE Std C57.106-2006,
including the dielectric strength of a sample of insulating oil taken from a new shunt reactor, which shall
not be less than the minimum kV values listed in Table 2 or Table 3 of that standard.
4.6 Magnetic characteristics
Shunt reactors may be identified, with regard to their magnetic characteristics, as linear, nonlinear, or
saturated. Most oil-immersed shunt reactors have a nonlinear magnetic characteristic where the normal
operation is in the unsaturated portion of the curve. (See Figure 1.)
NOTE—If the magnetic characteristic curve cannot be extended into the saturation region to determine the knee of the
curve at rated frequency, then conventional coordinates, which are rms voltage and current measurements, are adequate
(see 10.9).
Figure 1 —Nonlinear magnetic characteristics for shunt reactors
5. Basis for rating shunt reactors

See Clause 11 and Clause 12 for other requirements that may be specified for some applications.
5.1 kVA ratings
The kVA or MVA rating for shunt reactors is the apparent power for which the shunt reactors are designed,
and is based on rated voltage (line-to-ground voltage associated with nominal system voltage). Shunt
reactors shall also have the capability of continuous operation at maximum operating voltage (line-to-
ground voltage associated with maximum system voltage) without exceeding the temperature-rise limits
specified in Table 3.
9
Table 3 —Limits of temperature rise for continuously rated shunt reactors
Average winding Hottest-spot

Insulation
Type of shunt temperature rise winding
Item temperature
reactor by resistance temperature rise
class
(°C) (°C)
1 Oil-immersed — 65 80
Dry-type 105 55 65
130 80 90
155 100 115
180 125 140
220 150 180
2 Metallic parts in contact with, or adjacent to, the insulation shall not attain a temperature in excess of
that allowed for the hottest spot of the windings adjacent to that insulation.
3 Metallic parts other than those covered in Item 2 shall not attain excessive temperature rises.
4 An oil-immersed shunt reactor shall have an oil-preservation system. The temperature rise of the
insulating liquid in such a shunt reactor should not exceed 65 °C when measured near the top of the
tank.
NOTE 1— A shunt reactor with specified temperature rise shall have an insulation temperature index that
has been proven by experience or testing.
NOTE 2— The insulation temperature class in this table is supplied as a reference, and is based on the
“preferred temperature index” for insulation materials as defined in IEEE Std 1-2000.
NOTE 3— The maximum hot-spot temperature rise limits in this table are based on continuous operation
in a 30 °C daily average ambient with a 40 °C maximum.
If the actual average annual ambient temperature is lower, or the equipment sees cyclical loading (such as
the case for thyristor-controlled reactors), or the service duty is short term or seasonal, then consideration
can be given to increasing the allowable hottest-spot winding-temperature rise limit.
5.2 Terms in which rating is expressed
The rating of a shunt reactor shall be expressed in the following terms:

a) Rated voltage
b) Maximum operating voltage
c) Rated kVA
d) Rated current
e) Rated frequency
f) Basic lightning-impulse insulation level (BIL)
g) Number of phases
h) Oil and/or winding-temperature rise
i) Method of cooling
j) Delta or star connection (three-phase shunt reactor)
NOTE—Magnetic characteristics are not deemed to be a basis for rating. However, magnetic linearity for oil-immersed
shunt reactors employing a magnetic circuit may be important regarding switching considerations. However, it is
beyond the scope of this standard to define desired magnetic characteristics. The end user should define such
characteristics in their specification as required.
10
6. Tests
6.1 General
Unless otherwise specified, the tests described in 6.2 shall be made prior to delivery.
6.2 Types of tests
6.2.1 Routine, design, and other tests for shunt reactors
Types of tests for oil-immersed shunt reactors are listed in Table 4.A. Types of tests for dry-type shunt
reactors are listed in Table 4.B. Definitions of these various tests are included in IEEE Std C57.12.80-2002.
The listing of tests shown in Table 4.A and Table 4.B does not necessarily indicate the sequence in which
the tests shall be made. All tests are defined and shall be made in accordance with Clause 10.
6.2.1.1 Routine tests for shunt reactors
Routine tests shall be made on all shunt reactors in accordance with the requirements of Table 4.A and
Table 4.B, as applicable.
6.2.1.2 Design tests for shunt reactors
Design tests shall be made on shunt reactors in accordance with the requirements of Table 4.A and
Table 4.B.
6.2.1.3 Other tests for shunt reactors
When specified (as individual tests), other tests shall be made on shunt reactors as listed in Table 4.A and
Table 4.B. “Other” tests may be routine or design, as required or specified.
7. Losses and impedance
7.1 Total losses
7.1.1 Loss measurement conditions
The total losses of a shunt reactor shall be determined at rated voltage and rated frequency. If available test
power is insufficient for testing at rated voltage, then the manufacturer shall demonstrate to the user that
reduced-voltage testing produces sufficiently accurate results when extrapolated to the rated voltage level.
The manufacturer shall notify the user of reduced-voltage testing during the proposal stages.
7.1.2 Tolerance on losses
A tolerance on losses is utilized for two purposes: first, for commercial evaluation, and second, to provide
the basis of a quality check.
11
Table 4 .A—Routine, design, and other tests for oil-immersed shunt reactors
Test classification
Tests
Routine Design Other
Resistance measurements....................................................................................... X
The dc resistance measurements shall be made on the full winding.
Impedance measurements ...................................................................................... X
The impedance measurements shall be made on the assembled shunt reactor.
Total loss measurements ........................................................................................ X
Temperature tests ................................................................................................... X
Temperature test or tests shall be made on one unit when one or more units of a
given rating are produced by one manufacturer at the same time. These tests
may be omitted, however, when a record of a temperature test, made in
accordance with this standard, on a duplicate or essentially duplicate unit is
available.
Insulation power factor tests .................................................................................. X
Applied-voltage tests ............................................................................................. X
The applied-voltage tests shall be determined by the BIL of the neutral end of
the Y-connected shunt reactor.
Low-frequency overvoltage tests ........................................................................... X
Lightning-impulse tests:
Nominal system voltage of 115 kV or above .................................................... X
Nominal system voltage below 115 kV (only when specified) ......................... X
Neutrals of reactors for nominal system voltage above 115 kV (only when
specified)........................................................................................................... X
Switching impulse tests:
Nominal system voltage of 345 kV or above .................................................... X
Nominal system voltage below 345 kV (only when specified) ......................... X
Audible-sound tests:
Shunt reactors rated 50 MVA, three-phase (16.67 MVA, single-phase) or above,
or nominal system voltage of 115 kV or above (routine test performed with the
unit cold and design test performed with the unit at close to operating
temperature) X X
Shunt reactors rated below 50 MVA, three-phase (16.67 MVA, single-phase), or
nominal system voltage below 115 kV (only when specified) .......................... X
Vibration tests:
Shunt reactors rated 50 MVA, three-phase (16.67 MVA, single-phase) or above,
or nominal system voltage of 115 kV or above................................................. X
Determination of magnetic characteristics (linearity):
Shunt reactors rated below 50 MVA, three-phase (16.67 MVA, single-phase), or
nominal system voltage below 115 kV (only when specified) .......................... X
Shunt reactors rated 50 MVA, three-phase (16.67 MVA, single phase) or above,
or nominal system voltage of 115 kV or above. X
Mechanical tests.....................................................................................................
Mechanical tests shall be made to verify the pressure requirements for
maximum operating pressures and full vacuum filling. X
12
Table 4.B—Routine, design, and other tests for dry-type shunt reactors
Test classification
Tests
Routine Design Other
Resistance measurements....................................................................................... X
The dc resistance measurements shall be made on the full winding.
Impedance measurements ...................................................................................... X
The impedance measurements shall be made on the full winding.
Total loss measurements ........................................................................................ X
When specified, a total loss test may be performed on one unit of an order of
identical units to measure total losses at rated voltage or at maximum voltage
levels attainable with laboratory facilities.
Temperature tests ................................................................................................... X
Temperature test or tests shall be made on one unit when one or more units of a
given rating are produced by one manufacturer at the same time. These tests
may be omitted, however, when a record of a temperature test, made in
accordance with this standard, on a duplicate or essentially duplicate unit is
available.
Applied-voltage tests ............................................................................................. X
The applied-voltage tests shall be made only on support insulators when
specified.
Turn-to-turn test ..................................................................................................... X
This test is performed for nominal system voltages of 34.5 kV and below.
Lightning-impulse tests:
Nominal system voltage greater than 34.5 kV................................................... X
Nominal system voltage at or below 34.5 kV (only when specified) ................ X
Switching impulse tests.......................................................................................... X
Switching surge tests shall be made on shunt reactors rated 115 kV or
above (only when specified).
Audible-sound tests................................................................................................ X
The audible-sound-level test is a design test for dry-type shunt reactors only
when a maximum noise limit is specified.
Vibration tests ........................................................................................................ X
7.1.2.1 Tolerance on losses for commercial evaluation
As energy costs increase, losses become a more significant component of total operating cost, and, as such,
may be evaluated by the purchaser. Therefore, compliance to guaranteed losses becomes part of the
commercial contract. A tolerance on losses, to account, for instance, for measurement tolerances, etc., may
be part of the contractual agreement. Additionally, the contract may specify such guarantee criteria as
maximum loss per unit, average loss for all units, total “package” losses, etc. In any case, this is purely a
commercial matter between the purchaser and the manufacturer.
The only warning that should be stressed is that if a unit exceeds guaranteed loss, aside from the
commercial implications (which are a matter between the manufacturer and the purchaser), it is essential to
demonstrate that temperature rise limits for the insulation system employed are not exceeded.
7.1.2.2 Tolerance on losses as a basis of a quality check
Unless otherwise specified the total losses of a shunt reactor, demonstrated by tests and corrected to the
rated MVA base, shall not exceed the specified total losses by more than 6%. Failure to meet this loss
tolerance shall not warrant immediate rejection but shall lead to consultation between purchaser and
manufacturer. The manufacturer shall initiate an investigation in order to determine the cause of this
deviation. In order for acceptance to be considered, the manufacturer shall demonstrate to the purchaser, by
either calculation and/or test, that the deviation will not impair the ability of the unit to meet the other
requirements of this standard, including the temperature rise limits.
13
7.1.3 Stray losses
In the application of dry-type shunt reactors, due to the presence of an external magnetic field, there will be
losses in any adjacent metallic structures. The magnitude of these losses is a function of the proximity, type
of material, and geometric considerations.
7.2 Impedance
7.2.1 Tolerance
The tolerance for shunt reactor impedance at rated voltage shall be within ±2.5% of the specified value.
NOTE—In the case of nonlinear shunt reactors, for conditions above rated voltage, it may be necessary to define
impedance characteristics to prevent resonant overvoltages. Such characteristics should be specified by the user.
7.2.2 Tolerance between phases
In the case of a three-phase shunt reactor or a bank made of three single-phase shunt reactors, the maximum
deviation of impedance in any one phase shall be within ±2% of the average impedance of the three
phases. 10
7.2.3 Zero-sequence impedance
When specified, the zero-sequence impedance of oil-immersed iron-core shunt reactors shall be determined
at rated frequency as measured between the line terminals connected together and its neutral terminal. It is
expressed in ohms per phase. For three-phase shunt reactors, the zero-sequence impedance is three times
the impedance measured as indicated. The applied voltage should not exceed one-third of rated line-to-
neutral voltage, nor should the neutral current exceed the rated phase current.
8. Temperature rise
8.1 Life of insulating materials
The life of insulating materials commonly used in shunt reactors depends largely upon the temperatures to
which they are subjected and the duration of such temperatures.
Since the actual temperature is the sum of the ambient temperature and the winding-temperature rise, it is
apparent that the ambient temperature very largely influences the life of insulating materials used in shunt
reactors.
Other factors upon which the life of insulating materials depend are as follows:
a) Dielectric stress and associated effects
b) Vibration or varying mechanical stress
c) Repeated expansions and contractions
d) Exposure to air, moisture, etc.
10
For dry-type shunt reactors without magnetic-field shielding, this tolerance applies only when units are arranged in an equilateral
triangle configuration and isolated from any external magnetic influences.
14
8.2 Conditions under which temperature limits apply
Temperature limits shall not be exceeded when a shunt reactor is operated at maximum operating voltage.
8.3 Limits of temperature rise for continuous ratings
8.3.1 Limits of observable temperature rise 11
The temperature rise above the ambient temperature of shunt reactors and parts in contact with the
insulation or encapsulation material, when operated at maximum operating voltage, shall not exceed the
values given in Table 3. For procedures for determining temperature rise, see 10.5.
8.3.2 Limits of hottest-spot temperature rise
Shunt reactors shall be designed so that the hottest-spot conductor temperature rise above the ambient
temperature, when operated at maximum operating voltage, will not exceed the values given in Table 3.
9. Dielectric tests and insulation levels
9.1 Dielectric tests
9.1.1 General
In this current revision, a number of significant changes were made to dielectric tests and includes the
following:
a) In this revision, Table 5.A and Table 5.B of IEEE Std C57.21-1990 have been merged into one
table, Table 5, for shunt reactors used in effectively grounded systems. Test levels are network-
dependent and thus should be applicable to all technologies, e.g., dry-type or oil-immersed. Higher
insulation levels may be required for shunt reactors used in non-effectively grounded systems. In
such cases, specific standardized levels cannot be given because the required insulation levels
depend on surge arrester protective levels and protection strategy. For such applications, the rated
insulation levels shall be determined by system studies.
b) In Table 5, the nominal system voltage and maximum system voltage are included.
c) In Table 5, BIL values are based on the highest values associated with each nominal/maximum
system voltage. For shunt reactor applications, the highest BIL values are used since shunt reactors
are frequently (several times a day) switched on and off. This change essentially precludes the use
of reduced BIL levels for shunt reactors. Annex B (informative) provides background information
on the dielectric stresses imposed on shunt reactors during switching and is a basis for the
elimination of reduced BILs.
d) For oil-immersed shunt reactors, low-frequency tests shall consist of an applied-voltage test and a
low-frequency overvoltage test. For dry-type shunt reactors, an applied-voltage test is carried out
on the support insulators only and only when specified. A high-frequency overvoltage test or turn-
to-turn test is a routine test for dry-type shunt reactors; nominal system voltage 34.5 kV and lower.
e) Impulse tests, if performed, shall be made on each line terminal.
11
See 4.4 for the effect of altitudes on temperature rise.
15
f) The impulse test on the neutral of a shunt reactor shall be made only when specified for oil-
immersed units. If the insulation level of the neutral bushing is different from that of the neutral end
of the winding, the impulse-test voltage shall be for whichever insulation level is lower.
g) Switching impulse tests shall be performed in accordance with Table 5.
h) Insulation power factor tests are only required for oil-immersed shunt reactors. Insulation power
factor tests on dry-type air-core shunt reactors would be very difficult to measure and interpret,
since the insulation between the windings and ground is provided by the support insulators and
insulation resistance is essentially the surface resistance of the support insulators.
i) Consideration can be given to performing the dielectric tests for oil-immersed shunt reactors in the
following order: switching impulse test, lightning impulse test, overvoltage test with partial
discharge. However, it is most important to perform the overvoltage test with partial discharge
measurement as the last dielectric test.
9.1.2 Applied-voltage test
9.1.2.1 Oil-immersed shunt reactors
For oil-immersed shunt reactors, the applied voltage test is a test of the insulation to ground and the applied
voltage is in accordance with Table 5, Column 10.
9.1.2.2 Reduced insulation at the neutral
For an oil-immersed shunt reactor with reduced insulation at the neutral, the applied-voltage test shall be in
accordance with Table 5, based upon the insulation level of the neutral end.
9.1.3 Overvoltage tests
9.1.3.1 Low-frequency overvoltage test for oil-immersed shunt reactors
The low-frequency overvoltage test for oil-immersed shunt reactors is based on the maximum system
voltage as follows:
a) The low-frequency overvoltage test for oil-immersed shunt reactors of maximum system voltage of
48.3 kV and less shall consist of a 7200 cycle overvoltage test from Table 5, Column 7. This
voltage shall be applied across the winding with the neutral solidly grounded. No partial discharge
measurements are required.
b) The low-frequency overvoltage test for oil-immersed shunt reactors of maximum system voltage
of 72.5 kV and higher shall consist of an enhancement test level from Table 5, Column 9, for
7200 cycles without partial discharge measurement directly followed, without reducing the applied
voltage below the 1 h test level, by a 1 h low-frequency test from Table 5, Column 8. This voltage
shall be applied across the winding with the neutral solidly grounded. Partial discharge
measurements shall be made during the 1 h test.
9.1.3.2 Overvoltage tests for dry-type shunt reactors
a) Applied-voltage test. The applied-voltage test is a high-voltage test of the support insulators only.
b) Turn-to-turn overvoltage test. The turn-to-turn test for dry-type shunt reactors (nominal system
voltage 34.5 kV and below) shall be made by applying between the terminals of each winding a
train of high-frequency, exponentially decaying, sinusoidal voltages with a first-peak voltage at
least equal to the values as specified in Table 5, Column 11.
16
NOTE—The rationale for performing the turn-to-turn test in lieu of the impulse test as a routine test on dry-type air-
core shunt reactors is that the rise time to the first peak is the same order of magnitude as the front time for the impulse
test. Also, the test duration is longer (1 min/7200 overvoltages) and failure detection is simpler.
9.1.4 Lightning-impulse test
The lightning-impulse test shall include reduced full-wave, chopped-wave, and full-wave tests. Front-of-
wave tests may also be included when specified.
9.1.5 Insulation power factor test
Insulation power factor is the ratio of the power dissipated in the insulation, in watts, to the product of the
effective voltage and current, in voltamperes, when tested under a sinusoidal voltage and prescribed
conditions. It is recorded during factory tests to permit a comparison with the power factor measured in the
field.
9.2 Dielectric tests for line terminals
9.2.1 Line terminal BIL
The BIL chosen for each line terminal shall be such that the lightning-impulse, chopped-wave-impulse, and
switching impulse insulation levels include a suitable margin in excess of the dielectric stresses to which
the terminal will be subjected in actual service. For information on surge arrester characteristics and
application, refer to IEEE Std C62.11-2005 and IEEE Std C62.22-1997.
9.2.2 BIL and insulation test levels
The basic impulse insulation levels and insulation test levels are given in Table 5 for oil-immersed shunt
reactors and for dry-type shunt reactors.
9.2.3 Y-connected shunt reactors
Oil-immersed and dry-type shunt reactors designed for Y connection only shall be assigned insulation
levels for both line and neutral terminals.
9.2.4 Switching impulse test
When required, as indicated in Table 4.A and Table 4.B, switching impulse tests shall be performed on oil-
immersed and dry-type shunt reactors. A switching impulse test may be desired to demonstrate the
switching-surge insulation strength at the line terminals of the shunt reactor. A switching impulse test
consists of applying to each phase, from the line terminal to ground, a switching-impulse voltage transient
defined in Clause 10. The transients shall have a minimum crest value in accordance with Table 5.
NOTE—The required standard wave shape across the winding (e.g., > 100 μs / > 200 μs / > 1000 μs) can be difficult to
achieve, especially the time to the first zero crossing. A shorter wave shape may be accepted if the optimum impulse
generator configuration is used (use of parallel stages). The manufacturer should state at the bid stage if a deviation in
required wave shape is expected.
17
Table 5 —Insulation levels for shunt reactors (for use on effectively grounded systems)
Low-frequency
overvoltage test
Minimum (phase-to-
Maximum Nominal Chopped- Switching Applied
time to ground) Low-frequency overvoltage test Turn-to-turn
system system BIL wave impulse level voltage test
voltage 7200 cycles (phase-to-ground) overvoltage
voltage voltage (See Note 3) level (BSL) (See Note 8)
chopping without partial (See Note 7) (See Note 9)
(See Note 1) (See Note 2) (See Note 4) (See Note 5)
(See Note 11) discharge
measurement
(See Note 6)
Enhancement
One-hour level for
level with 7200 cycles
partial without
discharge partial
measurement discharge
measurement
kV rms kV rms kV crest kV crest µs KV crest kV rms kV rms kV rms kV rms kV peak.
Column 1 Column 2 Column 3 Column 4 Column 5 Column 6 Column 7 Column 8 Column 9 Column 10 Column 11
1.3 1.2 45 50 1.25 — 1.7 — — 10 41
2.6 2.5 60 66 1.5 — 4.3 — — 15 54
5.2 5.0 75 83 1.6 — 8.0 — — 19 68
9.1 8.7 95 105 1.8 — 13 — — 26 86
15.8 15 110 120 2.0 — 18 — — 34 99
26.3 25 150 165 3.0 — 30 — — 50 135
36.2 34.5 200 220 3.0 — 44 — — 70 180
48.3 46 250 275 3.0 — 56 — — 95 —
72.5 69 350 385 3.0 — — 65 71 140 —
121 115 550 605 3.0 460 — 105 118 230 —
145 138 650 715 3.0 540 — 125 142 275 —
169 161 750 825 3.0 620 — 145 165 325 —
242 230 900 990 3.0 745 — 210 249 395 —
362 345 1175 1295 3.0 975 — 315 355 520 —
550 500 1550 1705 3.0 1290 — 475 540 — —
800 765 1950 2145 3.0 1550 — 750 — — —
800 765 2050 2255 3.0 1700 — 750 — — —
NOTE 1—The maximum system voltage values given in Column 1 are the voltages for which shunt reactors can be continuously operated without exceeding the allowable
average and hot-spot temperature rise limits and audible sound limit. For tertiary-connected shunt reactors at 69 kV and below, the maximum tertiary voltage (Column 1) is
105% of nominal/rated unless otherwise specified.
18
Table 5—Insulation levels for shunt reactors (for use on effectively grounded systems) (continued)
NOTE 2—The nominal system voltage values given in Column 2 are used merely as reference numbers and do not necessarily imply a relation to operating voltages.
NOTE 3—Because shunt reactors may be frequently (several times a day) switched on and off, the use of reduced BIL levels is not recommended. For dry-type reactors having
a maximum system voltage less than 48.3 kV, the preferred method for testing the insulation between terminals is the turn-to-turn test. See Column 11.
NOTE 4—Chopped-wave test levels (Column 4) are 1.10 × BIL.
NOTE 5—Unless otherwise indicated, BSL values (Column 6) are equal to 0.83 × BIL. Because shunt reactors may be frequently (several times a day) switched on and off, the
use of reduced BSL levels is not recommended. This is a change from IEEE Std C57.21-1990.
NOTE 6—The low-frequency overvoltage test (phase-to-ground) is applicable only to oil-immersed shunt reactors having a maximum system voltage less than 72.5 kV. The
test voltage (Column 7) is approximately equal to 2.0 times the maximum system phase-to-ground voltage.
NOTE 7—The low-frequency overvoltage test is applicable only to oil-immersed shunt reactors having a maximum system voltage of 72.5 kV and above. Unless otherwise
indicated, the 1 h test voltage is equal to 1.5 times and the enhancement test voltage (Column 8 or Column 9) is equal to 1.7 times the maximum system phase-to-ground
voltage. The enhancement test is carried out directly prior the 1 h test. If the enhancement test level cannot be achieved in the manufacturer’s test facility because the power
demand is too high, this limitation/exception has to be stated by the manufacturer at bid stage. In this case, the test should be carried out at as high a level of enhancement as
possible or the enhancement test may be omitted by agreement between purchaser and manufacturer at the time of order.
NOTE 8—The applied voltage test as performed on dry-type air-core shunt reactors is a high-voltage test of the support insulators. The test level should be chosen according to
the BIL level specified for the reactor terminal having the lowest BIL. The applied voltage test levels for dry-type shunt reactors are the same as for oil-immersed shunt
reactors since they are applied interchangeably.
NOTE 9—The turn-to-turn test is not applicable for reactors having a nominal system voltage greater than 34.5 kV. For dry-type reactors having a nominal system voltage
equal to or lower than 34.5 kV, the turn-to-turn test is the preferred method for testing the insulation across the winding. The voltages are expressed in terms of peak voltage
because it is more applicable to the type of wave shape (decaying sinusoidal wave shape). All voltages are equal to 0.9 × BIL value since this is a test performed in lieu of the
lightning impulse. Because the wave shape is in between switching surge and lightning impulse, a de-rating factor of 0.9 is appropriate. See Column 9. A lightning impulse test
according to Column 3 may be performed as an alternative.
NOTE 10— For systems having a non-effectively grounded neutral (isolated or grounded through a high impedance value), the rated insulation levels should be determined by
using classic insulation coordination studies and depends on the protective level of the surge arresters installed across the shunt reactor. Insulation and test values should be
agreed upon between manufacturer and user.
NOTE 11— Spark gaps and other suitable means for voltage chopping may be used to carry out a chopped-wave impulse.
NOTE 12— Front-of-wave levels have not been listed in the revision of this standard since very few end users request this test.
NOTE 13— Historically, some of the BIL and BSL values included in previous editions of standards, such as IEEE Std C57.21-1990, were very high, especially for 765 kV
reactors. Instead of 2050 kV (BIL) and 1700 kV (BSL), 1950 kV (BIL) and 1550 (BSL) are recommended and included. The low-frequency overvoltage test with partial
discharge measurement for 765 kV reactors has been changed to 750 kV rms for 1 h. This change is based on the very good service experience of shunt reactors tested in this
manner. For instance, using 1950 kV BIL, 1524 kV (BSL), and 650 kV rms, for the 1 h test level, service experience was, in some cases, unsatisfactory.
19
Table 5—Insulation levels for shunt reactors (for use on effectively grounded systems) (continued)
NOTE 14— If an end user is employing shunt reactors at a maximum system voltage not included in this table, then the end user should specify both the maximum system
voltage and nominal system voltage. This applies to directly connected and tertiary-connected shunt reactors. The various test voltage levels should also be specified; possibly
using extrapolation from values listed in this table.
NOTE 15— In IEEE Std C57.21-1990, shunt reactors were required to be able to operate at voltages 5% higher than the nominal system voltage without exceeding the
temperature rise limits. If the end user wishes to utilize this rating practice, he should so stipulate in the specification that maximum operating voltage is based on 105% of
nominal system voltages.
NOTE 16— In the application of shunt reactors at voltages 69 kV and below, they are usually connected to a transformer tertiary. Transformer tertiary voltages, as such, do not
have to necessarily conform to typical distribution voltage standard values. Therefore, care must be taken in specifying maximum operating voltage levels corresponding to
nominal system voltage values. The maximum voltage on a transformer tertiary is impacted by the maximum voltage on the primary (transmission voltage) and by transformer
characteristics. Tertiary voltages do not necessarily follow typical values for distribution systems. Tertiary-connected shunt reactors may operate at distribution voltage levels
but they are part of the transmission system and defining parameters must be selected accordingly.
NOTE 17— One test level is listed for the overvoltage test on 800 kV shunt reactors. The service experience of some utilities with 800 kV shunt reactors showed that an
overvoltage test at 1.5 p.u. was not enough to verify the insulation system; the number of service failures was way too high. At a test level of 750 kV, service experience was
far better and failure rates were similar to those for lower rated voltages.
NOTE 18— The Working Group decided not to include an enhancement level for 800 kV reactors since very few labs are able to perform the test at the required voltage level
since about 30% more reactive power compensation is required. For instance one of Hydro Quebec’s normal 800 kV reactor ratings is 110 MVA (single-phase) and even IREQ
cannot perform the test at the required voltage level. Past experience with 800 kV shunt reactors showed good service experience when tested without enhancement. If the test
was carried out, an enhancement voltage of 850 kV (as is done for 800 kV power transformers) should be specified. However, purchasers should be aware that this new
requirement will require substantial changes for test laboratories and primarily involves the addition of substantial power factor compensation capacitor banks.
20
9.3 Basic lightning-impulse insulation levels and insulation test levels for neutral
terminals
The neutral terminal of a winding that is designed for Y connection only may have an insulation level
lower than that for the line terminal(s) in accordance with 9.3.1, 9.3.2, and 9.3.3.
9.3.1 Specified BIL
When specified, the neutral shall be designed for a specific BIL in addition to the low-frequency test.
9.3.2 Y-connected oil-immersed shunt reactors
Oil-immersed shunt reactors designed for Y connection only, with the neutral brought out and solidly
grounded directly or through a current transformer, shall have an insulation level at the neutral not less than
110 kV BIL.
9.3.3 Insulation level of neutral end of winding
The insulation level of the neutral end of the winding may differ from the insulation level of the neutral
bushing for which provision is made in the shunt reactor tank. In this case, the dielectric tests on the neutral
shall be determined by the insulation level of the neutral end of the winding, or the insulation level of the
neutral bushing, whichever is lower. Neutral BIL shall not, in any case, be lower than 110 kV.
The windings of single-phase or three-phase shunt reactors designed for Y operation shall be capable of
withstanding the applied-voltage test corresponding to the insulation test level of the line end for those cases
where the neutral is insulated to withstand the low-frequency test voltage assigned to the line terminal(s). The
windings (or insulators, in the case of dry-type shunt reactors) shall also be capable of withstanding a low-
frequency overvoltage test at a voltage in accordance with the low-frequency test values, specified in Table 5
for the nominal system voltage of the line end. The line end shall be capable of withstanding full-wave,
chopped-wave, and, if specified, front-of-wave impulse tests corresponding to its nominal system voltage.
10. Test code
10.1 General
This clause prescribes methods for performing tests specified in Clause 6. The test methods covered herein
are as follows:
a) Resistance measurements (see 10.2)

b) Dielectric tests and insulation power factor tests (see 10.3)
c) Losses and impedance (see 10.4)
d) Temperature-rise tests (see 10.5)
e) Audible-sound-level tests (see 10.6)
f) Vibration tests (see 10.7 and 10.8)
g) Magnetic characteristics measurements (see 10.9)
h) Seismic performance verification on oil-immersed and dry-type shunt reactors (see 10.10)
21
The order shown in the previous list does not necessarily indicate the sequence in which the tests shall be
made.
In this standard, an effort is made to use a uniform set of symbols without sacrificing simplicity or clarity.
The main symbols shall be as follows:
Symbol Quantity
E Voltage
I Current
P Active power (in-phase component)
Q Reactive power (quadrature component)
M Impedance, mutual
Z Impedance, self
R Resistance
X Reactance
T Temperature, as indicated in degree Celsius
t Time
Ø Impedance angle, degrees
F, C Factor, as indicated
K, k Ratio or factor, as indicated
h Hours
NOTE—Subscripts and other symbols used shall be as
locally identified.
10.2 Resistance measurements
10.2.1 General
Resistance measurements are of fundamental importance for two purposes:
a) The calculation of the conductor I2R loss

b) The calculation of winding temperatures at the end of a temperature-rise test
10.2.2 Determination of cold temperature
The cold temperature of the winding shall be determined as accurately as possible when measuring the cold
resistance.
10.2.2.1 Oil-immersed shunt reactors
Cold resistance measurements shall not be taken on a shunt reactor when it is located in a room in which
the temperature is fluctuating rapidly.
The temperature of the windings shall be assumed to be the same as the average temperature of the oil,
provided that
a) The windings have been under oil with no excitation and no current in the windings from 3 h to 8 h
(depending upon the size of the shunt reactor) before the cold resistance is measured.
b) The temperature of the oil has stabilized, and the difference between top and bottom temperature
does not exceed 5 °C.
22
10.2.2.2 Dry-type shunt reactors
Cold resistance measurement shall not be taken in less than 4 h after the shunt reactor has been moved from
one location to another, where the ambient temperatures differ by more than 5 °C but less than 10 °C.
Measurements should not be taken in less than 8 h if the temperature difference is more than 10 °C.
10.2.3 Oil-immersed shunt reactor windings out of insulating liquid
The temperature of the windings shall be recorded as the average of several thermometers or thermocouples
inserted between the coils, with care used to see that their measuring points are, as nearly as possible, in
actual contact with the winding conductors. It should not be assumed that the windings are at the same
temperature as the surrounding air.
10.2.4 Conversion of resistance measurements
Cold winding-resistance measurements are normally converted to a standard reference temperature equal to
the rated average winding-temperature rise plus 20 °C. In addition, it may be necessary to convert the
resistance measurements to the temperature at which the impedance and loss measurements were made.
The conversions are accomplished by using Equation (1):
T s + Tk ⎞
R s = R m ⎛ -----------------
⎝ T m + T k- ⎠
(1)
where
Rs is the resistance at desired temperature Ts

Rm is the measured resistance
Ts is the desired reference temperature
Tm is the temperature at which resistance was measured
Tk is equal to 234.5 °C for copper and 225 °C for aluminum
NOTE—The manufacturer shall use the appropriate value of Tk for the specified conductor material used and shall
advise the user accordingly.
10.2.5 Methods for measuring resistance
Bridge or drop-of-potential methods may be used for measuring direct-current (dc) resistance.
When drop-of-potential methods are used, the measuring equipment shall have a very high degree of
accuracy.
The dc resistance of the winding shall be measured as accurately as possible. The following precautions
shall be observed:
a) Current used when making dc resistance measurements shall not exceed 15% of the rated current.
Larger values may cause inaccuracy by heating the winding, and thereby changing its temperature
and resistance.
b) Measurements shall not be taken until after steady-state values have been reached.
c) If the drop-of-potential method for measuring dc resistance is used (see Figure 2), precaution d)
should also be taken:
23
d) Not less than four pairs of readings for current and voltage shall be taken. The average of the
resistances calculated from these measurements shall be considered to be the resistance of the
winding.
NOTE—When shutting down the dc resistance measurement circuit in Figure 2, the voltage should be lowered slowly.
It is also advisable to insert a resistor across the reactor terminals to control possible overvoltages due to the decaying
flux in the reactor.
Figure 2 —Connections for the drop-of-potential method of resistance measurement
10.3 General dielectric tests
10.3.1 Dielectric tests
10.3.1.1 Factory dielectric tests
The purpose of dielectric tests in the factory is to demonstrate that the shunt reactor has been designed and
constructed to withstand the specified insulation levels.
10.3.1.2 Test requirements
Test levels and other test parameters shall be as outlined in Clause 6 and Clause 9 of this standard, or as
otherwise specified.
10.3.1.3 Measurement of test voltages
Unless otherwise specified, the dielectric test voltages shall be measured in accordance with IEEE Std 4-
1995, with the following exceptions:
a) A protective resistance may be used in series with sphere gaps, on either the live or the grounded
sphere. Where it is unnecessary to protect the spheres from arc damage, it may be omitted.
b) The bushing-type potential divider method shall be considered a standard method for shunt reactor
tests.
c) The rectified capacitor-current method shall be considered a standard method for shunt reactor
tests.
10.3.1.4 Factory dielectric tests and conditions
10.3.1.4.1 Test sequence
The final dielectric test to be performed shall be the low-frequency overvoltage test.
24
10.3.1.4.2 Ambient temperature
One-hour low-frequency dielectric tests may be made at an ambient temperature under the conditions of
routine test.
10.3.1.4.3 Assembly
Shunt reactors shall be assembled prior to making dielectric tests. This includes bushings and terminal
compartments when necessary to verify air clearances. However, assembly of items such as radiators and
cabinets that do not affect dielectric tests is not necessary. Bushings shall, unless otherwise authorized by
the user, be those to be supplied with the shunt reactor.
10.3.1.4.4 Shunt reactors for connection to gas-insulated equipment
During dielectric testing of shunt reactors designed for direct connection to gas-insulated substations,
testing with the in-service bushings is preferred. Substitute air-oil bushings may be used, however, unless
otherwise specified by the user, live-part clearances and locations of the substitute bushings inside the
shunt reactor must be identical, within normal manufacturing tolerances, to those of the in-service
bushings. If the required internal clearances, or external air clearances, or both, cannot be achieved, suitable
arrangements will be required, as determined by the manufacturer and user in advance of the design of the
shunt reactor.
10.3.1.5 Tests on bushings
When tests are required on outdoor apparatus (air-to-oil) bushings separately from the shunt reactors, the
tests shall be made, in accordance with IEEE Std C57.19.00-2004, by the bushing manufacturer.
Details of separate testing of bushings for use on shunt reactors connected to gas-insulated equipment shall
be agreed upon by the manufacturer and user prior to the design of the shunt reactor.
10.3.2 Applied-voltage tests
10.3.2.1 Duration, frequency, and connections
A normal power frequency shall be used. The duration of the test shall be 1 min.
All terminals of the winding(s) under test shall be joined together and connected to the line terminal of the
power source.
All other terminals, the tank, and the core shall be connected to ground and to the other terminal of the
power source.
10.3.2.2 Protective gap
A protective gap set at a voltage 10% or more in excess of the specified test voltage may be connected
during the applied-voltage test.
10.3.2.3 Application of test voltage
The voltage should be started at one-quarter or less of the full value, and be brought up gradually to full
value in not more than 15 s. After being held for the time specified, it should be reduced gradually (in not
more than 5 s) to one-quarter or less of the maximum value, and the circuit should be opened.
25
10.3.2.4 Failure detection
Careful attention should be maintained for evidence of possible failure, which could include such items as a
breakdown of test voltage, an audible sound such as a thump, or a sudden increase in test-circuit current.
Any such indication should be carefully investigated by observation, by repeating the test, or by other tests
to determine if a failure has occurred.
10.3.3 Overvoltage tests
10.3.3.1 Low-frequency overvoltage test for oil-immersed shunt reactors of maximum

system voltage of 48.3 kV and less
The low-frequency overvoltage test for oil-immersed shunt reactors of maximum system voltage of
48.3 kV and less shall consist of a 7200 cycles overvoltage test from Table 5, Column 7. This voltage shall
be applied across the winding with the neutral solidly grounded. No partial discharge measurements are
required.
Three single-phase 7200 cycles tests may be substituted for a three-phase test.
In this case, the phase-to-phase insulation cannot be fully stressed and the end user should be notified of the
deviation by the manufacturer at the tender stage. Care should be taken to avoid overheating three-leg
reactors without a magnetic-return path during single-phase overvoltage tests.
NOTE—A three-phase supply should be used to perform the low-frequency overvoltage test on three-phase oil-
immersed shunt reactors connected in delta.
10.3.3.2 Low-frequency overvoltage test for oil-immersed shunt reactors of maximum

system voltage of 72.5 kV and higher
The low-frequency overvoltage test for oil-immersed shunt reactors of maximum system voltage of
72.5 kV and higher shall consist of an enhancement test level from Table 5, Column 9, for 7200 cycles,
without partial discharge measurement, directly followed, without reducing the applied voltage below the
1 h test level, by a 1 h low-frequency test from Table 5, Column 8. This voltage shall be applied across the
winding with the neutral solidly grounded. During this period, partial discharge measurements in
accordance with IEEE Std C57.113-1991 (apparent charge measurement) shall be continuously monitored
and reported in the test protocol for each 5 min intervals.
NOTE 1—The enhancement test level duration of 7200 cycles is also consistent with IEC 60076-3:2000 [B15].
Single-phase reactors shall be tested from a single-phase source. Three-phase reactors should preferably be
excited from a three-phase source or, if agreed upon by the manufacturer and the user, three individual 1 h
tests can be performed on each individual phase by using a single-phase supply. In this case, the phase-to-
phase insulation is not fully stressed and the end user should be notified of the deviation by the
manufacturer at the tender stage. Care should be taken to avoid overheating three-leg reactors or reactors
without a magnetic-return path during single-phase overvoltage tests.
NOTE 2—A three-phase supply should be used to perform the low-frequency overvoltage test on three-phase oil-
immersed shunt reactors connected in delta.
10.3.3.2.1 Frequency
The test frequency shall be increased, relative to operating frequency, as required to avoid core saturation,
winding overheating, or injurious heating of the metallic parts of the shunt reactor.
26
10.3.3.2.2 Failure detection
Failure may be indicated by the sudden rise of the partial discharge level, an indication of smoke and
bubbles rising in the oil, an audible sound such as a thump, or a sudden increase in test current.
Any such indication shall be carefully investigated by observation, by repeating the test, or by other tests to
determine if a failure has occurred.
In terms of interpretation of the partial discharge measurements, the results are acceptable and no further
discharge test required, if all of the following criteria are met:
a) The magnitude of the partial discharge does not exceed 500 pC (picocoulombs) during the 1 h test.
b) The increase in partial discharge level during the 1 h does not exceed 150 pC compared to the
“bench mark” partial discharge level taken immediately before the enhancement period begins.
c) The partial discharge level during the 60 min period does not exhibit any steadily rising trend, and
there is no sudden, sustained increase in levels during the last 20 min of the test.
Judgment should be based on the continuous monitoring and on the 5 min readings so that sporadic short-
duration excursions of the partial discharge level above the allowable 500 pC limit caused by external
effects such as cranes or other known external ambient sources shall be disregarded.
It is recommended to conduct the overvoltage test with ultrasonic transducers installed on the reactor tank.
These transducers may help to distinguish between internal or external partial discharge sources (locations).
Also, the test may be extended until an uninterrupted continuous period of 60 min with acceptable
performance, as previously defined, has been obtained.
As long as no breakdown occurs, and unless very high partial discharges are sustained for a long time, the
test is regarded as nondestructive. A failure to meet the partial discharge acceptance criteria shall require
consultation between user and manufacturer about further investigation.
10.3.3.2.3 Partial discharge measurements
Partial discharge measurements are not applicable to dry-type shunt reactors nor to oil-immersed shunt
reactors rated 72.5 kV and below.
10.3.3.2.4 Internal partial discharges
Apparent internal partial discharges (apparent charge) shall be measured at the terminals of the reactor
windings under test and reported in units of picocoulombs.
10.3.3.2.5 Instrumentation
A partial discharge meter shall be used to measure the apparent charge generated by any internal partial
discharges.
The partial discharge detector, based on IEEE Std C57.113-1991, is used to measure the partial discharge
levels at the terminal(s). The partial discharge meter shall be coupled to the line terminal(s) of the
winding(s) under test through the voltage tap of the bushing(s) or through suitable coupling capacitor(s)
connected in parallel with the tested winding(s). General principles and circuits are described in IEEE Std
C57.113-1991.
27
External shielding may be used to avoid corona in air, such as may occur at the bushing terminals or
grounded projections located in their vicinities. Radio-frequency chokes or tuned filters may be used to
isolate the reactor under test and the partial discharge measuring circuit from the remainder of the test
circuit, including its energy source.
10.3.3.2.6 Calibration
The test circuit shall be calibrated according to IEEE Std C57.113-1991.
10.3.3.3 Turn-to-turn overvoltage test for dry-type shunt reactors
The turn-to-turn test is performed by repeatedly charging a capacitor and discharging it, through sphere
gaps, into the reactor windings. The type of overvoltage that the reactor is subjected to is more
representative of a switching overvoltage, with an exponentially decaying sinusoidal wave shape. The test
duration is to be 1 min, and the initial crest value of each discharge is given in Column 11 of Table 5. The
ringing frequency is a function of the coil inductance and charging capacitor, and is typically on the order
of 100 kHz. The test shall consist of not less than 7200 overvoltages of the required magnitude.
Primary verification of winding insulation integrity should be based on oscillographic methods. A surge
oscilloscope and camera are used to record the last discharge superimposed on a reduced-voltage discharge;
on the order of 10% to 30% of full wave. A change in period or rate of envelope decay, between the
reduced and full waves, would indicate a change in coil impedance and thus an interturn failure.
Secondary verification of insulation integrity is made by observation. A failure can be detected either by
noise, or by smoke or spark discharge in the reactor windings.
Part (a) and part (b) of Figure 3 shows the schematic of the test circuit and representative oscillograms of
applied test voltage. The use of oscillograms for failure detection is based on change in ringing frequency
and a change in rate of envelope decay (damping).
For measurement of the waveforms generated during the turn-to-turn test, a digital data acquisition system
can be used to advantage and is preferred over historical methods. Part (c) and part (d) of Figure 3 illustrate
typical waveforms captured on a digital oscilloscope. Through use of PC-based software that can also be
used to control the test equipment, the waveform can be displayed on a PC monitor. The user can easily
observe and manipulate the captured voltage wave shapes allowing for accurate and in-depth evaluation of
the shunt reactor under test. Characteristics of the waveform can be displayed and recorded for reference.
In part (d) of Figure 3, the two waveforms, reduced and full wave, are superimposed. Damping
characteristics and resonant frequency are displayed allowing the user to determine the dielectric integrity
of the shunt reactor under test. Additionally, software can be custom designed to provide features that a
user may desire such as mathematical methods to accurately determine the first peak (note that this is the
reason that the read-out voltage in part (c) and part (d) of Figure 3 differs from a direct measurement of the
first peak) and shifting or equalizing of the two waveforms to evaluate differences. Note that the measured
peak voltage is the “smoothed” value and does not consider the oscillating transient occurring at the
beginning of the applied voltage.
28
(a)
NOTE 1—I, II: Oscillograms showing reactors which passed the turn-to-turn test.
NOTE 2—III: Oscillogram showing a reactor of the same rating as those of oscillograms I and II, but having a turn-to-
turn fault.
(b)
Figure 3 —Sample oscillograms
29
(c)
(d)
Figure 3—Sample oscillograms (continued)
30
10.3.4 Impulse test procedures
Lightning-impulse tests are required as a routine test for oil-immersed shunt reactors with a nominal system
voltage of 115 kV and above, and for dry-type shunt reactors with nominal system voltages greater than
34.5 kV. The tests shall consist of, and be applied in, the following order: one reduced full-wave, two
chopped-waves, and one full-wave. The time interval between application of the last chopped-wave and the
final full-wave should be less than 10 min to avoid recovery of dielectric strength if a failure were to occur
prior to the final full-wave. Oil-immersed shunt reactors below 115 kV and dry-type shunt reactors 34.5 kV
and below shall be tested only when specified. Dry-type shunt reactors rated 34.5 kV and below shall be
tested with the turn-to-turn overvoltage test.
If front-of-wave tests are also specified, impulse tests are generally applied in the following order: one
reduced full-wave, two front-of-waves, two chopped-waves, and one full-wave. A reduced full-wave shall
be applied first; and the full-wave shall be the last wave to be applied to the terminal under test. Other
reduced full-waves may be applied at any time during the intervening sequences.
Refer to IEEE Std C57.98-1993 for guidance and information on impulse-testing techniques, interpretation
of oscillograms, and failure detection criteria.
10.3.4.1 General
Impulse tests shall be made without excitation.
10.3.4.2 Full-wave test
This is a wave that rises to crest in 1.2 μs and decays to one-half of crest value in 50 μs from the virtual
time zero. The crest value shall be in accordance with the assigned BIL (see Table 5), subject to a tolerance
of ±3%, and no flashover of the bushing or test gap shall occur. The tolerance on time to crest should
normally be ±30%, and the tolerance on time to one-half of crest shall normally be ±20%. As a practical
matter, however
a) The time to crest shall not exceed 2.5 μs, except for windings of large impulse capacitance (low-
voltage, high-MVAr and some high-voltage, high-MVAr windings). For convenience in
measurement, the time to crest may be considered as 1.67 times the actual time between points on
the front of the wave at 30% and 90% of the crest value, as described in IEEE Std 4-1995.
b) To demonstrate that the large capacitance of the winding causes the long front, the impulse
generator series resistance may be reduced, which should cause superimposed oscillations. Only the
inherent generator and lead inductances should be in the circuit.
c) The impedance of some windings may be so low that the desired time to the 50% voltage point on
the tail of the wave cannot be obtained with available equipment. In such cases, shorter waves may
result and may be acceptable provided that the optimum impulse generator configuration is used
(use of parallel stages). In addition, the manufacturer shall declare at the bid stage if a shorter wave
shape is foreseen during tests.
The virtual time zero can be determined by locating points on the front of the wave at which the voltage is,
respectively, 30% and 90% of the crest value, and then drawing a straight line through these points. The
intersection of this line with the time axis (zero-voltage line) is the virtual time zero.
If there are oscillations on the front of the waves, the 30% and 90% points shall be determined from the
average smooth-wave front sketched in through the oscillations. The magnitude of the oscillations should
preferably not exceed 10% of the applied voltage.
When there are high-frequency oscillations on the crest of the wave, the crest value shall be determined
from a smooth wave sketched through the oscillations. If the period of these oscillations is 2 μs or more, the
actual crest value shall be used.
31
10.3.4.3 Reduced full-wave test
This wave is the same as a full-wave, except that the crest value shall be between 50% and 70% of the full-
wave value given in Table 5.
10.3.4.4 Chopped-wave test
This wave is inherently a full wave, except that the crest value shall be at the required higher level and the
voltage wave shall be chopped at or after the required time to sparkover but not later than 6.0 μs after
virtual origin. The virtual front time of the chopped wave may be different from the virtual front during the
full-wave test because of the presence of the chopping gap. Nevertheless, the tolerance on the virtual front
time for the chopped-wave test should remain as defined for the full-wave test.
The gap or other equivalent chopping device shall be located as close as possible to the terminals of the
reactor without disrupting its electrical field distribution. The distance between the chopping device and the
test object shall not exceed a lead length greater than the total height of the reactor (tank + bushing for oil-
immersed and coil + insulators + pedestals for dry-type). The impedance between the tested terminal and
the grounded end of the chopping device shall be limited to that of the necessary leads.
NOTE 1—This method will ensure that the chopping time (dv/dt of the voltage collapse) is as fast as possible.
NOTE 2—No criteria can be given on the steepness of the voltage collapse since the measuring circuit may not
accurately measure the chopping time. Nevertheless, by using the method described in this clause, the chopping time
should not generally exceed 1.0 μs.
NOTE 3—The use of a chopping gap made of sphere gap(s) is the preferred chopping method since it usually gives a
faster chopping time. Nevertheless, the use of a rod-rod chopping gap is permissible since this more accurately
represents the flashover over in air of insulators in service. The rod-rod gap naturally adds damping since its arc voltage
is greater than a sphere gap.
However, only for cases where the overswing to the opposite polarity is greater than 30%, it is permissible
to add a series-connected resistor in the chopping circuit to limit the amount of overswing to the opposite
polarity. When a resistor is added in the chopping circuit, the resistor shall not decrease the overswing to
the opposite polarity below 30% of the amplitude of the chopped wave.
10.3.4.5 Wave polarity
For oil-immersed shunt reactors, the test waves are normally of negative polarity to reduce the risk of
erratic external flashover in the test circuit. For dry-type shunt reactors, the test waves shall be of positive
polarity, unless otherwise specified.
10.3.4.6 Impulse oscillograms
All impulses applied to a shunt reactor shall be recorded by a cathode-ray oscillograph or by a suitable
digital transient recorder, unless their crest voltage is less than 40% of the full-wave level. These
oscillograms shall include voltage oscillograms for all impulses and ground-current oscillograms for all
full-wave and reduced full-wave impulses. Sweep times should be in the order of 2 μs to 5 μs for front-of-
wave tests (if performed); 5 μs to 10 μs for chopped-wave tests, 50 μs to 100 μs for full-wave tests, and
100 μs to 600 μs for ground-current measurements.
When reports require oscillograms, voltage and current oscillograms of the first reduced full-wave voltage,
voltage oscillograms of the last two chopped-waves, and voltage and current oscillograms of the last full-
wave shall represent a record of the impulse test to shunt reactors.
32
10.3.5 Connections for impulse tests
In general, the tests shall be applied to each terminal, one at a time.
10.3.5.1 Terminals not being tested
One terminal of the winding under test shall be grounded through a low-resistance shunt so that ground
current measurements can be made. The terminals of the winding or windings that are not being tested shall
be grounded.
Exceptions: All grounds shall be direct, except as previously described, or at the neutral terminal that may
be grounded through the same neutral grounding impedance as is to be used in service. If this impedance is
unavailable, the neutral shall be directly grounded.
10.3.6 Impulse tests on shunt reactor neutrals
One reduced and two full-waves are applied directly to the neutral with an amplitude equal to the insulation
level of the neutral. A wave having a front time of not more than 10 μs and a tail time to half-crest not less
than 40 μs shall be used. Exception: If the inductance of the winding is so low that the desired voltage
magnitude and duration to the 50% point on the tail of the wave cannot be obtained, a shorter wave tail may
be acceptable provided that the optimum impulse generator configuration is used (use of parallel stages). In
addition, the manufacturer shall declare at bid stage if a shorter wave shape is foreseen during test.
10.3.7 Detection of failure during impulse test
Because of the nature of impulse test failures, one of the most important matters is the detection of such
failures. There are a number of indications of insulation failure.
10.3.7.1 Ground-current oscillograms
In this method of failure detection, the impulse current in the grounded end of the winding tested is
measured by means of a cathode-ray oscillograph, or by a suitable digital transient recorder, connected
across a suitable shunt inserted between the normally grounded end of the winding and ground. Any
differences in the wave shape between the reduced full-wave and final full-wave, detected by comparison
of the two current oscillograms, may be indications of failure or deviations due to non-injurious causes.
They should be fully investigated and explained by a new reduced-wave and full-wave test. Examples of
probable causes of different wave shapes are operation of protective devices, core saturation, or conditions
in the test circuit external to the shunt reactor.
The ground-current method of detection is not suitable for use with chopped-wave tests.
It is difficult to shield the measuring circuit completely from the influence of the high voltage of the surge
generator. Some stray potentials are frequently picked up that may produce an erratic record for the first
1 μs or 2 μs. Such influences, if they occur at the start of the current wave (and to lesser extent at the start
of the voltage wave), should be disregarded.
Where the impedance of the shunt reactor under test is high, with respect to its series capacitive reactance,
current measurements may be difficult to make because of the small impulse current. In order to reduce the
initial large-capacitance current and to maintain a reasonable amplitude for the remainder of the wave, a
capacitor may be included in the current-measuring circuit. The capacitor should not be larger than required
to achieve this result.
10.3.7.2 Other methods of failure detection
Other methods of failure detection include the following:
33
a) Voltage oscillograms. Any unexplained differences between the reduced full-wave and final full-
wave detected by comparison of the two voltage oscillograms, or any such differences observed by
comparing the chopped-waves to each other and to the full-wave up to the time of chopping, are
indications of failure. Deviations may be caused by conditions in the test circuit external to the
shunt reactor, and should be fully investigated and confirmed by a new reduced-wave and full-
wave test.
b) Failure of gap to flashover. In making the chopped-wave test, failure of the chopping gap, or any
external part, to flashover, although the voltage oscillogram shows a chopped-wave, is a definite
indication of a failure either within the shunt reactor or in the test circuit.
c) Noise. Unusual noise within the shunt reactor, at the instant of applying the impulse, is an
indication of trouble. The cause of such noise should be investigated.
10.3.8 Switching impulse tests on oil-immersed shunt reactors
10.3.8.1 Test procedures
When a switching impulse withstand level is specified across the winding, a switching impulse test across
the windings shall be performed. In some cases, particularly when the inductance is low, the energy
available from the impulse generator may not be sufficient to obtain the required wave shape. In any case, a
minimum available energy of 50 kJ is desirable. See Annex C for more background information.
NOTE—Since extra-high-voltage shunt reactors are normally Y-connected and grounded and, consequently, the neutral
insulation level is low, a switching impulse test between the winding and ground is not performed.
10.3.8.2 Number of tests
The test series shall consist of one reduced voltage at 50% to 70% of the specified test level followed by
three full voltage transients at the specified test level.
10.3.8.3 Oil temperature
Switching impulse tests may be performed at oil temperatures assumed under normal operation or at the
temperatures attained under the conditions of routine tests.
10.3.8.4 Switching impulse waves
10.3.8.4.1 Polarity
For the switching impulse test, negative polarity is usually employed. If positive or both polarities are
required for the switching impulse test, the purchaser shall notify the manufacturer at the tender stage.
10.3.8.4.2 Wave shape
10.3.8.4.2.1 Switching impulse test across the reactor
The switching impulse voltage wave shall have a crest value in accordance with the assigned insulation
level, subject to a tolerance of ±3%, and shall exceed 90% of the crest value for at least 200 μs. The actual
time to crest shall be greater than 100 μs and the time to the first voltage zero on the tail of the wave shall
be at least 1000 μs. The energy available from the impulse generator may not be sufficient to produce the
required wave shape, especially for lower inductance shunt reactors. In cases where the specified wave
shape is not achievable, the manufacturer shall provide, preferably at bid stage or as a minimum at design
review or final electrical design stage, to the purchaser, details of the achievable wave shape for his
approval.
34
10.3.8.4.2.2 Time to crest
The actual time to crest shall be defined as the time interval from the start of the transient to the time when
the maximum amplitude is reached.
10.3.8.4.2.3 Time to first voltage zero
The time to the first voltage zero on the tail of the wave shall be defined as the time interval from the start
of the transient to the time when the first voltage zero occurs on the tail of the wave.
10.3.8.4.2.4 Ninety-percent time
A smooth wave sketched through any oscillations on the switching impulse voltage oscillogram may be
used to determine the time that the applied wave is in excess of 90% of the specified crest value.
A digital test record shall be taken of each applied voltage transient. The test is successful if there is no
sudden collapse of voltage indicated on the digitally recorded wave shape. For switching impulse tests
applied across the windings, successive digitally recorded wave shapes may differ because of the influence
of magnetic saturation on impulse duration
10.3.8.6 Sequence of tests
The switching impulse tests shall precede the low-frequency dielectric test.
10.3.8.7 Switching impulse test connections for three-phase shunt reactors
Connections used for making switching impulse tests are similar to those used for single-phase low-
frequency overvoltage tests.
Where the design permits, each winding may be tested separately at the test voltage shown in Table 5 with
the neutral grounded.
10.3.9 Switching impulse tests on dry-type air-core shunt reactors
10.3.9.1 Test procedures
Switching impulse tests are required as an “other” test for dry-type shunt reactors rated 115 kV and above.
Functionally, the switching impulse can be applied both across the winding and to ground. However, for a
switching impulse applied across the winding of a dry-type shunt reactor, it is technically impossible for
any test laboratory in existence today to obtain the required wave shape of 250 μs front and 2500 μs to 50%
of crest value. For instance, in the case of a 200 kJ impulse generator used to test shunt reactors of
1000 mH (which is a typical inductance rating), the best achievable wave shape is 400 μs zero crossing.
Another consideration is that BIL levels are at least 10% higher than switching impulse levels. For these
reasons, the switching impulse across the winding of a dry-type shunt reactor is classified as “other.” The
number of shots should be as for the recommended impulse design test.
For the switching-impulse type test to ground, there is no difficulty in obtaining the required wave shape. In
the case of dry-type shunt reactors, the major insulation to ground is provided by the support insulators.
Therefore, this test is essentially a test of the support insulators. Since the insulators constitute a self-
restoring insulation system, the test should be consistent with Procedure B in IEEE Std 4-1995, and the test
should consist of 15 shots of positive polarity. Two flashovers are allowed.
35
The test shall be made under wet conditions as specified in IEEE Std 4-1995. If the test equipment cannot
provide full wet conditions for all insulators simultaneously, the specified parameters for wet testing shall
be met at least for one insulator of the support structure.
10.3.9.2 Wave polarity
For dry-type shunt reactors the test waves shall be of positive polarity, unless otherwise specified.
10.3.9.3 Wave shape
The wave-shape shall be the standard 250/2500 μs and the crest value shall be in accordance with the
specified test level subject to a tolerance of ±3%. The tolerance on time to crest should be ±20% and the
tolerance on time to one half of crest shall be ±60%.
A digital test record shall be taken of each impulse wave. The test is considered successful if there is no
collapse of voltage indicated on the digitally recorded wave.
10.3.10 Insulation power factor tests
10.3.10.1 General
The insulation power factor is the ratio of the power dissipated in the insulation, in watts, to the product of
the effective voltage and current, in voltamperes, when tested under a sinusoidal voltage and prescribed
conditions.
The methods described herein are applicable to shunt reactors of present-day design that are immersed in
oil.
10.3.10.2 Preparation for tests
The test specimen shall have all of the following:
a) All windings immersed in oil

b) All windings short-circuited
c) All bushings in place
d) The temperature of windings and oil near the reference temperature of 20 °C.
Where the temperature is other than 20°C, the results shall be reported along with the actual temperature.
Note that there is no consensus on temperature correction factor for insulation power factor.
10.3.10.3 Instrumentation
The insulation power factor may be measured by special bridge circuits or by the voltampere-watt method.
The accuracy of measurement should be within ±0.25% of the insulation power factor, and the
measurement should be made at or near a frequency of 60 Hz.
10.3.10.4 Voltage to be applied
The voltage to be applied for measuring the insulation power factor shall not exceed one-half of the low-
frequency test voltage given in Table 5 for any part of the winding, or 10 000 V, whichever is lower.
36
10.3.10.5 Procedure
Insulation power factor tests shall be made from windings to ground.
10.3.10.6 Temperature correction factors
Temperature correction factors for the insulation power factor (if required or utilized) depend upon the
insulating materials, their structure, their moisture content, etc.
F PT
F P20 = ---------
K (2)
where
FP20 is the power factor corrected to 20 °C

FPT is the power factor measured at T °C
K is the correction factor
T is the test temperature
NOTE—In IEEE Std C57.12.90-2006, the temperature correction for insulation power factor tests has been eliminated.
The power factor is reported along with the measured top oil temperature and the lower oil temperature, if available.
Temperature correction of the insulation power factor test results for a trending basis may be applied by the user.
10.4 Losses and impedance
10.4.1 Reference temperature for losses and impedance
The reference temperature for shunt reactors to which losses and impedances are corrected shall be equal to
the limiting winding-temperature rise by resistance, covered in Table 3, plus 20 °C.
NOTE—In the case of loss-evaluated shunt reactors, consideration may be given to the use of a reference temperature
for loss calculations to be the average winding rise, as determined by temperature-rise test, plus 20 °C.
10.4.2 Impedance test
10.4.2.1 Definition
The impedance test is a determination of the ratio of the rated phase voltage to the current that flows with
rated voltage and frequency applied to the shunt reactor terminals.
10.4.2.2 Three-phase shunt reactor
The impedance of a three-phase shunt reactor shall be measured with three-phase voltage applied to the
shunt reactor terminals.
10.4.2.3 Test power
When available test power is insufficient for testing at rated voltage, the manufacturer shall demonstrate to
the user that reduced-voltage testing produces sufficiently accurate results when extrapolated to the rated
voltage level. The manufacturer shall notify the user of reduced-voltage testing during the proposal stages.
37
10.4.2.4 Unbalanced phase currents
If the currents in a three-phase shunt reactor are not balanced, the impedance current shall be taken as the
average of the three values.
CAUTION
Care should be taken to avoid overheating “three-leg” reactors without a “magnetic return path” during the
zero sequence impedance measurement.
10.4.2.5 Zero-sequence impedance
The zero-sequence impedance shall be measured at rated frequency between the line terminals, connected
together, and its neutral terminal. It is expressed in ohms per phase and is calculated by using 3 × E/I,
where E is the test voltage and I is the test current. This measurement shall be performed with a test current
corresponding to a neutral current equal to the rated phase current, and the test voltage not above one-third
of the line-to-neutral rated voltage.
10.4.3 Loss measurements
10.4.3.1 General
Since shunt reactors operate at extremely low power factors, small variations in frequency, deviations from
the true sine wave in applied voltage, errors in measuring components, and electromagnetic interference
may introduce significant errors in loss measurements. Proper test conditions and precision components,
specifically designed for low power factor measurements, are essential for an accurate determination of
shunt reactor losses.
a) Impedance bridges are frequently used to measure losses. They are generally more accurate than
wattmeter measurements. While many configurations of impedance bridge networks are possible,
the choice of a particular network shall be determined by the measurement problem at hand and the
testing facilities available.12
b) If wattmeters are used to measure losses, connections to the shunt reactor will be the same as those
shown in Figure 4, Figure 5, and Figure 6. The voltage is adjusted to the desired value at rated
frequency, and simultaneous readings of amperes, volts, watts, and frequency are taken. Because of
the extremely low power factor, corrections shall be considered for phase angle and losses in the
instruments and instrument transformers.
10.4.3.2 Loss tests on dry-type shunt reactors with no shields or magnetic shunts
In these shunt reactors, the losses consist of the dc resistance (I2R) losses in the conductor, and the eddy
losses in the conductor and any metallic framework of the clamping structure.
Since the losses in these shunt reactors are proportional to I2, the losses shall be measured at 100% voltage
or at a reduced voltage if equivalent precision of measurement can be demonstrated to the user’s
satisfaction. The losses are to be corrected to rated current and a reference temperature. In some cases, the
actual average winding rise, as determined by the temperature-rise test plus 20 °C, may be used. This is an
attempt to reflect actual site service losses and actual site average ambient temperature.
12
See [B5]–[B12] and [B21]–[B24] in Annex D.
38
NOTE—F = frequency meter; AV = average-voltage meter.
Figure 4 —Connection for loss measurement test of a single-phase shunt reactor

(a) without instrument transformers and (b) with instrument transformers
Figure 5 —Three-wattmeter method with shunt reactor neutral available
39
Figure 6 —Three-wattmeter method using artificial wattmeter neutral N
10.4.3.3 Loss tests on shunt reactors with enclosures but no magnetic shunts
The losses in these shunt reactors consist of the I2R losses in the conductor, eddy losses in the conductor,
and stray losses in the metallic enclosure. The losses in these shunt reactors are proportional to I2.
Therefore, the losses shall be measured at 100% rated voltage, or at a reduced voltage if better precision of
measurement can be demonstrated. The eddy and stray losses in the enclosure may have a different
temperature coefficient than the winding.
In this case, it may be necessary to make loss measurements at more than one temperature to verify the
temperature correction equation [see Equation (3)]. The losses are to be corrected to rated current, and a
reference temperature equal to the sum of the limiting average winding-temperature rise obtained from
Table 3 plus 20 °C.
⎛ T +T ⎞ ⎛ ⎞
Ps = I m2 Rm ⎜ s k ⎟ + ( Pm − I m2 Rm ) ⎜ Tm + Th ⎟⎟ (3)
⎜ T +T ⎟ ⎜ T +T
⎝ m k ⎠ ⎝ s h ⎠
Im is the current in reactor when losses were measured

Pm is the measured reactor losses at temperature Tm and current Im
Ps is the reactor losses at reference temperature Ts and measured current Im
Rm is the winding dc resistance at temperature Tm
Th is the temperature constant for enclosure (if different from Tk, to be determined by test)
Tm is the winding temperature at which loss measurements were made
Ts is the limiting winding-temperature rise plus 20 °C
10.4.3.4 Losses in shunt reactors with magnetic shunts or gapped iron cores
10.4.3.4.1 Definition
The losses in these shunt reactors consist of I2R losses in the conductor, eddy losses in the conductor, stray
losses in the clamping structure and enclosure, and losses in the magnetic shunt or core.
40
For these shunt reactors, it is required to measure the losses at rated voltage and correct them to reference
temperature, which is equal to the limiting average winding-temperature rise by resistance from Table 3
plus 20 °C.
The preferred method of measuring losses in these shunt reactors13 is the use of a current-comparator bridge
or an impedance bridge.
10.4.3.4.2 Three-phase shunt reactors
When three-phase shunt reactors are tested, three-phase excitation shall be applied to the shunt reactor. If a
single bridge is used, care shall be exercised to control the temperature, and accurately determine the
temperature of the windings during the test.
10.4.3.4.3 Test power
When available three-phase power is insufficient, single-phase testing can be performed, provided the
manufacturer can demonstrate to the user that measurements from single-phase tests can be converted to
results valid for three-phase excitation. The manufacturer shall notify the user during the proposal stages of
this condition. The losses may be segregated into I2R conductor losses, eddy losses in conductors and
clamping structure, and iron losses.14
10.4.3.4.4 Losses corrected to reference temperature
The losses corrected to the reference temperature are expressed in Equation (4) as follows:
⎛ T + Tk ⎞
Ps = I m2 Rm ⎜⎜ s ⎟
⎟
⎝ Tm + Tk ⎠
(4)
+ (
Pm − I m2 Rm )
⎛ T + Tk
− Pf ⎜⎜ m
⎞
⎟ + Pf
⎟
⎝ Ts + Tk ⎠
where
Pf is equal to the losses in magnetic shunt or core at the voltage corresponding to Im
Other terms are as defined under Equation (3).
A more realistic way of obtaining a reasonable temperature correction is by establishing the effective
overall temperature coefficient for the whole reactor experimentally.
If this is impracticable, the iron loss shall be deemed independent of temperature.
13
See Annex D of “Transformer-ratio-arm bridge for high voltage capacitance measurements” [B21].
14
Since iron losses cannot be measured separately on a shunt reactor, another method must be used for their determination. Therefore,
calculated iron losses provided by the manufacturer should be used. In previous editions of this standard, a method of iron loss
segregation achieved by measuring the losses in the reactor at two temperatures at least 10 °C apart was suggested. However, the
working group that produced the 2008 version of this standard determined that there were significant accuracy issues due to the
inherent tolerances in the measured data (temperature and total measured losses).
41
10.4.4 Correction of losses for current
10.4.4.1 Impedance constant
If the impedance of the shunt reactor remains constant, the losses measured at other than rated current may
be corrected to rated current and temperature, as expressed in Equation (5):
2
⎛I ⎞
Pr = ⎜ r ⎟ Ps (5)
⎜I ⎟
⎝ m⎠
where
Im is the current in the reactor when losses were measured

Ir is the rated current (see 3.2.3)
Pr is the losses at rated current
Ps is the reactor losses at reference temperature Ts and measured current Im
10.4.4.2 Impedance not constant
If the impedance of the shunt reactor is not constant, empirical methods may be required to correct the
various components of the losses to rated current and reference temperature.
10.5 Temperature-rise tests
10.5.1 Voltage and frequency
The temperature-rise test shall be made at rated frequency and maximum operating voltage (see Clause 8).
When the available test power does not permit making the test at maximum operating voltage, then the
manufacturer shall demonstrate to the user that reduced-voltage testing produces sufficiently accurate
results when extrapolated to the maximum operating voltage level. The manufacturer shall notify the user
of reduced-voltage testing during the proposal stages.
10.5.2 Test voltage for shunt reactors rated 34.5 kV or less
All shunt reactors having a rated voltage of 34.5 kV or less should be tested at maximum operating voltage.
10.5.3 Correction to maximum operating voltage
The temperature rises shall be corrected to maximum operating voltage by calculating the additional
temperature rise that would result from increasing the losses from those measured at the temperature test
voltage level to those associated with the maximum operating voltage level. The losses used for this
extrapolation will be corrected to the ultimate, average temperature of the windings at both the test level
and the maximum operating voltage level, respectively.
10.5.4 Determination of average measured winding temperature by the hot-resistance

method
The average measured temperature of a winding may be determined by either Equation (6) or Equation (7):
R
T = ------ ( T k + T o ) – T k
Ro
(6)
42
R–R
T = ---------------o ( T k + T o ) + T o
Ro
(7)
where
T is the temperature, in degrees Celsius, corresponding to hot resistance R

To is the temperature, in degrees Celsius, corresponding to cold resistance Ro
Ro is the cold resistance determined in accordance with the rules of this standard
R is the hot resistance
The time for the measuring current to become stable should be noted during the cold-resistance
measurements in order to assure that sufficient time elapses for the induction effect of the winding to
disappear before hot resistance readings are taken.
Record the elapsed time between the instant of shutdown and each hot-resistance measurement.
10.5.5 Correction of observed temperature rise for variation in altitude
When tests are made at an altitude not exceeding 1000 m (≈3300 ft) above sea level, altitude corrections
shall not be applied to the temperature rise.
When a shunt reactor, which is tested at an altitude less than 1000 m (≈3300 ft), is to be operated at an
altitude in excess of 1000 m (≈3300 ft), it shall be assumed that the observed temperature rise will increase
in accordance with the relation in Equation (8):
⎛ A ⎞
Increase in temperature rise at altitude A in meters (feet) = Observed rise ⎜⎜ −1⎟⎟ F (8)
A
⎝ o ⎠
where
Ao is equal to 1000 m (≈ 3300 ft)

F is the empirical factor given in the right-hand column of Table 6
Table 6 —Emperical factor for temperature correction for altitude
Method of cooling Empirical factor F
Oil-immersed, self-cooled (ONAN) 0.04

Dry-type, self-cooled (AN) 0.05
The observed rise in Equation (8) is as follows:
a) Top-oil temperature rise, or average-oil temperature rise, and winding-temperature rise over the
ambient temperature for oil-immersed shunt reactors
b) Winding-temperature rise over the ambient temperature for dry-type shunt reactors
43
10.5.6 Temperature-rise tests on all shunt reactors
10.5.6.1 Test conditions
All temperature-rise tests shall be made under normal (or equivalent to normal) conditions, based on the
methods of cooling.
10.5.6.2 Assembly for test
Shunt reactors shall be completely assembled and, if oil-immersed, shall be filled to the proper level.
10.5.6.3 Ancillary equipment
If the shunt reactors are equipped with thermal indicators, bushing-type current transformers, etc., such
devices shall be assembled with the shunt reactors.
10.5.6.4 Test room
The temperature-rise test shall be made in a room that is as free from drafts as practicable; typically the air
speed should be less than 0.5 m/s.
10.5.6.5 Ambient temperature determination
The temperature of the surrounding air, the ambient temperature, shall be determined by at least three
thermocouples or thermometers spaced uniformly around the shunt reactor under test. They should be
located at about one-half the height of the shunt reactor, and at a distance of 1 m to 2 m (3 ft to 6 ft) from
the shunt reactor. They should be protected from drafts and abnormal heating.
To reduce to a minimum the errors due to time lag between the temperature of the shunt reactor and the
variations in the ambient temperature, the thermocouples or thermometers shall be placed in suitable
containers. These containers shall have such proportions as will require not less than 2 h for the indicated
temperature within the container to change 6.3 °C if suddenly placed in air that has a temperature 10 °C or
less than the previous steady-state indicated temperature within the container.
10.5.6.6 Temperature rise of metal parts
When measured, the temperature rise of metal parts (other than the winding conductor) in contact with or
adjacent to insulation, and the temperature rise of other metal parts, shall be determined by thermocouple or
by thermometer.
Provisions shall be made to measure the surface temperature of iron or alloy parts surrounding or adjacent
to the outlet leads or terminals carrying currents. Readings shall be taken at intervals, or immediately after
shutdown.
The determination of the temperature rise of metal parts within the case, other than winding conductors, is a
special test. This test shall be made when so specified, unless a record of this test made on a duplicate, or
essentially duplicate, unit can be furnished. This test will not be made unless definitely specified, because
provision for the proper placement of the thermocouples and leads must frequently be made during the
design of the shunt reactor. Comparisons with other shunt reactors having metal parts of similar design and
arrangement, but not necessarily having the same rating, will, in many cases, be adequate.
10.5.6.7 Use of thermocouples
A thermocouple is the preferred method of measuring surface temperature. When used for this purpose, the
thermocouple should, when practical, be spot-welded to the surface. When this is not practical, the
44
thermocouple should be soldered to a thin metal plate or foil approximately 25 mm (1 in) square. The plate
is to be placed, and securely fastened, against the surface. In either case, the thermocouple should be
thoroughly insulated thermally from the surrounding medium, and care should be exercised to prevent the
solder and the spot weld from disturbing the accuracy of the thermocouple.
NOTE 1—The use of thermocouples can be hazardous due to parts being at high voltage. Other temperature measuring
methods may have to be used.
NOTE 2—To mitigate electromagnetic interference, thermocouples employing twisted-wire pairs and shielding, plus
time averaging interface circuitry, should be considered.
10.5.6.8 Use of restricted cooling
It is permissible to shorten the time required for the test by the use of restricted cooling, or any other
suitable method. If the determination of the thermal time constants is required by the user, then restricted
cooling cannot be used to shorten test time.
10.5.6.9 Determination of temperature rise of winding
The temperature rise of the windings shall be determined by the resistance method, or by thermometer
when so specified.
10.5.6.10 Ultimate temperature rise
The ultimate temperature rise is considered to be reached when the temperature rise becomes constant; that
is, when the temperature rise does not vary more than 2.5% or 1 °C, whichever is greater, during a
consecutive 3 h period (for oil-immersed shunt reactors), or 2 h period (for dry-type shunt reactors).
10.5.7 Temperature-rise tests on oil-immersed shunt reactors
10.5.7.1 General
The top-oil temperature shall be measured by a thermocouple or alcohol thermometer immersed

approximately 50 mm (2 in) below the top-oil surface.
The average temperature of the oil shall be determined when the average temperature method is used.
The average-oil temperature is equal to the top-oil temperature minus one-half the difference in
temperature of the moving oil at the top and the bottom of the cooling means, as determined by suitable
measurements.
For shunt reactors with external cooling means, this temperature difference may be closely approximated
by careful determination of the temperature on the external surface of the oil inlet and oil outlet of the
cooling means by the use of thermocouples.
10.5.7.2 Temperature-rise tests on oil-immersed self-cooled (Class ONAN) shunt reactors
The ambient temperature shall be taken as that of the surrounding air, which should be not less than 10 °C
or more than 40 °C. Corrections for variation of ambient temperature within this range shall not be applied.
Temperature tests may be made with ambient temperature outside the range specified if suitable correction
factors are available.
45
10.5.8 Temperature-rise tests on oil-immersed shunt reactors—Methods
10.5.8.1 Using the top-oil temperature rise
Determine the top-oil temperature rise over the ambient temperature in the following manner:
a) Apply maximum operating voltage at rated frequency to the shunt reactor.

b) Run until the top-oil temperature rise over the ambient temperature does not change more than
2.5% or 1 °C, whichever is greater, during a consecutive 3 h period.
c) Measure the ultimate top-oil temperature rise.
Determine the average temperature rise of the winding over the top-oil temperature in the following
manner:
1) Immediately after determining the top-oil temperature rise as described in step a), step b), and
step c), shut down and measure the winding resistance, and calculate the average winding-
temperature rise over the top-oil temperature at the end of the run.
2) Correct these rises back to the instant of shutdown.
The top winding-temperature rise over the ambient temperature is the sum of the top-oil temperature rise
over the ambient temperature plus the average winding-temperature rise over the top-oil temperature.
10.5.8.2 Using the average oil temperature rise
Determine the average oil temperature rise over the ambient temperature in the following manner:
a) Apply maximum operating voltage at rated frequency to the shunt reactor.

b) Run until the top-oil temperature rise over the ambient temperature does not change more than
2.5% or 1 °C, whichever is greater, during a consecutive 3 h period.
c) Measure the ultimate average oil temperature rise.
NOTE—Methods for this measurement are given in 10.5.7.
Determine the average temperature rise of the winding over the average oil temperature in the following
manner:
1) Immediately after determining the average oil temperature rise as described in step a), step b), and
step c), shut down and measure the winding resistance and calculate the average winding-
temperature rise over the average oil temperature.
2) Correct these rises back to the instant of shutdown.
The average winding-temperature rise over the ambient temperature is the sum of the average oil
temperature rise over the ambient temperature plus the corrected average winding-temperature rise over the
average oil temperature.
10.5.9 Temperature-rise tests on dry-type shunt reactors—Methods and correction factors
10.5.9.1 Ambient temperature correction
When the ambient air temperature is other than 30 °C, a correction factor shall be applied to the
temperature rise of the winding by multiplying it by the correction factor C, which is given by the ratio in
Equation (9):
46
T + 30 ° C
C= k (9)
Tk + Ta
where

Ta is the ambient air temperature, in degrees Celsius
When temperature-rise tests by the thermometer are required, place at least one thermometer in each coil
assembly. It is important that the coil thermometers be placed in the air ducts in such a manner as to
indicate the winding temperature without restricting the ventilation.
Once the temperature rise has become constant, the test current shall be interrupted. Immediately thereafter,
the coil thermometers, and any other temperature indicating devices, should be read continually in rotation
until the temperature begins to fall. If any of the thermometer temperatures are higher than those observed
during the run, the highest temperature should be recorded as the final thermometer temperature.
10.5.9.2 Temperature-rise tests on dry-type self-cooled (Class AA) shunt reactors
The ambient temperature shall be taken as that of the surrounding air, which should be not less than 10 °C
or more than 40 °C.
10.5.10 Correction back to shutdown—Cooling curve method
Take a series of at least four, preferably more, readings on each winding, and record the time after
shutdown for each reading.
The readings should be time spaced to assure accurate extrapolation back to shutdown.
The overall reading time should exceed 4 min, and may extend considerably beyond.
The first reading on each winding should be taken as quickly as possible after shutdown, but not before the
measuring current has become stable. The first reading shall be taken within 4 min.
Plot the resistance time data on suitable coordinate paper, and extrapolate the curve back to the instant of
shutdown.
The resistance value so obtained shall be used to calculate the average winding temperature at the instant of
shutdown.
The resistance time curve obtained on one phase may be used to determine the correction back to shutdown
for the other phases of the windings, etc., provided that the first reading on each of the other windings has
been taken within 4 min after shutdown.
NOTE—With the availability of spreadsheet software, it is very easy to make the extrapolation for the other phases.
If necessary, the temperature test may be resumed so that the first readings on any group of windings may
be completed within the required 4 min.
47
10.6 Audible-sound-level test
10.6.1 General
The audible sound for all dry-type and oil-immersed shunt reactors has one major source, which is the
magnetic force caused by the reaction between the winding current and the flux density in the medium or
composite medium. (This applies to all core and coil assemblies.) This force generates audible sound. Its
harmonics will be in discrete tones, whose frequencies are even multiples of the shunt reactor’s excitation
frequency. The audible-sound-level test should be performed per the sound pressure method discussed in
the following clauses. The sound intensity method may also be used per agreement between manufacturer
and end user.
NOTE 1—The sound level for oil-immersed and dry-type shunt reactors may vary with temperature.
The A-weighted measurement characteristic best relates how a remote listener hears the complex sound
generated by the shunt reactor, and shall be used to determine the average sound-level performance of the
shunt reactor.
For some purposes, a frequency distribution of a shunt reactor’s sound is desirable; when specified, it shall
be measured in frequency bands (either octave or one-third octave band), or as discrete frequencies.
NOTE 2—The sound level of a shunt reactor may be expressed in terms of either sound pressure level or sound power
level.
10.6.2 Instrumentation
10.6.2.1 Applicable instrumentation
Sound-level measurements shall be made with instrumentation that meets ANSI S1.4-1983 for Type-2
meters. Integrating sound-level meters are available and using a “walk around” measurement methodology
versus discrete microphone positioning allows an energy average sound level to be determined directly.
10.6.2.2 Octave or one-third octave band frequency measurements
Octave or one-third octave band frequency measurements shall be made, when specified, with
instrumentation that meets ANSI S1.4-1983 for Type-2 meters, together with ANSI S1.11-2004 for Type-
E, Class 11 performance, or their equal.
10.6.2.3 Discrete frequency measurements
Discrete frequency measurements shall be made when specified, or when necessary due to test conditions.15
Instrumentation is not standardized at present. However, typical analyzer bandwidth characteristics deemed
suitable are one-tenth octave, 1%, 3%, or 10% of the selected frequency, or 3 Hz, 10 Hz, or 50 Hz.
10.6.2.4 Use of suitable wind screen for microphone
A suitable wind screen may be used where the air velocity due to winds, prevailing drafts, or microphone
locations in the proximity of the shunt reactor cause the readings to be in error. Suitable corrections, if
necessary, shall be made for readings with wind screens to ensure that only the wind noise effects are
negated.
15
For a discrete frequency application, see Annex D and “Measurement of self-cooled transformer sound levels in relatively high
ambients” [B13].
48
10.6.3 Test conditions
10.6.3.1 Test environment
Measurements shall be performed in an environment having an ambient sound level that is below the
combined sound level of the shunt reactor and the ambient for the frequency band in which measurements
are being made. This level shall be at least 5 dB, but preferably >10 dB below the combined sound level.
The ambient sound level shall be established by averaging measurements taken immediately preceding and
immediately following the shunt reactor tests for at least four microphone locations, spaced equally around
the shunt reactor. For an average ambient sound level of 5 dB or more below the combined sound level of
shunt reactor and ambient, the following corrections shall be applied, according to Table 7.
Ambient sound corrections shall be governed by the average sound-level measurements with identical
frequency bandwidths for the combined shunt reactor and ambient sound, and the ambient sound alone.
Where ambient conditions differ from the above and are steady, suitable corrections may be feasible. The
details and method of making such ambient corrections shall be determined by those responsible for the
design and application of the shunt reactor.
Where the difference is less than 5 dB, and it is only desired to know a sound level that the shunt reactor
does not exceed, a correction of –1.6 dB may be used.
Table 7 —Correction to sound level
Difference between average sound level of combined Correction to be applied to average sound level
shunt reactor and ambient and average sound level of combined shunt reactor and ambient to obtain
of ambient (dB) average sound level of shunt reactor (dB)
5 1.6
6 1.3
7 1.0
8 0.8
9 0.6
10 0.4
Over 10 0.0
10.6.3.2
The shunt reactor shall be located so that no acoustically reflecting surface, other than the floor or ground,
is within 3.0 m (10 ft) of the shunt reactor.
10.6.3.3
The shunt reactor shall be connected for, and energized at, maximum operating voltage and rated
frequency. The audible-sound design test shall be carried out at close to operating temperature. This may be
most easily achieved by performing this test at the end of the temperature-rise test. Three-phase shunt
reactors shall be energized from a three-phase source and single-phase shunt reactors from a single-phase
source. When available test power is insufficient for testing at rated voltage, then the manufacturer shall
demonstrate to the user’s satisfaction that reduced-voltage testing produces sufficiently accurate results
when extrapolated to the rated voltage level. The manufacturer shall notify the user of reduced-voltage
testing during the proposal stages. If this cannot be demonstrated to the user, a field test can be performed.
49
A “walk around” routine test is performed on oil-immersed shunt reactors with the shunt reactors cold. The
“walk around” routine test is carried out with an integrating sound-level meter. In order to provide a
calibration point, a “walk around” sound test should be carried out at the beginning of the audible-sound
design test (usually done at the beginning of the temperature-rise design test) with the unit cold. Routine
audible-sound test results are compared to this “bench mark.”
NOTE 1—It may be advantageous to incorporate the sound-level test into the temperature-rise test, especially since this
allows the determination of sound level at close to operating temperature and thus will reflect in-service operating
conditions. For this reason, the audible-sound test should be performed at the end of the temperature-rise test. The
sound level so determined is at maximum operating voltage. If no temperature-rise-type test is carried out (and a
similar-type test report is not available) but an audible-sound-type test is required, sound-level measurements shall still
be made with the unit at close to operating temperature. This can be demonstrated when the measured sound level
stabilizes with ± 2 dB(A). The measured sound level can be used in calculating the sound level at nominal voltage.
NOTE 2—If the audible-sound design test cannot be performed at maximum operating voltage due to test lab
limitations, the manufacturer shall inform the end user at the bid stage.
NOTE 3—The audible-sound test is required to assess environment impact and as a quality or design verification
check. This usually implies that the audible-sound test should be carried out at maximum operating voltage. However,
in some cases, if audible-sound level guarantees are related to environmental impact, then it may be more appropriate
to determine audible sound at nominal/rated voltage. This is based on the assumption that shunt reactors are employed
to bring high system operating voltages, due to light loads, down to nominal/rated values. In this case, the audible-
sound test can still be carried out at the end of the temperature-rise test when the unit is at operating temperature and
the measured audible-sound level can be used to calculate the audible-sound level at nominal/rated voltage, or in fact,
the audible-sound level at any specified voltage can be calculated.
10.6.3.4
The audible-sound test shall be carried out at maximum operating voltage, however, when specified, sound-
level tests can be conducted at a specific voltage other than maximum operating voltage.
10.6.4 Microphone positions
10.6.4.1
The reference-sound-producing surface of a shunt reactor is a vertical surface that follows the contour of a
taut string stretched around the periphery of the shunt reactor or integral enclosure. This is to include
radiators, tubes, switch compartments, terminal chambers, etc., but excludes bushings and minor extensions
such as valves, oil gauges, thermometers, conduit terminal boxes, and projections at or above cover height.
In consideration of the safety and consistency of measurement, the reference-sound-producing surface near
unenclosed live parts of field-assembled items, such as switches, switchgear and terminal compartments or
wall-mounted bushings, SF6 air-to-oil adapter bushings, etc., shall be moved outward from the taut string
contour to be consistent with safe worker clearances, as determined by the manufacturer for the system
voltage of the live-part terminations involved.
10.6.4.2
For oil-immersed shunt reactors, the first microphone location shall coincide with the main drain valve (see
Figure 7). Additional points shall be located at 1.0 m (3.3 ft) intervals, proceeding clockwise in a horizontal
direction as viewed from above, along the reference-sound-producing surface defined in 10.6.4.1.
There shall be no fewer than four microphone location points, which may result in intervals of less than
1.0 m (3.3 ft) for small shunt reactors. The microphone shall be located on a straight line perpendicular to
the reference-sound-producing surface at each microphone location point. The microphones shall be spaced
0.3 m (1 ft) from the reference-sound-producing surface.
50
Figure 7 —Microphone locations for audible-sound tests of oil-immersed shunt reactors
10.6.4.3
For oil-immersed shunt reactors having an overall tank or enclosure height of less than 2.4 m (8 ft),
measurements shall be made at half-height. For shunt reactors having an overall tank or enclosure height of
2.4 m (8 ft) or more, measurements shall be made at one-third and two-thirds height.
10.6.4.4
For single-phase dry-type shunt reactors less than 2.4 m (8 ft) tall, microphone stations shall be at half-
height. For single-phase dry-type shunt reactors greater than 2.4 m (8 ft) tall, microphone stations shall be
at one-third and two-thirds height. For two- and three-coil stacked arrangements, microphone stations shall
be at mid-height of each reactor. If measurements at the previously mentioned heights are not possible due
to bus bar layout, microphone stations shall be located at the mid-height of the base reactor. In plan view,
the microphone stations for dry-type shunt reactors shall be laid out clockwise, sequentially along the
circumference of a circle having its center at the geometric center of the shunt reactor, and a radius equal to
the reactor radius plus 3 m (10 ft). The first station will be on a radial line through the bottom terminal, or
as close to it in the clockwise direction as is permitted to comply with minimum clearance distances to live
parts. For the case of tall coils, two-stack coils, and three-stack coils that can only be measured at one
height above grade, the radius of the circle along which measurements will be taken is equal to the reactor
radius plus 3 m (10 ft), or one-half of the overall reactor stack height, whichever is greater.
For side-by-side arrangements of single or stacked reactors, microphone stations are determined by the
same method as for a single coil or single stack, if the stations do not overlap. If the microphone stations do
overlap, measurements shall only be taken around the outermost perimeter of the resulting contour (see
Figure 8).
An integrating sound-level meter can be used in place of a standard sound-level meter. The energy average
sound level is recorded by traversing the reactor(s) envelope of the previously defined contours at a
constant rate of speed.
51
Figure 8 —Microphone locations for audible-sound tests of dry-type shunt actors
52
10.6.4.5
The magnetic field in the near vicinity of dry-type shunt reactors may be strong enough to affect the sound
instrumentation adversely. The sound-level meter should be kept at a distance from the shunt reactor that
will ensure that the ratio of sound pressure level to the electromagnetic noise susceptibility of the sound-
level meter shall not exceed 30 dB.
10.6.4.6
When sound-level tests are made at the factory, the mounting conditions that will be utilized at the final
installation should be simulated as much as practicable.
10.6.4.7
If it should become expedient to measure the sound level of a single-phase dry-type shunt reactor after it
has been installed as part of a three-phase bank of shunt reactors, then it will be desirable that those
responsible for the design and application of the apparatus agree upon a suitable technique of making
corrections for ambient sound of other phases, for reflecting surfaces, and for unavoidable alterations in the
microphone locations.
10.6.5 Sound-level measurements
10.6.5.1
Sound levels shall be measured in conformance with 10.6.1, 10.6.3, and 10.6.4 using the sound-level-meter
A-weighting characteristic.
10.6.5.2
The average A-weighted sound level is defined as the arithmetic mean of the respective A-weighted sound-
level measurements taken at each microphone location defined in 10.6.4.
10.6.5.3
When specified, measurements shall also be taken using the sound-level-meter C-weighting characteristic.
10.6.5.4
If necessary due to ambient conditions, the sound level may be measured using discrete frequency
components (see 10.6.6.2). 16
10.6.5.5
The shunt reactor should be located so that no acoustically reflecting surface, other than the floor or
ground, significantly affects the sound-level measurements. There should be no reflecting surfaces with a
normal projected area of more than one-fourth of the square of the distance between the unit and the
surface.
10.6.5.6
The unit shall be connected for, and energized at, rated voltage and frequency. The harmonic factor of the
voltage shall not exceed 1%.
16
See footnote 15.
53
NOTE—The harmonic factor is defined in IEEE Std C57.12.80-2002 and The Authoritative Dictionary of IEEE
Standards Terms [B20].
10.6.6 Optional frequency-analysis measurements
10.6.6.1 Weighting characteristic
When specified, frequency analysis shall be made in accordance with 10.6.2, 10.6.3, and 10.6.4 for either
octave, one-third octave, or discrete-frequency measurements.17 Either the A-weighting, the C-weighting, or
the flat-response-meter characteristic may be specified. The weighting characteristic used shall be reported
with the data. Integrating sound-level meters are now readily available and the use of such an instrument to
carry out a frequency analysis is the preferred test methodology.
10.6.6.2 Octave frequency analysis
Octave or one-third octave band frequency-analysis measurements shall cover the interval of midband
frequencies from 63 Hz through 4000 Hz. Discrete frequency-analysis measurements shall cover the
fundamental through the seventh-harmonic component (fundamental component is twice the excitation
frequency).
10.6.6.3 Calculation of average sound level for each frequency band
The average sound level for each frequency band measured shall be determined by taking the average of
the individual readings about the shunt reactor. Thus, Equation (10) follows:
⎛1 n ⎞
L X = 10 log10 ⎜
⎜n
⎝
∑10
i =1
0.1 Li ⎟ dB
⎟
⎠
(10)
where
LX is the average level for the X frequency band

Li is the level in the X frequency band at the ith measurement point
n is the total number of measurement points
If the components are with A weighting, the average is then with A weighting.
It should be noted that the value calculated using Equation (10) may have to be corrected for the following
factors:
a) Background or ambient noise level

b) Acoustic influences of the location where sound readings are taken, e.g., reverberant properties of
the test lab.
10.6.6.4 Calculation of sound level at each microphone position
If the average sound level in dB(A), as defined in 10.6.5.2, is to be determined from the individual A-
weighted frequency-band measurements at each microphone position, then an A-weighted sound level shall
be calculated for each microphone position from the frequency-band measurements as follows in
Equation (11):
17
See footnote 15.
54
⎛ m ⎞
L A = 10 log10 ⎜ ∑ ⎟ dB(A)
0.1 L j
10 (11)
⎜ ⎟
⎝ j =1 ⎠
where
LA is the average calculated A-weighted sound level

Lj is the band level with A weighting for the jth band
m is the total number of bands
It should be noted that the value calculated using Equation (11) may have to be corrected for the following
factors:
a) Background or ambient noise level

b) Acoustic influences of the location where sound readings are taken, e.g., reverberant properties of
the test lab.
10.6.6.5 Calculation of average sound level
An average sound-level value L A shall be calculated from the measured values of the A-weighted sound
level LAi by using Equation (12):
⎛1 N ⎞
LA =10 log10 ⎜
⎜N
⎝
∑10
i =1
0.1L Ai ⎟ dB(A)
⎟
⎠
(12)
where
LA is the average A-weighted sound level in dB(A)

LAi is the measured sound level at the ith measurement point in dB(A)
N is the total number of measurement points
10.7 Vibration tests on oil-immersed shunt reactors
10.7.1 General
The design and construction of oil-immersed shunt reactors should be such as to avoid the detrimental
effects of excessive stress due to vibration. Areas of primary concern in the control of vibration to ensure
proper performance are as follows:
a) Vibration of core and coil assembly
NOTE—Movement of the core and coil assembly and shielding structure, caused by the time-varying magnetic forces,
results in vibration of the tank and ancillary equipment. However, due to the variations in the present design practices,
it is not practical to define a test code for acceptance tests covering the vibration of the core and coil assembly.
Development tests are not included in the scope of this test code.
b) Vibration of tank with associated stresses developed in plates, braces, and welded seams
c) Vibration of instruments, accessories, and cooling equipment
55
10.7.2 Preparation for tests
The shunt reactor under test shall be completely assembled in normal operating condition with cooling
equipment, gages, and accessories mounted and connected.
The shunt reactor should be mounted on a level surface that will provide proper bearing for the base, in
order to eliminate the generation of abnormal tank stresses.
10.7.3 Method of measurement
The vibration of shunt reactor components shall be measured by transducers, optical detectors, or
equivalent measuring devices. The measuring equipment shall be accurate within ±10% at the second
harmonic of the exciting frequency. The peak-to-peak amplitude shall be determined by direct
measurement, or calculated from acceleration or velocity measurements.
10.7.4 Test conditions
The shunt reactor shall be energized at rated voltage and frequency. Three-phase excitation is required for
three-phase units. When available test power is insufficient for testing at rated voltage and/or three-phase
excitation, the manufacturer shall demonstrate to the user that reduced-voltage testing shall produce
sufficiently accurate results at rated conditions. The manufacturer shall notify the user of this condition
during the proposal stages. Factory tests at ambient temperature are acceptable, but vibration measurements
should be made, if a temperature-rise test is required, as soon as possible after the thermal test. If tests are
made at ambient temperature, readings should be taken as quickly as possible to minimize changes in the
temperature of the shunt reactor components.
10.7.5 Number of measurements
10.7.5.1 General
The minimum number of readings required will be 48. Additional readings will depend upon the size and
complexity of the shunt reactor core and coil construction, and the tank design.
10.7.5.2 Tank
Each of the four sides of the shunt reactor tank shall be divided into 12 rectangular areas of approximately
the same size. These areas are suitably marked and numbered for reference. Points of maximum excursion
for each marked area of the tank shall be located by appropriate methods for detecting vibration patterns in
the tank. Vibration readings shall be measured at these points and recorded in the final report. Readings
may also be taken and recorded along tank seams, cover, and any other locations mutually acceptable to the
shunt reactor manufacturer and user. When test conditions, such as safety regulations, do not permit
relocation of the measuring equipment when the shunt reactor is energized, those responsible for the design
and application of the shunt reactor will determine, in advance, a suitable technique for making the required
measurements.
10.7.5.3 Vibration amplitude levels
The average amplitude of all local maximum points shall not exceed 60 μm (2.36 mil) peak-to-peak. The
maximum amplitude within any rectangular area shall not exceed 200 μm (7.87 mil) peak-to-peak.
10.7.5.4 Instruments, accessories, and cooling equipment
Instruments, accessories, and oil-cooling equipment shall be observed for evidence of vibration during the
factory test (see 10.7.2, 10.7.3, and 10.7.4).
56
10.8 Vibration tests on dry-type shunt reactors
10.8.1 Design and construction of dry-type shunt reactors
The design and construction of dry-type shunt reactors should be such as to avoid the detrimental effects of
excessive stress due to vibration.
10.8.2 Test conditions and description
Since all parts of dry-type shunt reactors above base-support insulators are at elevated potential, a non-
energized test of natural frequencies shall be performed when specified.
This test shall consist of an impact force on the reactor components of concern, and the determination of
the component natural frequencies and percentage damping due to this impact load. An impact-load
hammer and accelerometers shall be used to record the impact force and response of the component being
tested.
The purpose of the test is to demonstrate that the natural frequency of the component is not coincident with
a frequency in the vicinity of double the system frequency.
10.9 Magnetic characteristic measurements
Due to iron-saturation effects, the inductance of oil-immersed shunt reactors is not linear. Therefore, the
inductance shall be determined as a function of the ac applied voltage from zero up to the specified
maximum overvoltage for which the linearity is specified.
If the laboratory equipment and/or the thermal performance of the reactor permit the use of an ac voltage
source at rated frequency up to the specified maximum overvoltage for which the linearity is specified, then
the measurement of the reactor linearity with an ac supply at rated frequency is preferred.
If the laboratory equipment and/or reactor design does not permit the use of an ac source at rated frequency,
the linearity of the reactor shall be determined by the incremental inductance method. By using this
method, the linearity measurement shall be performed by registering the current decay in the short-circuited
shunt reactor. The reactor shall be initially “charged-up” by using a dc current from a power rectifier to a
value of current equal to or greater than the corresponding ac current peak value related to maximum
overvoltage for which the linearity performance is guaranteed.
This will ensure that the inductance value at any applied voltage value, including overvoltage, can be
determined.
The reactor is then short-circuited through a current shunt and the dc current source is disconnected.
The current decay through the reactor shall be registered for several seconds with an appropriate sampling
interval in order to provide adequate information from which the curve of incremental inductance versus
current can be derived. At every instant of the recorded curve, the relation expressed in Equation (13) holds
true:
− L di =R i (13)
dt
where
R is the resistance of the discharge circuit including reactor winding, connecting leads, and
measuring shunt
57
L is the inductance
di/dt is the derivative of the current decay curve recorded
The circuit resistance shall be kept as low as possible and should be measured before and after the test in
order to take into account the effect of temperature rise on the discharge circuit components. The mean
value of the resistance measured before and after the test shall be used as the total circuit resistance value.
The slope of the current curve at every point represents the inverse value of the incremental inductance.
The incremental inductance at voltages lower than maximum operating voltage can be determined with an
ac source at rated frequency.
10.10 Seismic performance verification on oil-immersed and dry-type shunt

reactors
When specified, a seismic performance verification shall be carried out using analytical methods, by testing
under simulated seismic conditions, or by combined test and analysis such as that described in IEEE Std
693-2005. Although not directly applicable to electrical substation equipment, building codes such as the
Uniform Building Code [B25] are at times referenced regarding the seismic performance of shunt reactors.
11. Construction for oil-immersed shunt reactors
11.1 Bushings
11.1.1 Insulation class
Shunt reactors shall be equipped with bushings of an insulation class not less than that of the winding
terminal to which they are connected, unless otherwise specified.
11.1.2 Electrical characteristics
Electrical characteristics of outdoor shunt reactor bushings shall be as listed in IEEE Std C57.19.00-2004.
11.1.3 Dimensions
Bushings for use with outdoor shunt reactors shall have dimensions as listed in IEEE Std C57.19.01-2000.
11.2 Bushing-type current transformers
11.2.1 Dimensions
Bushing-type current transformers, used with bushings having dimensions in accordance with IEEE Std
C57.19.01-2000, shall have an inside diameter adequate to accommodate the maximum D dimensions for
those bushings, as shown in the applicable tables.
11.2.2 Specification
When specified, bushing-type current transformers, or provision for their addition in the future, shall be
multi-ratio with accuracy classification and taps, as specified by IEEE Std C57.13-1993.
58
11.2.3 Secondary leads
All bushing current transformer secondary leads shall be brought to an outlet box.
11.2.4 Terminal blocks
Nonsplit terminal blocks shall be provided in a weather-resistant case of the nonsplit type, located near the
shunt reactor base for terminating alarm circuits specified in 11.3.4, and current transformer secondaries
specified in 11.2.2.
11.2.5 Bushing-type current transformer removal provisions
On shunt reactors 1000 kVA and larger, provision shall be made for removing bushing-type current
transformers from the shunt reactor tank without removing the entire tank cover of the shunt reactor in
which they are to be used.
11.3 Accessories
11.3.1 Accessories to be provided
The accessories in the following table, described in 11.3.2 through 11.3.12, shall be provided.
Accessories Clause
Liquid-level indicator 11.3.2
Liquid-temperature indicator 11.3.3
Temperature and liquid-level
11.3.4
indicator alarm contacts
Pressure-vacuum gage 11.3.5
Drain and filter valves 11.3.6
Jacking facilities 11.3.7.3
Nameplate 11.3.8
Ground pad 11.3.10
Pressure-relief device 11.3.12
11.3.2 Liquid-level indicator
A liquid-level indicator shall be mounted so as to be readable at the level of the base. Dial markings shall
show 25 °C level and the minimum and maximum levels. The words “liquid level” shall be shown on the
face of the dial or on a suitable nameplate adjacent to the indicator.
11.3.3 Liquid-temperature indicator
A dial-type thermometer shall be mounted on the side of the tank.
The temperature indicator shall have resettable maximum temperature limits with corresponding contacts.
The thermometer shall be either a direct-stem-mounted unit or a temperature-sensing unit for remote eye-
level indication. Either unit shall be mounted in a closed well located at a suitable level to indicate the top-
oil temperature. For the dimensions of the well, see Figure 9.
The dial markings shall cover a minimum range of 0 °C to 120 °C. The words “liquid temperature” shall be
shown on the dial of the thermometer, or on a suitable nameplate mounted adjacent to the indicator.
59
Figure 9 —Dimensions of thermometer well
11.3.4 Temperature and liquid-level-indicator alarm contacts
11.3.4.1 Alarm contacts
Ungrounded alarm contacts for liquid-level indicators and temperature indicators shall be suitable for
interrupting
a) 0.02 A dc inductive load

b) 0.20 A dc noninductive load
c) 2.5 A ac noninductive or inductive load
d) 250 V maximum in all cases
The liquid-level-indicator alarm contacts shall be nonadjustable, and shall be set to close at the minimum
safe operating level of the liquid.
The liquid-temperature-indicator alarm contacts shall be adjustable over a range of 65 °C to 110 °C.
The winding-temperature-indicator alarm contacts shall be adjustable over a range of 95 °C to 125 °C.
11.3.4.2 Contact wiring and wire color coding
Contacts shall be in accordance with Figure 10, using cable with the color coding in Figure 10.
11.3.5 Pressure-vacuum gage
A pressure-vacuum gage shall be provided for shunt reactors of sealed-tank and gas-oil-sealed construction.
11.3.6 Drain and filter valves
A combination drain and lower filter valve, of the ball or globe type, shall be located on the side of the tank.
This valve shall provide for drainage of the liquid to within 25 mm (1 in) of the bottom of the tank.
The drain valve shall have a built-in 10 mm (3/8 in) sampling device that shall be located in the side of the
valve between the main valve seat and the pipe plug.
60
Figure 10 —Contact wiring and wire color coding
The device shall be supplied with a 5/16-in 32 male thread for the user’s connection, and shall be equipped
with a cap.
The size of the drain valve shall be 25 mm (1 in) for shunt reactors through 2500 kVA, and 50 mm (2 in)
for larger kVA ratings, and shall have NPT (national pipe thread taper) threads (in accordance with
ANSI/ASME B1.20.1-1983) with a non-ferrous metallic pipe plug in open ends.
Valves should not be located below any control cabinets. Shunt reactors through 2500 kVA shall have a
25 mm (1 in) upper filter plug, or cap, located above the maximum liquid level.
Shunt reactors above 2500 kVA shall have an upper filter valve, of the ball or globe type, located below the
25 °C liquid level.
The size of the upper filter valve shall be 50 mm (2 in), and it shall have 2-in NPT threads (in accordance
with ANSI/ASME B1.20.1-1983) with a nonferrous metallic-pipe plug in open ends.
11.3.7 Lifting, moving, and jacking facilities
11.3.7.1 Lifting facilities
Lifting eyes shall be provided for lifting the cover only.
Adequate facilities shall be provided for lifting the core and coil assembly from the tank. Lugs for lifting
the complete shunt reactor shall be provided. The bearing surfaces of the lifting lugs shall be free from
61
sharp edges, and each lifting lug shall be provided with a hole having a minimum diameter of 21 mm
(13/16 in) for guying purposes.
11.3.7.2 Moving facilities
The base of the shunt reactor shall be designed to permit rolling in the direction of center lines, and
provision shall be made for pulling the reactor in these directions.
The base should be designed so that the center of gravity of the shunt reactor, as normally prepared for
shipment, should not fall outside the base support members for a tilt of the base of 15° from the horizontal,
with or without oil in the shunt reactor.
11.3.7.3 Jacking facilities
Jacking facilities shall be located near the corners of the tank. Dimensions and clearances for jacking
provisions shall be as shown in Figure 11.
NOTE 1—Dimensions E, F, G, and H are free clearances.
NOTE 2—Where required in manufacturer’s standards designs, any dimensions may be in excess of those shown.
NOTE 3—E applies to nonremovable coolers only.
NOTE 4—Weight includes completely assembled reactor and fluid.
Figure 11 —Provision for jacking
62
11.3.8 Nameplate
A stainless-steel diagram nameplate (conforming to ASTM A167-1999, Grade 2) shall be provided. The
nameplate shall include the following information:
a) The words “shunt reactor”

b) Serial number (see NOTE 1)
c) Cooling class (see 3.6)
d) Number of phases
e) Frequency
f) Rated kVA (see NOTE 1)
g) Rated voltage (see NOTE 1)
h) Maximum operating voltage
i) Temperature rise(s), in degrees Celsius
j) Name of manufacturer
k) Year of manufacture
l) Measured impedance in ohms per phase at maximum operating voltage
m) Approximate masses in kilograms (pounds) (see NOTE 2)
n) Number of cubic meters (gallons) of oil (see NOTE 3)
o) Connection diagram (see NOTE 4)
p) Service conditions, if special
q) Reference to instruction book or sheet
r) BIL (see NOTE 5)
s) Symbols (see NOTE 7)
t) Patent numbers at manufacturer’s option
u) Winding material
v) Type of oil
w) Sound level (see NOTE 6)
NOTE 1—The height of letters and numerals showing kVA, serial number, and voltage ratings shall be 4 mm (5/32 in)
whether engraved or stamped. Height of other letters and numerals shall be optional with the manufacturer.
NOTE 2—The approximate weight in kg (lb) for the following items shall be shown for oil-immersed shunt reactors:
a) Core and coils
b) Tank and fittings
c) Oil
d) Total weight
e) Untanking (heaviest piece)
NOTE 3—The number of cubic meters (gallons) of insulating liquid, referred to elsewhere in this standard as oil, shall
be shown for the main tank and for each oil-filled compartment.
63
NOTE 4—All leads brought outside the tank and all windings shall be identified on the nameplate, or on the
connection diagram.
⎯ A schematic view shall be included to show the relative location of external leads and internal terminals.
⎯ All internal leads and terminals that are not permanently connected shall be designated or marked with
numbers or letters in a manner that will permit convenient reference, and will obviate confusion with
terminal and polarity markings.
⎯ Where development of windings is shown, the scallop symbol should be used in accordance with IEEE
Std 315-1975.
NOTE 5—The full-wave BIL, in kV, of line and of neutral terminals shall be designated.
NOTE 6—Noise level can be recorded if required.
NOTE 7—The following symbols shall be used where applicable:
11.3.9 Additional nameplate information
In addition to the information specified in 11.3.8, the following shall be included on the nameplate when
applicable:
a) Indication of potential transformers, potential devices, current transformers, voltage limiting

devices, winding-temperature devices, etc., when used.
b) Polarity and location identification of current transformers to be shown, if used for metering or
relaying. (Polarity need not be shown if current transformers are used for winding-temperature
equipment.)
c) Maximum operating pressures of oil preservation system—kPa (lbf/in2) positive, and kPa (lbf/in2)
negative.
d) Tank designed for—kPa (lbf/in2) positive, and kPa (lbf/in2) vacuum filling.
e) Oil level below top surface of the highest point of the highest manhole flange at 25 °C—mm (in).
Oil level changes—mm (in) per 10 °C change in oil temperature. (This applies only to shunt
reactors that have a gas cushion above the oil.)
11.3.10 Ground pads
Tank grounding provisions shall consist of two copper-faced steel pads or two stainless-steel pads without
copper facing, each 50 mm × 90 mm (2 in × 3½ in), with two holes horizontally spaced on 44.5 mm (1¾ in)
centers, and drilled and tapped for ½ in 13 National Coarse Thread. The minimum thickness of copper
facing shall be 0.4 mm (0.015 in).
64
Thread protection for the ground pad shall be provided.
Ground pads shall be welded on the base, or on the tank wall near the base, and shall be located diagonally
opposite from each other, so as not to interfere with the jacking facilities.
11.3.11 Control wiring
All control wiring shall have a minimum of 600 V insulation class.
11.3.12 Pressure-relief device
A pressure-relief device shall be provided on the shunt reactor cover.
11.4 Terminal markings
The winding or windings of a shunt reactor shall be distinguished from one another by marking each
terminal with an H followed by one of the numbers 1, 2, or 3, used as a subscript.
Any neutral end of the winding or windings of a shunt reactor shall be designated by using a 0, 01, 02, or
03, as a subscript for the H following the form of H0, H01, H02, or H03.
The terminal markings of a three-phase oil-immersed shunt reactor shall increase from right to left when
facing the highest voltage side of the tank, in the form H1, H2, H3.
A single-phase shunt reactor shall have the following terminal designation: H1, H0.
11.5 Oil preservation
The nitrogen for use with inert-gas-protected shunt reactors shall be dry nitrogen with less than 0.5% by
volume of impurities and less than 0.03% by weight of moisture.
The nitrogen shall be supplied in 5.66 m3 (200 ft3) cylinders equipped with connection no. 580 of
ANSI/CGA V-1-2005. The filling pressure is to be 15.2 MPa (2200 lbf/in2) at 21 °C. Before it is filled, the
cylinder shall be thoroughly cleaned of water or other impurities so that the nitrogen will not be
contaminated. The cylinder with outlet fitting, after being filled with nitrogen, shall be soap-film tested at
all joints for leaks. It shall not leak.
11.6 Oil-preservation systems
One of the oil-preservation systems described in 11.6.1, 11.6.2, or 11.6.3 shall be provided on shunt
reactors conforming to this standard.
11.6.1 Sealed-tank system
A sealed-tank system is one in which (1) the interior of the shunt reactor will be sealed from the
atmosphere throughout a top-oil temperature range of 100 °C, and (2) the gas-plus-oil volume will remain
constant, such that the internal gas pressure will not exceed 69 kPa (10 lbf/in2) positive or 55 kPa (8 lbf/in2)
negative.
A pressure-vacuum bleeder device set to operate at the maximum operating pressures (positive and
negative) indicated on the nameplate shall be furnished.
65
11.6.2 Inert-gas pressure system
An inert-gas pressure system is a system in which the interior of the shunt reactor will, by means of a
positive pressure of inert gas maintained from a separate inert-gas source and reducing-valve system, be
sealed from the atmosphere throughout a top-oil temperature range of 115 °C, and the internal gas pressure
will not exceed 34 kPa (5 lbf/in2).
11.6.3 Conservator or expansion-tank system
A conservator or expansion-tank system is one that, by means of an auxiliary tank partly filled with oil
connected to the completely filled main tank, seals the oil in the main tank from the atmosphere by means
of an air cell or diaphragm throughout a top-oil temperature range of 115 °C. The internal top-oil pressure
in the main tank will not exceed 34 kPa (5 lbf/in2).
11.7 Tanks
11.7.1 Operating pressures
Maximum operating pressures (positive and negative), for which the shunt reactor is designed, shall be
indicated on the nameplate. The main shunt reactor tank, and any compartment attached thereto that is
subject to the operating pressures, shall be designed to withstand, without permanent deformation, a
pressure 25% greater than the maximum operating pressures, 18 resulting from the system of oil preservation
used.
11.7.2 Vacuum filling
Tanks shall be designed for vacuum filling (essentially full vacuum) in the field on all shunt reactors with a
high-voltage system-voltage level of 69 kV and above, and on all shunt reactors rated 10 000 kVA and
larger, any system-voltage level.
11.7.3 Cover construction
A bolted or welded main cover shall be provided.
11.7.4 Core ground
A single-core ground shall be provided and accessible without removing oil.
11.7.5 Manholes
Shunt reactors 1000 kVA and larger shall have manholes in the cover. Manholes, if circular, shall be a
minimum of 460 mm (18 in) in diameter. If rectangular or oval, they shall have minimum dimensions of
360 mm × 460 mm (14 in × 18 in).
11.8 Shunt reactor finish
The finish for shunt reactor tanks shall consist of a pigment paint. 19
18
Individual designs may not necessarily reach the maximum pressures indicated in the definition of oil preservation systems.
19
Metallic flake paints, such as aluminum, zinc, etc., have properties that increase the temperature rise of shunt reactors, except in
direct sunlight. Temperature limits and tests are based upon the use of a pigment paint finish.
66
11.9 Other equipment accessories
11.9.1 Power supply voltage for shunt reactor controls
The power supply voltage for the shunt reactor controls shall be provided by the user.
11.9.2 Alarm and protection devices
Devices required for the protection of shunt reactors are listed as follows:
a) A thermally operated alarm-circuit device, with the thermal element mounted in a well and
responsive to the top-oil temperature of the shunt reactor. The device should have two sets of
contacts with factory settings as given in the following table:
Contact Function
1 Spare
2 Initiate alarm or actuate relay
Non-grounded alarm contacts shall be suitable for interrupting:

1) 0.02 A dc inductive load
2) 0.20 A dc non-inductive load
3) 2.5 A ac inductive or non-inductive load
4) 250 V maximum in all cases (ac or dc)
b) A weather-resistant cabinet enclosing the switching equipment, located on the shunt reactor at a
height suitable for operation by a person standing at the level of the base.
c) Wiring for control.
11.9.3 Surge arresters
The following types of construction are available for surge protection:
a) Provision only for the mounting of surge arresters.

b) Mounting complete with surge arresters.
c) Surge-arrester ground pad, consisting of a tank-grounding pad (in accordance with 11.3.10)
mounted near top of the tank, may be specified for each set of arresters. Exception: Where the
separation of the arrester stacks is such that individual pads for grounding each phase arrester
represent better design, individual ground pads may be supplied.
NOTE—Material for connecting surge arresters to live parts and to ground pads is not included in 11.9.3.
12. Construction and installation of dry-type shunt reactors
12.1 General description
All parts of dry-type shunt reactors are “live,” unlike oil-immersed units where the tank is grounded. The
only external live parts of an oil-immersed shunt reactor are the bushings.
67
Dry-type shunt reactors do not have an iron core. Therefore, the magnetic field is not constrained, and it
will occupy the space around the dry-type reactor. Although the magnetic field reduces in strength with
increase in distance from the reactor, the presence of this field shall be taken into consideration in many
dry-type shunt reactor installations.
Key issues are discussed in 12.2 through 12.8, and some guidance is given to the user of dry-type shunt
reactors.
12.2 Safety
A dry-type shunt reactor is not enclosed in a grounded steel tank. All parts of the reactor must be
considered to be live. This also includes the situation in which the breaker used to switch the shunt reactor
is located at the neutral end. Thus, even when the reactor is not carrying current, it is floating at line
potential (unless disconnected on the line side).
Therefore, dry-type shunt reactors are highly recommended to be installed such that accidental contact by
station personnel is not possible. Two methods that can be used to achieve this are fencing, and elevating
the reactor at a safe distance above ground.
When fencing is employed, consideration shall be given to the stray magnetic field of the air-core reactor.
Metallic fencing should be broken up into electrically insulated sections if it is located very near the
reactor. Consideration should also be given to the fact that when metallic fencing is used in high-voltage
substations, capacitively coupled voltage may appear on the fence sections. Grounding of the fence sections
is important. Care should be taken not to form shorted electrical loops (continuous conducting path linked
by the magnetic flux of the reactor) by using a single ground connection per section of fence. The use of
non-metallic fencing, which is now readily available, can eliminate many of the induced current problems.
Another option that is widely used is to elevate the reactor a distance off the ground so that live parts are not
accessible to station personnel. Typically, 8 ft 6 in pedestals are used. This meets the minimum clearance
requirements of the National Electrical Safety Code® (NESC®) (Accredited Standards Committee C2-2007
[B1]).20 Reactors can be mounted on concrete pilings (care shall be taken to ensure that excessive heating of
reinforcement bar does not occur), fiberglass pedestals, metallic structures specially designed such that closed
loops do not occur, etc. It should be emphasized that in all cases, the supporting structure shall also be
appropriately designed to avoid eddy-current heating of metallic parts. Other factors, such as correct selection
of materials (stainless steel) and orientation, shall be observed. Otherwise, very high temperatures could be
encountered that would be hazardous to station personnel and/or cause weakening of the support structure.
When employing dry-type shunt reactors, care should be taken in the installation of the station ground grid in
the vicinity of the reactors. The ground grid should be designed so as not to have shorted loops. Otherwise,
currents could be induced in the grid. Grounding of other ancillary support structures or equipment in the
vicinity of the reactors should be accomplished without creating closed loops in the grounding system.
In all, it is preferable that support structures and fencing be designed or reviewed by the manufacturer of
the shunt reactor.
12.3 Magnetic clearances
Since dry-type shunt reactors have no magnetic core, the magnetic field occupies a space around the reactor
and, depending on the MVA of the unit, can be of a substantial strength even at some distance from the
reactor. This alternating magnetic field can induce currents in nearby metallic geometries.
20
National Electrical Safety Code and NESC are both registered trademarks and service marks of the Institute of Electrical and
Electronics Engineers, Inc.
68
There are some simple rules of thumb that can be employed. Clearance to small metallic parts not forming
closed loops should be at least one-half the coil diameter radially from the edges of the reactor. Larger
geometries or closed loops should be located at least one coil diameter from all the surfaces of the reactor.
These are rules of thumb that generally can keep the user out of trouble. However, it is advisable that the
manufacturer be consulted. They should have at their disposal very accurate field-plot programs and
sophisticated analysis tools to calculate losses and temperature rise in metallic geometries located in
magnetic fields.
What this means is that circuit breakers, equipment housings, current transformers, surge arresters, and
other equipment should be safely located to ensure that the magnetic field of the reactor does not adversely
affect equipment performance. In addition, any metallic support structures for bus bar, etc., should also be
designed to avoid overheating due to induced eddy currents.
It should be emphasized that in cases where space is limited, the previously presented “rules of thumb”
may be substantially reduced. With careful review by the manufacturer, the use of special materials such as
fiber-reinforced plastics, austenitic stainless steel, and nonmagnetic shielding can allow the trouble-free
installation of dry-type shunt reactors in limited space.
12.4 Connections
Since bus connections with dry-type shunt reactors are metallic, care should be taken in properly selecting
the type of bus, since it is also exposed to the magnetic field of the reactor and thus can be susceptible to
eddy-current heating. As a rule of thumb, the eddy losses in any metallic geometry in a magnetic field are
proportional to the major geometry to the fourth power. Therefore, large tubular bus would have
significantly higher eddy losses than stranded cable. In the case of stranded cable, eddy losses are
substantially reduced due to the stranding. Bus bar can be used if it is designed to be streamlined to the
magnetic field (i.e., rectangular bus appropriately orientated). A good concept is to use bus bar for longer
runs, and make the final 2 m to 3 m (6 ft to 9 ft) of connection to the reactor using stranded cable.
By using cable for the last 2m to 3 m (6 ft to 9 ft), eddy losses can be kept to a minimal level. Beyond the
2 m to 3 m (6 ft to 9 ft) distance, the field of reactors should be substantially reduced.
If terminal temperature rise is a concern, the utilization of higher-ampere-capacity cable than is required
will effectively heat sink the terminal area and maintain a low temperature rise in a bolted connection.
However, it must be emphasized that the terminal temperature rise is a function of the contact resistance,
throughput current, and the eddy heating.
The connector itself is also very important if terminal temperature rises are to be kept within bounds. The
connector used should also attempt to have geometries that are streamlined to the magnetic field.
Connectors that have large contact surface area, but do not provide a large frontal area to the magnetic
field, are preferable.
Again, the manufacturer of dry-type shunt reactors is the best source of information regarding the type of
connector and bus run to be used when connecting dry-type shunt reactors.
It should be recognized that not all of the previous precautions will be necessary on every shunt reactor
installation. Generally, the larger the MVA rating, the more important these factors become. For example,
for units rated roughly 15 MVA and less, many satisfactory installations exist where tubular bus is used for
direct connection to the reactor terminal without problems.
12.5 Installed sound level
The measurement of sound level is a test sometimes performed on dry-type shunt reactors as it is with oil-
immersed equipment. How the reactor is installed will also have an impact on observed sound level. This is
69
true also for oil-immersed equipment. Foundation design, ground cover, and nearby reflecting surfaces can
all have an impact on the perceived sound level. Avoidance of standing waves is very critical to achieving
low sound level.
12.6 Concrete foundations
There are several critical factors in designing foundations for dry-type air-core shunt reactors. As with any
piece of large equipment, a rule of thumb to ensure that no vibration occurs is to design the concrete
support base to have at least triple the mass of the equipment to be installed. If bedrock at site is very near
the surface, or if the soil conditions warrant it, a base of reduced mass can be used.
Another factor that is an extremely critical consideration is the use of rebar. In most cases, the distribution
of the coil mass on the concrete base results in the concrete being very lightly loaded. Therefore, rebar may
not be necessary. If rebar is necessary, nonmetallic or stainless-steel rebar should eliminate potential eddy
heating problems. In addition, where metallic rebar is used, the crossover points should be electrically
isolated to prevent closed loops. Pieces of hose slipped over the rebar at the crossover is sufficient. If
clearances greater than the coil diameter are used below the reactor, then precautions of isolating the rebar
to prevent shorted loops and selecting stainless-steel or nonmetallic material may not be necessary.
The final major consideration is the anchoring system used to secure the coil-support structure to the
concrete base. The anchors must be located deep enough in the concrete, and be designed to resist the
overturning load imposed on the coil and its structure due to wind loading or seismic excitation. This is
most important for reactors located on top of tall support structures (e.g., fiberglass pedestals) to give
required clearance for station personnel.
Again, the manufacturer of dry-type shunt reactors is the best source of definitive information.
12.7 Switching—Circuit breakers
The type of breaker used and the location of the breaker are very complicated subjects. Breakers can be
located on the line side of the reactor. Many installations utilize a breaker on the neutral side of the reactor.
Another factor is how often the reactor is switched. This will dictate the type of breaker used for the operation.
More detailed discussions in this area are beyond the scope of this document. However, one of the most
significant considerations is that circuit-breaker interrupting characteristics are very much dependant upon
circuit-breaker technology and the electrical characteristics of the circuit. The high-frequency
characteristics of station bus work shall be taken into consideration. In conjunction with this, it is also
reasonable to ensure that detailed knowledge of the electrical characteristics of the reactors at frequencies
near the natural frequency of the reactors be incorporated into any decision-making process.
The previous comments apply equally to both oil-immersed iron-core shunt reactors and dry-type shunt
reactors. See Annex B for a more in-depth tutorial on the dielectric stresses imposed on shunt reactors
during switching. IEEE Std C37.015-1993, IEEE Application Guide for Shunt Reactor Switching, should
also be consulted.
12.8 Protection practices for air-core shunt reactors
While modern dry-type air-core reactors are generally very reliable pieces of equipment that seldom fail, it
is advisable to take at least some precautions in their protection to minimize the extent and cost of potential
coil failure. IEEE Std C37.109-2006, IEEE Guide for the Protection of Shunt Reactors, should be consulted
for information on protection practices.
Many alternative protection schemes exist, each with an associated cost.
70
When protecting dry-type shunt reactors, the protection engineer should endeavor to provide a low-cost
scheme, which ensures that failure of one unit in a three-phase bank will not result in consequential failure
of the units in adjacent phases. This objective should be in addition to the common priority of ensuring that
a coil fault will not damage the associated transformer.
Sometimes shunt reactors are supplied in a “double neutral” or split phase configuration and additional
testing may be required—including determination of inherent unbalance.
71
Annex A
(normative)
Specific requirements for dry-type air-core thyristor-controlled reactors
A.1 General
A.1.1 Scope
The information in this annex applies to dry-type air-core thyristor-controlled reactors (TCRs) used in an
SVC. SVCs are shunt-connected devices with a primary function of supplying vars to an electrical system.
Background information can be found in supporting documents IEEE Std 1031-1991 [2] and the CIGRE
report, “Static VAR Compensators” [3].21
SVCs are used to either
⎯ Control the voltage of transmission class electrical power systems by supplying lagging or leading
vars within the operation range of the SVC; specifically to reduce system overvoltages caused by
line switching and load rejection, to provide dynamic voltage support or damping, and to provide
other functions as specified, or
⎯ Compensate for the reactive power demand of large dynamic loads such as electric arc furnaces and
rolling mills in steel works; specifically to correct the power factor, to mitigate flicker which is a
result of supply voltage variations, to provide dynamic voltage support, etc.
SVCs typically consist of fixed or mechanically switched capacitors (which may be part of the harmonic
filtering scheme) or thyristor switched capacitors, and shunt reactors connected line to line and in series
with firing delay angle-controlled thyristors.
NOTE—The reference system frequency used in the annex for TCRs is 60 Hz. Since this standard is used outside
North America, 50 Hz may be the system operating frequency. In this case, calculations may have to be modified
accordingly.
A.1.2 Annex precedence
Where information in this annex conflicts with that provided elsewhere in this standard, the annex
information shall be considered to take precedence for dry-type air-core TCRs.
A.2 Definitions
A.2.1 “Effective” resistance
Effective resistance is the value of resistance obtained by dividing total losses by the current squared at a
specific frequency.
21
In this annex, the numbers in brackets correspond to those of the supporting published material in A.9.
72
A.2.2 Rated inductance
Rated inductance is the inductance value at power frequency, on a single phase basis.
NOTE—Two stacked, electrically coupled reactors may be employed to make up one single phase unit. For two
stacked reactors connected line to line, the rated inductance reported on each coil nameplate is the self-inductance of
that coil plus mutual inductance.
A.2.3 Rated current
The rms value of maximum power-frequency current and, when applicable, simultaneous harmonic
currents, continuously flowing through the reactor.
NOTE—Typically on the rating plate, due to space limitations, the currents shown are the fundamental and the RSS
(root sum square) of fundamental plus all harmonics whereas on drawings, where more room is available, fundamental
current plus major harmonics (corresponding to the operating point that produces the highest losses) are shown.
A.3 General requirements
A.3.1 General
A TCR consists of an air-core reactor of fixed inductance and a bi-directional thyristor valve. In a thyristor
valve, the thyristors are connected in series to meet the required blocking voltage levels at a given power
rating. A thyristor valve is brought into conduction by simultaneous application of a gate pulse to all
thyristors of the same polarity. The valve will automatically block immediately after the ac current crosses
zero. Current conduction will resume in the reverse direction when the gate signal is applied to the
thyristors of reverse polarity in the valve.
The current in the reactor can be controlled from maximum (thyristor valve fully conducting) to zero
(thyristor valve blocked) by delay angle control; firing of the thyristor valve is delayed with respect to the
peak of the applied voltage in each half cycle, thus the duration of the current conduction intervals are
controlled. The magnitude of the reactor current can be varied continuously by increasing the delay angle
from 90 electrical degrees to 180 electrical degrees. The firing angle control results in a nonsinusoidal
current waveform through the reactor; the TCR generates odd harmonics of positive and negative current
half-cycles that are identical. The TCR shall be designed for full fundamental current and the full range of
reduced fundamental current plus corresponding harmonics.
NOTE—When full conduction is not allowed, the TCR shall be designed for the maximum fundamental current
allowed by the valves plus the corresponding harmonic currents at that particular firing angle. This condition shall be
clearly stated in the TCR specification.
A.3.2 Design
TCRs are typically connected line to line in a delta configuration with either one coil or two coils
connected in series between lines. TCRs may be a single coil per phase, two coils per phase (side-by-side
mounting), or two coils per phase (two coils stacked). Typically each phase reactor is divided into two
coils—one on each side of the thyristor valve—in order to limit the short-circuit currents resulting if one of
the coils were to become shorted. The reactor(s) also limit the di/dt seen by the thyristors in the valve when
they are fired. The di/dt can be limited by just one single reactor but is not as effective as a two-coil-per-
phase configuration due to ground capacitance effects associated with the split inductance.
73
NOTE 1—Air-core TCRs are surrounded by a stray magnetic field created by the winding ampere turns. The location
of the reactor relative to metallic structures should be considered with regard to inductive heating effects during normal
operation. See 12.3 for further information.
NOTE 2—In the case of two-coil stacked configured TCRs, the magnetic coupling should be considered in the design
value of the per-coil self-inductance.
A.3.3 Design consideration for audible sound
Due to the presence, at specific operating points, of large harmonic currents that produce forced vibrations
in the more sensitive human audible range of the sound spectrum, the acoustic noise of TCRs may be quite
high. TCRs shall therefore be designed so as to avoid, as far as possible, mechanical or acoustic resonances
at major audible sound frequencies, especially near specified harmonic currents.
When a TCR is provided with a sound mitigation enclosure, special attention shall be given to the design
and manufacture of electrical connections inside the enclosure, as they cannot be inspected during
operation. Fire retardant materials should be used in the sound enclosure.
A.3.4 Environmental aspects
Since the maximum voltage drop may be as high as the line-to-line voltage across a TCR, the
environmental conditions shall be stated clearly in the inquiry, e.g., degree of salt contamination or
humidity, location of site if in the tropics or close to the sea, installation in desert areas or in heavy
pollution areas such as steel mills. The environmental aspects shall be considered in the design, for
example, in selecting materials and dimensioning of surface (e.g., creepage distance) and turn-to-turn
stresses, to ensure safe operation during the expected lifetime of the reactor. IEC/TR 60815:1986 is a useful
reference document as it contains a table listing ESDD levels versus site descriptors.
In addition, other environmental conditions such as actual maximum and average ambient temperatures at
the site (versus standard weather bureau data), seismic requirements, wind loads, etc., should be included in
the reactor specification. IEEE Std 693-2005 provides information regarding seismic design/qualification
requirements.
A.4 Ratings
Reference to 60 Hz voltages and currents apply equally to 50 Hz (including associated harmonics) systems.
A.4.1 Voltage drop rating
The voltage drop across the reactor is due to the rated 60 Hz current or reduced 60 Hz current and
corresponding harmonic currents. The voltage drop is a function of the magnitude of the current
components and their phase relationships. See Figure A.1.
74
Figure A.1—Example of the voltage across a TCR and the corresponding current
through the reactor (delay angle……36°)
The graph is an example of the voltage across a TCR and the corresponding current through the reactor
(delay angle…………36°)
The voltage drop at any operating point can be determined by the RSS method using 60 Hz and harmonic
frequency voltage drops, e.g., as follows in Equation (A.1):
1/ 2
⎛ ∞ ⎞
VD RSS =⎜
⎜ ∑VD 2H ⎟
⎟
(A.1)
⎝ H =1 ⎠
where
VDRSS is the RSS voltage drop

VD is the voltage drop at harmonic H
H is the harmonic number
NOTE—The value of voltage drop obtained by the “RSS” method is usually lower than the peak / 2 voltage but is
typically used to evaluate voltage stresses in the reactor windings and across the reactor surface.
A.4.2 Current rating
Fundamental 60 Hz current, and reduced fundamental plus corresponding harmonic currents where
applicable, shall be specified. This is critical as the current loading spectrum is required for both thermal
and acoustic design purposes.
75
A.4.3 Thermal short-time and mechanical short-circuit current rating
In the case where two coils are employed to provide the phase reactance the thermal short-time and
mechanical short-circuit current rating for the individual coils is effectively the resultant current obtained
when one of the coils is shorted. The reactor specification should provide the relevant current magnitude
and duration. Since the magnitude is on the order of two times the rated fundamental current, short-circuit
testing or verification are not usually required due to the very low mechanical stresses developed under this
condition.
NOTE—An overcurrent can be seen by the TCRs due to a misfiring of the thyristor valve. The magnitude of the
current is usually smaller than that associated with a shorting of one of the TCRs.
A.4.4 Rated losses of a thyristor-controlled reactor
For dry-type TCRs, the total losses shall be taken as the sum of the conductor I2R fundamental and
harmonic losses, the conductor fundamental and harmonic eddy losses and the fundamental and harmonic
stray losses in the windings and manufacturer supplied framework, but not including stray (eddy) losses in
the user supplied support structure, surrounding structures (bus supports, etc.), or mounting pad (rebar).
A harmonic current spectrum should be provided to the reactor manufacturer for losses evaluation
purposes. Usually, the current spectrum used for losses evaluation is less stringent than the spectrum
defined for assigning the current rating of the TCR. The losses evaluation current spectrum should
represent a realistic daily average loading of the reactor or losses for evaluation purposes can be based on
duty or time at various current loadings. If a loss evaluation spectrum is not provided, it is assumed that the
rated current spectrum is to be used for losses evaluation. Unless otherwise specified by the user in the
technical specification, the reference temperature for losses shall be the actual average winding rise, as
determined by the temperature-rise test plus 20 °C.
A.4.5 Reactor duty for rating
Unless otherwise specified, the reactor duty for rating is continuous operation at rated power frequency
current (full thyristor conduction). Where full conduction is not allowed, reactor duty for rating is
continuous operation at maximum power frequency current plus associated harmonic currents.
A.4.6 Limits of temperature rise for continuous rating
Limits of temperature rise for TCRs at rated current loading shall be as given in Table 3.
A.4.7 Reference temperature
The reference temperature for a TCR shall be the average temperature rise of the reactor at rated current
plus 20 °C. The temperature rise shall be based upon the temperature-rise test or calculations made by the
manufacturer when temperature-rise test results are not available.
NOTE 1—In some cases, if loss evaluation is not a consideration or it is required to simplify losses calculation and
evaluation, a specification may state that the reference temperature should be 75 °C.
NOTE 2—For loss evaluated (guaranteed losses) TCRs operating at low average temperature rise, the use of actual
average rise plus 20 °C for the correction of losses is more representative of service conditions.
76
A.4.8 Insulation between the two coils of a phase reactor and to ground
When the thyristor-controlled phase reactor is split into two coils, insulators are most often provided
between the coils in a two-coil stack configuration. These insulators shall be rated based on the phase-to-
phase voltage. In the case of the insulators at the bottom of the coil stack or the insulators used to mount
side-by-side configured reactors, this rating shall be consistent with the line-to-ground system voltage
requirements.
A.4.9 Insulation across the coil
The insulation system of a TCR should be designed to withstand both the steady-state voltage drop caused
by the 60 Hz and harmonic currents, and the transient overvoltages caused by lightning and switching.
Depending on the physical configuration of the TCR(s), the insulation levels across the reactor, the
insulation requirements between two stacked coils, the insulation level between the bottom coil of a two-
coil stack, and the insulation level between a single coil per phase-to-ground shall be considered as
required. BIL and SIL (switching impulse levels) across the reactor shall be included in the TCR
specification, as applicable.
A.4.10 Mechanical considerations in the support structure design
In the case where thyristor-controlled phase reactors are supplied in a two-coil stack configuration, care
shall be taken in the design of the intercoil and bottom coil to ground support structure. The coupling
between the two coils can result in large cyclical attractive forces, which may, if proper mechanical design
mitigation is not employed, cause severe vibration and fatigue-related problems.
A.5 Testing
A.5.1 General
All routine tests may be carried out on a per-coil basis. Type tests for two-coil stacked TCRs should be
carried out on a two-coil configuration, if applicable.
A.5.2 Routine tests for thyristor-controlled reactors
Routine tests for TCRs are as follows:
a) Measurement of winding dc resistance

b) Measurement of inductance
NOTE 1—In the case of stacked TCRs, inductance is measured on a per-coil basis but including coupling. In other
words, the inductance is measured for each individual reactor with the two coils energized in the stacked configuration.
c) Measurement of total losses
NOTE 2—Losses are usually measured on a per-coil basis without structure; the commercial guarantee, if any, is
usually on a per-coil basis. When requested losses can be measured with coils in a stacked configuration, including
structure.
d) Lightning impulse test or turn-to-turn overvoltage test depending on the specified nominal
insulation voltage level (see Table 5)
77
NOTE—For TCRs employed at system voltages 34.5 kV and lower, the turn-to-turn test (as per 10.3.3.3) is the
recommended dielectric withstand test. Based on test equipment currently available and in use, the time to the first crest
is typically on the order of 0.1 μs. This value is an order of magnitude smaller than the 1.2 μs time to crest for a
standard lightning impulse. The voltage distribution is slightly more nonlinear versus that for a lightning impulse but is
not significantly different due to the fact that the capacitance to ground of dry-type TCRs is on the order of hundreds of
picofarads; self- and mutual inductances and interturn capacitance dominate the high-frequency equivalent circuit and
results in an almost linear voltage distribution. In any case, the turn-to-turn test results in higher stresses in the end
turns.
A.5.3 Design tests for thyristor-controlled reactors
Design tests are:
a) Temperature-rise test
A.5.4 Other tests for thyristor-controlled reactors
“Other” tests are:
a) Switching impulse test to ground (insulators)

b) Applied potential test (insulators)
c) Chopped-wave impulse test
d) Audible-sound test or calculation
e) Seismic verification test or calculation
f) Thermal short-time current test or calculation
g) Mechanical short-circuit test or calculation
h) Measurement of variation of inductance and resistance with frequency
A.5.5 Notes on routine tests for thyristor-controlled reactors
A.5.5.1 Measurement of winding dc resistance
The method of measurement is as generally outlined in 10.2.
A.5.5.2 Measurement of inductance
The inductance shall be measured at the rated frequency only.
A.5.5.3 Measurement of total losses and Q-factor
The method of measurement is as generally outlined in 10.4. The losses of an air-core reactor at a given
frequency are assumed to be proportional to the square of the current and may be measured at reduced
current (or voltage).
The “effective” resistance shall be determined at 60 Hz and at rated harmonic frequencies at any
convenient ambient temperature. The resistance shall be corrected to the reference temperature specified in
A.4.7.
78
The inductance and the Q-factor at a specific frequency are derived from the loss and impedance test by use
of Equation (A.2) and Equation (A.3):
1 Ex
L= ⋅ (A.2)
2π f I
Ex I
Q= (A.3)
Pz
where
f is the frequency, in hertz

L is the inductance, in henries, at frequency f
Q is the Q-factor at frequency f at reference temperature
Ex is the reactance voltage, in volts (see 10.4)
I is the current, in amperes (see 10.4)
Pz is the total losses at frequency f, in watts, corrected to the reference temperature specified in
10.4.1
NOTE—Pz = Pr + Ps, where Pr is the I2R losses and Ps is the stray losses.
If a bridge method is employed, inductance and losses at specific frequencies can be obtained directly from
the measurement.
Test arrangements should be agreed between the manufacturer and the purchaser prior to testing.
The measurement of the losses shall be performed after the dielectric testing and, if so specified, be
performed prior to the dielectric test.
The power-frequency winding loss is determined by measurements at power frequency. Losses for
specified individual harmonic frequency components are calculated or directly obtained from
measurements at the respective harmonic frequencies and added to the power-frequency loss. The total loss
is referred to reference temperature as specified in A.4.7.
A.5.5.4 Lightning impulse test
The test is as generally outlined in 10.3.
The test shall be made with waves of positive polarity applied to each terminal with the other terminal
grounded.
NOTE—The lightning impulse test is generally not required on thyristor-controlled shunt reactors that operate at
distribution voltage class levels. The turn-to-turn test is the dielectric test that is normally performed to verify the
winding insulation system. If a lightning impulse test is required, it shall be specified at time of order by the purchaser.
For TCRs that operate at voltages higher than 34.5 kV, a lightning impulse test is performed.
A.5.5.5 Turn-to-turn test
This test is as generally outlined in 10.3.3.3. Test levels are given in Column 11 of Table 5.
79
A.5.6 Notes on design tests for thyristor-controlled reactors
A.5.6.1 Temperature-rise test
The temperature-rise test is carried out at power frequency. The temperature-rise test is usually carried out
at an equivalent 60 Hz current corresponding to the operating point (fundamental and harmonic) that
produces the highest losses. In other words, the temperature-rise test should be carried out at rated losses if
they are the maximum losses; minus adjustments as described below. If the reactor is provided with an
enclosure, the enclosure shall be erected for the temperature-rise test. The test procedure is outlined in 10.5.
The equivalent test current is obtained from Equation (A.4):
m
I 2 t ⋅ Rt = I F2 ⋅RF + ∑
n=2
2
I Hn ⋅RHn (A.4)
where
It is the equivalent test current (50 or 60 Hz)

IF is the fundamental current, corresponding to the operating point that produces the highest losses
IHn is the nth harmonic current, corresponding to the operating point that produces the highest losses
Rt is the winding resistance at test current frequency corrected to maximum operating temperature
RF is the winding resistance at fundamental frequency corrected to maximum operating temperature
RHn is the winding resistance at nth harmonic, corrected to maximum operating temperature
m is the highest specified harmonic
Losses that do not directly influence the temperature rise of windings (e.g., losses in support structure
elements such as insulators, extension brackets, and pedestals) shall not be included. Therefore, it is not
directly based on measured loss values since winding loss (I2R and stray) and other stray losses cannot be
separated by measurement; theoretical values for stray losses in structure elements should be deducted from
the loss measurement. Losses in winding clamping structure should be included if they are determined to
have an effect/influence on the winding-temperature rise and/or the connector temperature rise. In addition,
when the reactor is provided with an enclosure, the possible effect on temperature rise due to any induced
losses in the enclosure should be considered. Stray losses external to the winding shall be defined and
explained by the manufacturer.
NOTE—In order to avoid discussion between the purchaser (end user) and the manufacturer regarding the magnitude
of stray losses in the support structure elements, it is recommended that the loss measurement be performed on the coil
only, minus the “in service” support structure. On this basis, the loss measurement results can be used to determine the
temperature-rise test current.
If terminal temperature-rise measurement is required, the customer is highly recommended to supply the
appropriate length of incoming bus/cable and associated connectors as would be used for the final
installation. The terminal temperature-rise measurement taken during the temperature-rise test at the
equivalent test current is then corrected for the maximum ambient temperature specified and for the actual
winding termination electrode/clamping structure loss at the required operating point. The maximum
temperature rise of the connectors shall be in accordance with 8.2.4 of IEEE Std C57.16-1996. If the
appropriate connectors and incoming bus/cable cannot be supplied, then the terminal temperature rise
cannot be assessed.
The test shall provide verification of temperature limits achieved at rated current as described in A.4.2.
Acceptable average winding and hot-spot temperature shall be as per A.4.6.
80
A.5.6.2 Measurement of variation of inductance and resistance with frequency
The variation of inductance and resistance with frequency shall, when required, be measured at any
convenient ambient temperature. The test may be carried out on a per-coil or per-phase basis. The
frequency range shall be 60 Hz to 3600 Hz for the inductance and dc to 3600 Hz for the resistance, unless
otherwise specified. The measured values, excluding stray losses in structural elements such as the coil
mounting system, shall be corrected to maximum operating temperature and used for calculation of the
equivalent current in A.5.5.1.
A.5.7 Notes on “other” tests for thyristor-controlled reactors
A.5.7.1 Audible-sound test
A.5.7.1.1 General
The audible sound of a reactor may be verified by calculation or by a separate test. It should be noted that
the full current spectrum cannot be supplied simultaneously. In addition, the magnitude of the currents that
can be supplied to the reactor should preferably be those derived from the specified fundamental and
harmonic currents. However, harmonic current sources that are in use have limited power capability, and in
many cases, the specified magnitude of harmonic current(s) cannot be supplied to the reactor; the required
power being in excess of the power supply capabilities. In such cases, it is general practice to carry out the
sound-level measurement for harmonic currents at reduced current levels.
Generally, an audible-sound test shall be performed as prescribed in 10.6; test setup, instrumentation,
microphone positions, etc. Two acceptable audible-sound test methodologies can be used; discrete
microphone position or “walk around” using an integrating sound-level meter, the latter being the preferred
method. Therefore, IEEE Std C57.16-1996 and IEEE Std 1277-2000 should be consulted in conjunction
with the specific information below. The sound radiated from the TCR depends on both the fundamental
and harmonic currents. It should be noted that, in service, TCRs are normally subjected to simultaneous
harmonic currents of different frequencies, amplitude, and phase relationship, and thus sound levels
measured in the laboratory may be different from those measured at site. The factory test can be used for
the verification of the sound-level design calculations. Unless otherwise specified, it is recommended to
consider only the fundamental and most significant harmonic currents out of the harmonic current
spectrum.
Since simultaneous fundamental and harmonic current usually cannot be applied for testing, it is
recommended to test the reactor successively with power-frequency and the harmonic currents.
Additionally, the TCR shall also be tested at currents that reflect the interaction of currents having different
frequencies. The harmonic power supply usually consists of a power amplifier or variable frequency
inverter, and associated resonant circuit elements; rotating machines may also be used. The objective is to
have a relatively “pure” sinusoidal harmonic test current. As stated above, the sound-level test may be
performed with the TCR energized at a reduced current level at various harmonics providing prior
agreement is reached between the manufacturer and user. This approach is usually acceptable if the
measured sound level is well above ambient. Sound level at the rated harmonic current can be calculated
from the measured sound-level value.
For a reactor current spectrum, the sound equivalent currents are given as follows:
81
Amplitude of Frequency of
TCR current Sound frequency
test current test current
I1, I2, I3, ... I1 f1 2 f1
I2 f2 2 f2
I3 f3 2 f3
… … …
For any pair of currents, the following currents need to be considered due to interactive effects:
Amplitude of Frequency of
TCR current Sound frequency
test current test current
I1, I2 (2 * I1 * I2) 1/2 (f2 + f1) /2 f2 + f1
(2 * I1 * I2) 1/2 (f2 − f1) /2 f2 − f1
where f1, f2, f3, ... are the frequencies of the filter reactor rms currents I1, I2, I3, ... that are interacting.
Usually f1 is the fundamental and f2, f3, ... are the frequencies of the significant harmonic currents.
A.5.7.1.2 Terminology
The following terminology is important for proper application of sound measurement methodology.
a) Ambient sound pressure level is the sound pressure level measured in the test facility without the
reactor energized.
b) Sound pressure level, in decibels, is 20 times the logarithm to the base ten of the ratio of the
measured sound pressure to a reference pressure of 20 μPa.
c) Sound power level, in decibels, is 10 times the logarithm to the base ten of the ratio of the emitted
sound power to a reference power of 10−12 watt.
d) Semi-reverberant facility is a room with a solid floor and an undetermined amount of sound
absorbing materials on the walls and ceiling.
A.5.7.1.3 Instrumentation
Sound pressure level measurements shall be made with instrumentation that meets the requirements of
ANSI S1.4-1983 for Type-2 meters.
Sound measuring instrumentation shall be calibrated before and after each measurement session. Further, it
should be demonstrated prior to the measurement that the magnetic field of the reactor does not affect the
readings of the sound-level meter. Should the calibration change by more than 2 dB, or should the readings
change by more than 1 dB due to the magnetic field, the measurements shall be declared invalid.
A.5.7.1.4 Test conditions
The thyristor-controlled shunt reactor shall be completely assembled.
The installed mounting configuration of the reactor should be simulated as much as possible. In order to
obtain rated current(s), it may be necessary to carry out the test on a per-coil basis. Sound level for a two-
coil stack can be calculated from the single-coil test results.
82
Should the reactor be tested within a semi-reverberant facility, it should be located in an asymmetrical
manner with respect to the room geometry. When reactor sound emissions are measured in an enclosed
space, sound reflections from walls or other large objects can influence the results because the sound is
essentially a discrete tone that may be affected by room acoustics, room geometry, or reflecting objects.
A.5.7.1.5 Microphone positions
The reference-sound-producing surface of a dry-type air-core TCR is its outside winding surface. For TCRs
with a winding less than 2.4 m (8 ft) tall, microphone locations shall be at mid-height of the winding. For
TCRs greater than 2.4 m (8 ft) tall, microphone positions shall be at one-third and two-thirds winding
height.
Continuous measurement by a “walk around” test with an integrating sound-level meter is the preferred test
method. The measurement contour as described in the following paragraph is to be employed when using
the capabilities of an integrating sound-level meter. It is important that the contour be traversed at a near-
constant speed (uniform walking speed). Experience has shown that results obtained using the “walk
around” audible-sound measurement methodology are in agreement with those obtained using the discrete
microphone position methodology.
As an alternative test method, the discrete microphone position method can be used. The choice is to left to
the manufacturer. In the plan view, the microphone locations shall be laid out clockwise, sequentially at
intervals of 1 m along the circumference of a circle having its center at the geometric center of the TCR,
and a radius equal to the TCR radius plus 3 m (10 ft). The first station will be on a radial line through the
bottom terminal, or as close to it in the clockwise direction as is permitted to comply with minimum
clearance distances to live parts.
For the most often employed two-coil stack arrangement for TCRs, the microphone positions described
apply to each individual reactor.
A.5.7.1.6 Sound-level measurement
Sound pressure levels shall be measured in conformance with 10.6 and A.5.7.1.1, A.5.7.1.3, A.5.7.1.4, and
A.5.7.1.5 using the sound-level-meter A-weighting characteristic.
A.5.7.1.7 Calculation of average sound pressure level
An average sound pressure level at each tested current and frequency value, LpA, shall be calculated from
the measured values of the A-weighted sound pressure level, LpAi, by using Equation (A.5):
⎛ N ⎞
⎜1 0.1L pAi ⎟
L pA= 10 log10 ⎜
⎜N
10 ∑ ⎟⎟ (A.5)
⎝ i =1 ⎠
where
LpA is the average sound pressure level (dB)

LpAi is the measured sound pressure level at locations, i (dB)
N is the total number of measurement locations
The total sound pressure level can be computed using Equation (A.6):
83
⎡ n ⎤
∑
Ltot = 10 log ⎢ 10 0.1 Li
⎢ i =1
⎥
⎥
(A.6)
⎣ ⎦
where
Ltot is the total sound pressure level

Li is the sound pressure level (dB) of each individual component
n is the number of individual components
NOTE—The sound pressure level of components not included in the test shall be estimated by calculation.
A.5.7.1.8 Calculation of A-weighted sound power level
The sound power level (LwA) shall be computed
LwA = Ltot + 10 log10 ( S ) (A.7)
where
LwA is the sound power level (dB)

Ltot is the total A-weighted sound pressure level (dB)
S is the measurement surface area in square meters
The measurement surface area (S) is the area (in square meters) of a hypothetical envelope of the TCR on
which the sound measurements are located. The determination of S is as follows.
For a dry-type TCR located near ground, the best estimation of the surface is a hemispherical shell. Thus, a
dry-type TCR, S, is calculated as follows in Equation (A.8):
S = 2πRm2 (A.8)
where
Rm is the measurement radius
A.6 Tolerances
A.6.1 Inductance
Tolerances on rated inductance shall be specified and guaranteed.
A.6.2 Losses
TCRs consist of single coils per phase or two coils per phase; more typically two stacked coils per phase. In
the case of two coil stacks, the losses of the top and bottom coils may differ. Therefore, the per phase losses
of a TCR consist of the losses of a single coil or of two coils. For the purpose of establishing a tolerance on
losses, the losses of any coil of the same design should not differ from the average of all units of the same
design by more than 6%. For the case of a single coil per phase or two identical coils per phase, the average
84
loss shall be calculated by using the measured losses of each individual unit. In the case of a two-coil
stacked TCR where the top and bottom coils are of different design an average loss value shall be
calculated for each of the top and bottom coil by using the measured losses for each individual unit of the
top and bottom coils, respectively. In this case, the losses of any top coil or bottom coil shall not differ by
more than 6% from the average loss value for the top and bottom coils, respectively.
A.7 Nameplate information
A.7.1 Recommended information
The following information should be included on the nameplate:
a) Apparatus type
b) Manufacturer’s name
c) Serial number
d) Year of manufacture
e) Temperature class
f) Temperature rise in degrees Celsius
g) Type of cooling
h) Rated power frequency
i) Rated inductance
j) Measured inductance
k) Rated fundamental current (as applicable)
l) Rated current
m) RSS current or fundamental plus major harmonic currents (corresponding to the operating point
that produces the highest losses)
n) BIL
o) Maximum operating voltage
p) Number and year of standard
q) Total weight
r) Noise level (if required)
A.7.2 Optional Information
Optional information that may be required as follows:
a) Number of coils per phase

b) Drawing number
c) Customer ID number
d) Service: indoor, outdoor
e) Maximum ambient temperature
85
f) Altitude
g) Harmonic current spectrum
A.8 Thyristor-controlled reactor design information
The information in A.8.1 and A.8.2 is required to properly design a TCR.
A.8.1 Thyristor-controlled reactors for standard applications
Maximum system voltage kV

Rated system voltage kV
System frequency Hz
Rated inductance mH
— Tolerance %
Fundamental current rating A
Harmonic currents (at corresponding fundamental current) A (Hz/amperes)
Maximum continuous voltage to ground kV
Lightning impulse withstand level (LIWL)
— Across reactor kV crest
— Low-voltage terminal to ground kV crest
Mounting: side-by-side or two-coil stacked
Indoor or outdoor
Loss evaluation at specified current loading $ / kW
Maximum stray capacitance to ground and across the reactor pF
A.8.2 Environmental conditions
Seismic (g)
Pollution levels: salt, industrial, etc.
Ambient temperature:
— Annual average °C
— Annual maximum °C
— Annual minimum °C
Elevation above see level m
Humidity conditions
Ice loading conditions
Maximum wind velocity km/h
Isokeraunic levels
Audible noise levels at the reactor (dBA at ___ m)
and if required at the boundary of the SVC installation dBA
Solar radiation level (if extreme)
86
A.9 Supporting published material
[1] IEC/TR 60815:1986, Guide for the Selection of Insulators in Respect of Polluted Conditions.22
[2] IEEE Std 1031-1991, IEEE Guide for a Detailed Functional Specification and Application of Static
VAR Compensators.23, 24
[3] Static VAR Compensators, CIGRE Report of Working Group 38-01, Task Force No. 2 on SVC, 1986.
22
IEC publications are available from the Sales Department of the International Electrotechnical Commission, Case Postale 131, 3,
rue de Varembé, CH-1211, Genève 20, Switzerland/Suisse (http://www.iec.ch/). IEC publications are also available in the United
States from the Sales Department, American National Standards Institute, 11 West 42nd Street, 13th Floor, New York, NY 10036,
USA.
23
This publication is available from the Institute of Electrical and Electronics Engineers, 445 Hoes Lane, Piscataway, NJ 08854, USA
(http://standards.ieee.org/).
24
The IEEE standards or products referred to in A.9 are trademarks of the Institute of Electrical and Electronics Engineers, Inc.
87
Annex B
(informative)
Dielectric stresses imposed on shunt reactors during switching
B.1 Background
The focus of this annex will be on stresses that tertiary shunt reactors and directly connected shunt reactors
can be exposed to and must be able to withstand during daily switching operations. Tertiary-connected
shunt reactors are usually limited to voltages below 72.5 kV and are applied on the tertiary of a
transformer, which is connected to the transmission line being compensated. High-voltage and extra-high-
voltage shunt reactors are directly connected to a station bus or to the transmission line being compensated.
IEEE Std C37.015-1993 considers both circuit breakers and circuit switchers and focuses on methodology
and formulas to calculate transient stress levels but does not explicitly deal with stresses that shunt reactors
should be able to withstand. Shunt reactor switching is a daily event (sometimes multiple times per day)
and exposes the shunt reactor to repetitive stresses—impacting tertiary shunt reactors and direct connect
shunt reactors up to 800 kV. The issue is how the withstand level resulting from the repetitive dielectric
stresses imposed on the shunt reactor during current chopping relates to the SIWL (switching impulse
withstand level). Similarly, the voltage stresses generated by circuit-breaker reignition phenomena are
similar to those that may be simulated by a chopped-wave test.
Circuit breakers are not ideal. They are subject to current chopping and reignition; levels are a function of
the breaker type. For instance, air blast breakers are capable of chopping inductive load currents of 50 A or
more, depending on the circuit parallel capacitance.
NOTE—Other switching devices such as load-break switches (circuit switchers) are used to switch shunt reactors and
the information in this annex can be applied as applicable.
B.2 Circuit-breaker testing
The purpose of circuit-breaker tests is as follows:
a) To determine the interrupting window without reignition

b) To determine the current chopping characteristics of the circuit breaker
Test results include the following:
1) Reignition performance versus arcing time

2) Chopping number (overvoltage) versus arcing time
3) Reignition voltage versus arcing time
Circuit-breaker tests for shunt reactor switching are covered in IEC and IEEE guides. See supporting
document IEC 62271-110:2005 [1]25 and IEEE Std C37.015-1993. These two documents are technically
identical since they are the result of a cooperative effort between a working group of the IEEE High-
Voltage Circuit Breaker Subcommittee and a Task Force of the IEC Technical Committee SC17A.
25
In this annex, the numbers in brackets correspond to those of the supporting published material in B.7.
88
B.3 Description of dielectric stresses imposed on shunt reactors during switching
Circuit breakers are used to switch tertiary-connected shunt reactors and direct-connected shunt reactors. In
the ideal case, there is no current chopping and no reignition during the switching operation, and the
maximum voltage seen by the shunt reactor will be the phase-to-ground peak voltage. However, ideal
circuit breakers do not exist and current chopping and reignitions will occur. In fact, both conditions can
exist in the same circuit-breaker application exposing the shunt reactor to dielectric stresses during
switching operations.
B.3.1 Current chopping
Current chopping is the premature forcing to zero of the interrupted current due to arc instability.
Severity of current chopping is defined by the magnitude of the current chopped. The tendency of some
circuit-breaker types to chop current is described by a chopping number, which is a characteristic of the
circuit breaker. The magnitude of the chopped current is a function of chopping number and the total
parallel capacitance seen by the circuit breaker used to switch the shunt reactor. This parallel capacitance
includes the source and load-side capacitances and any grading capacitance across the circuit breaker.
Typical chopping number ranges are as follows:
Air blast circuit breakers: 15 – 25 × 104 A/F0.5
SF6 circuit breakers: 4 – 17 × 104 A/F0.5
Minimum oil circuit breakers: 7 – 10 × 104 A/F0.5
Vacuum circuit breakers: Not applicable (see below)
In simple terms, a higher chopping number implies a greater tendency to chop current.
Vacuum circuit breakers are commonly used to switch shunt reactors connected at medium-voltage class
levels through a transformer tertiary winding. However, the level of current chopping is related to contact
material and therefore the chopping number approach is not applicable.
Where applicable, the magnitude of the chopped current (ich) can be calculated from the chopping number
and the parallel capacitance seen from the circuit-breaker terminal as follows in Equation (B.1):
ich = λ CT (B.1)
where
λ is the chopping number

ich is the magnitude of chopped current
CT is the effective capacitance across the terminals of the circuit breaker
Current chopping results in an overvoltage across the shunt reactor and across the circuit breaker. For a
specific shunt reactor, the overvoltage is proportional to the current chopped. The amplitude of the current
chopped is often proportional to the arcing time of the circuit breaker. A longer arcing time implies that the
“interrupting strength” of the circuit breaker may be greater, and thus the circuit breaker may chop more
current. Since during the current chopping event energy is transferred between the reactors and the parallel
capacitance, the overvoltage, using the equivalent circuit in Figure B.1, is given by Equation (B.2) and
Equation (B.3):
89
1 1
L × (ich )2 = × CT × (ΔV )2 (B.2)
2 2
⎛ L ⎞
ΔV = i ch × ⎜⎜ ⎟⎟ (B.3)
⎝ CT ⎠
where
CT = CP + CL (assuming Cs >> CL)

CP is the stray capacitance across the circuit breaker
CS is the effective capacitance of the system seen on the source side of the circuit breaker
CL is the effective capacitance in parallel with the shunt reactor (shunt reactor stray effective
capacitances, ancillary equipment stray capacitances, lumped capacitance of any mitigation)
By inspection of Equation (B.2) and Equation (B.3), it can be seen that a reactor with a small stray
capacitance will be affected more by current chopping. Dry-type reactors tend to have a lower stray
capacitance to ground versus oil-immersed reactors. Oil-filled reactors also employ a bushing that has
substantial capacitance, and winding capacitance to ground is high due to the close physical proximity of
tank and core. Winding interturn capacitance tends to make less of a contribution to total stray capacitance.
For a shunt reactor with directly grounded neutral, the frequency of the overvoltage is given by
Equation (B.4):
1
f transient = (B.4)
2π L C L
Typical range of frequencies is 1 kHz to 40 kHz for oil-immersed reactors (stray capacitance 0.8 nF to
4 nF) and > 50 kHz for dry-type reactors (stray capacitances ∼ 100 pF to 500 pF).
NOTE—If the shunt reactor is ungrounded, then the L-term in Equation (B.4) becomes 1.5 L.
The transient overvoltage can thus be severe depending on the circuit-breaker chopping characteristics (as
applicable), the shunt reactor rating, and stray capacitance.
The methodology to arrive at the maximum value of λ for a particular circuit breaker can be found in
Annex E of IEEE Std C37.015-1993. Note that shunt reactor switching tests are not performed on medium-
voltage circuit breakers, and this particular circuit-breaker application is treated by reference to low level
fault duties.
B.3.2 Reignition
Reignition occurs in a circuit breaker when the transient recovery voltage (TRV) applied to the circuit
breaker after current extinction is higher than the “dynamic” voltage withstand capability of the circuit
breaker during an opening operation. Reignition occurs normally when arcing times are small and the
contacts of the breaker have not yet reached the full clearance required to withstand the voltage stress. For
high-voltage SF6 circuit breakers, typically arcing times lower than 4 ms will lead to reignitions.
Typical frequencies of the transient overvoltages associated with reignition of circuit breakers are from
50 kHz to 1.0 MHz. The frequency is mostly dependent on the stray capacitance of the shunt reactor
(Cs >>> CL) and the inductance of the reignition circuit (Lb) between the source-side capacitance and the
shunt reactor. See Figure B.1.
90
Ls is the supply-side (short-circuit) inductance

Cs is the supply-side capacitance
LP, CP are the stray inductance and capacitance across circuit breaker CB
Lb is the inductance of reignition circuit
CL is the effective capacitance parallel to the reactor
L is the inductance of reactor
Figure B.1—Single-phase equivalent circuit for a solidly grounded shunt reactor
B.3.3 Insulation coordination recommendations for shunt reactors
Insulation coordination recommendations are as follows:
a) For medium-voltage and high-voltage applications (< 362 kV), the SIPL (switching impulse
protection level) of the shunt reactor surge arrester is relatively low and the margin between the
SIPL and the SIWL is greater than 32%. For these cases, the parallel arrester will normally provide
sufficient protection against current chopping even for shunt reactors with aged insulation systems.
In the case of extra-high-voltage (≥ 362 kV) shunt reactors, due to the low margin between the
SIPL and SIWL (typically 25%) and since reactor switching is a daily event, it is recommended
that the calculated overvoltage peak to be less than 0.68 × SIWL [0.8 (aging) × 0.85 (margin)
= 0.68]. Especially for extra-high-voltage shunt reactors, the arrester in many cases cannot provide
consistent protection; primarily if a specific dielectric strength reduction of the reactor insulation
system due to aging is considered. This is most likely to be the case for air blast and dual pressure
SF6 circuit breakers if the chopping overvoltage peak is high enough to cause arrester operation.
Similar results have been reported for some single pressure 800 kV SF6 circuit breakers having a
severe chopping behavior.
Good background information on the application of surge arresters can be found in IEEE Std
C62.22-1997.
b) The phase-to-ground overvoltage produced by a reignition is not the worst dielectric stress
generated during this type of event. The stresses produced by this event are caused by the high rate
of change voltage excursion applied to the shunt reactor during a reignition. This stress can be
extremely severe and the surge arrester connected across the shunt reactor will not protect it.
NOTE—The descriptor “high rate of change voltage excursion” is used to describe the overvoltage seen by shunt
reactors during switching due to reignition. This description is more appropriate as the dv/dt is not as fast as that
experienced during the chopped-wave test and the voltage excursion is with respect to an absolute voltage value. The
high rate of change voltage excursion seen in service during reignitions are always slower because the inductances
involved are always greater than during chopped-wave tests if the chopping gap is located as required in IEEE Std 4-
1995.
91
In other words, at reignition, the voltage from the high-voltage terminal to ground may, for
instance, instantaneously change from negative to positive voltage with the result that the reactor is
stressed by the full-voltage excursion from negative to positive, i.e., by the full high rate of change
voltage excursion. Depending on the rate of change of voltage excursion, this will lead to increased
transient voltage stress between turns for windings exhibiting nonlinear voltage distribution (mainly
in the turns closest to the shunt reactor high-voltage terminal). (The high rate of change voltage
excursion will not have as severe an impact for windings with virtually linear distribution.)
For medium-voltage shunt reactors (≤ 45 kV), the high rate of change voltage excursion produced
during single or multiple reignition events (more particularly applicable to vacuum circuit breakers)
is 2 × LIPL (lightning impulse protection level). The shunt reactor is generally well protected by
the surge arrester [less than 68% of the CWWL (chopped-wave withstand level)] because the
margin between the protective level of the surge arrester and the chopped-wave rated level of the
shunt reactor is large. As an example, a 25 kV shunt reactor has a CWWL of 165 kV and the LIPL
of the surge arrester is 51 kV. Therefore, 2 kV × 51 kV is < 165 kV, resulting in a very good
protection margin.
For reactors rated 69 kV to 161 kV (preferred BIL), the high rate of change voltage excursion
produced by a reignition will not typically be in excess of the required margin for reactors with
aged insulation—more than 68% of the CWWL. For these cases, reignition can be accepted as an
accidental event. It should be noted that for some cases (reduced BIL ratings of each nominal
system voltage), the high rate of change voltage excursion produced by a reignition will exceed the
dv/dt produced by a chopped wave. For these cases, reignitions of circuit breaker should be
avoided.
For reactors rated above 161 kV, the high rate of change voltage excursion produced by a reignition
is always higher than the chopped-wave rating of the reactor. For these cases, the high rate of
change voltage excursion produced by a reignition will exceed the high rate of change voltage
excursion produced by a chopped wave, and reignitions of the circuit breaker should be avoided.
B.4 Internal resonance
All shunt reactors have multiple internal self-resonance frequencies. A reactor winding is a complex
network and has not just one resonance frequency but several eigenvalues. The excitation of one or several
internal resonances depends on the boundary conditions. The lowest frequency fo is that with the high-
voltage terminal open and the other reactor terminal grounded. This is essentially the oscillation frequency
of the TRV stressing the breaker. If there is a fixed potential at both reactor terminals, e.g., after switching
a reactor on, then you may have internal oscillations of the reactor winding with both terminals being a
node. The main frequency f1 of this oscillation will be higher than fo. In the case of dry-type reactors, it is a
function of the windings only. For oil-immersed reactors, it is a function of the windings and associated
bushings—particularly in the case of the TRV. If the shunt reactor is subjected to a transient overvoltage
that excites the critical frequency, then the internal voltage seen by the winding insulation system can be
many times higher than the applied voltage across the terminals. Therefore, the application of a lightning
arrester across the shunt reactors will not completely protect the reactor. It is recommended that
calculations be carried out for shunt reactor installations to ensure that the frequency of switching
overvoltages (current chopping and reignition) do not coincide with critical self-resonance frequencies of
the shunt reactors. Background information can be found in a number of documents listed in the
bibliography (see Annex D).
B.5 Mitigation
Surge arresters shall be used to protect shunt reactors. Lightning arresters can be applied directly across the
shunt reactor (terminal to terminal) and line to ground. Consideration should be given to installing a
lightning arrester in both configurations depending on the grounding practice for the shunt reactor
installation.
92
The circuit breaker employed with a shunt reactor installation should be qualified for switching reactors
and its characteristics should be known so that an evaluation of the overvoltage stresses seen by the shunt
reactor during switching can be carried out. Additional transient recovery control networks such as resistive
capacitive (RC) devices have been employed in conjunction with medium-voltage (MV) tertiary-connected
shunt reactors. It is highly recommended that detailed overvoltage calculations be performed for each shunt
reactor application and appropriate measures taken to ensure that dielectric stresses seen by the shunt
reactor(s) and associated equipment are kept within acceptable limits.
The type of circuit breaker and circuit-breaker modifications are also major considerations. Opening
resistors were used in the past on air blast circuit breakers. Some SF6 circuit breakers employ an arrester in
parallel with the circuit breaker. Another newer option is to employ a synchronous switching strategy with
the circuit breaker. (For economic reasons, this is currently the preferred solution for high-voltage and
extra-high-voltage shunt reactors.)
B.6 Maximum switching transient overvoltage
In conjunction with considerations due to the frequent switching operations seen by shunt reactors, the
following maximum overvoltage levels should not be exceeded unless agreed upon by the user and
manufacturer. They are substantially lower than traditional lightning impulse (BIL) and switching impulse
(SIL) values. BIL and SIL are based on random, infrequent events, and the switching overvoltage
associated with shunt reactor switching is highly repetitive. Voltage aging of the reactor insulation system
has also been considered.
a) The maximum overvoltage produced by current chopping shall not exceed 68% of the rated SIWL
rating of the reactor or 56% of the rated BIL rating (68% × 83% × LIWL) if SIWL is not defined:
This is valid for both oil-immersed and dry-type shunt reactors and is consistent with the data in
Table 5.
b) The maximum high rate of change voltage excursion produced during reignition (if reignitions are
allowed as a possible daily event) to be less than 68% of the rated CWWL of the reactor.
c) Shunt reactors rated 230 kV and above are typically not designed to withstand the stresses
produced by a reignition of the circuit breaker unless special measures have been taken in the
reactor design or in the circuit-breaker design.
d) In no case should the high rate of change voltage excursion produced during a reignition exceed the
rated CWWL of the shunt reactor.
B.7 Supporting published material
[1] IEC 62271-110:2005, High-Voltage Alternating Current Circuit Breakers—Inductive Load

Switching.26
[2] IEEE PC57.142 (Draft 5.1, May 6, 2008), Draft Guide to Describe the Occurrence and Mitigation of
Switching Transients Induced by Transformer, Switching Device, and System Interaction.27, 28, 29
26
USA.
27
This IEEE standards project was not approved by the IEEE-SA Standards Board at the time this publication went to press. For
information about obtaining a draft, contact the IEEE.
28
29
The IEEE standard or product referred to in B.7 is a trademark of the Institute of Electrical and Electronics Engineers, Inc.
93
Annex C
(informative)
Dielectric tests including information on wave shape control
The maximum half-value time T2 of an impulse-wave tail can be derived from the resonance frequency of
the impulse generator capacitance (Cg) with the test object reactance (Lt), as expressed in Equation (C.1):
π
T2 = --- Lt Cg
3 (C.1)
This is a theoretical value applying to an undamped oscillation with an opposite polarity peak of 100%.
Various amounts of circuit damping will reduce this value accordingly. With a limitation of 50% for the
opposite polarity peak, for instance, as expressed in Equation (C.2):
T 2 ≈ 0.5L t C g
(C.2)
Values of T2 close to Equation (C.1) can be achieved with the use of an inductor in parallel with the series
(front) resistor of the impulse circuit, with compromises generally required between wave duration,
opposite polarity peak, wave front time, and peak overshoot.
The manufacturer may also elect to test a low-impedance winding by inserting a resistor of not more than
500 Ω in the grounded end of the winding. Although this will improve the impulse wave shape, the largest
portion of the test voltage will be across the resistor, and not across the test-coil windings. Therefore, a
shorter impulse-wave tail is preferable to the insertion of a series resistor between the test object and
ground.
In the case of the oil-immersed equipment, at the manufacturer’s preference, low-impedance windings may
be tested by tying together the terminals of windings with the same insulation level.
More information on the testing of low impedance windings can be found in IEC 60076-4:2002 [B16].
94
Annex D
(informative)
Bibliography
[B1] Accredited Standards Committee C2-2007, National Electrical Safety Code® (NESC®).30, 31
[B2] ANSI C63.2-1996, American National Standard Specifications for Electromagnetic Noise and Field
Strength; Instrumentation 10 kHz to 40 GHz.32
[B3] ANSI C84.1-2006, American National Standard for Electric Power Systems and Equipment—
Voltage Ratings (60 Hz).
[B4] ASTM D117-2002, Standard Guide for Sampling, Test Methods and Specifications, and Guide for
Electrical Insulating Oils of Petroleum Origin.33
[B5] Bruckel, W., “Measurement connection for testing high-voltage reactors” (in German), Messen und
Pruefen/Automatik, Nov. 1976.
[B6] Craig, S., and Kayser, H., “Iron-loss measurements by ac bridge and calorimeter,” Journal of the
Institution of Electrical Engineers, vol. 95, pt. 2, pp. 205–216, 1948.
[B7] Debourg, H., Jenkins, R. S., Slettenmark, I., Tengstrand, C. A., and Wester, C. E., “Calorimetric loss
measurement on alternators and reactors,” Paper 119, International Conference on Large High Tension
Electric Systems (CIGRE), 20th session, 1964.
[B8] Deutsch, F., “Measuring the active power losses of large reactors,” Brown Boveri Review, vol. 47,
pp. 268–278, Apr. 1960.
[B9] Erb, W., and Kraaij, D. J., “Design and testing of reactors for 735 kV,” Brown Boveri Review,
vol. 52, pp. 864–875, Nov./Dec. 1965.
[B10] Foley, A. H. A., “Self-balancing transformer core loss bridge,” AIEE Transactions, vol. 76, pt. 1,
pp. 567–573, 1957.
[B11] Grundmark, B., “High-voltage shunt reactors—Trends in design and testing,” Paper 12-03,
International Conference on Large High Tension Electric Systems (CIGRE), 23rd session, 1970.
[B12] Hague, B., Alternating Current Bridge Methods. London: Pitman and Sons, 1957.
[B13] Heroines, R. T., and Graham, D. C., “Measurement of self-cooled transformer sound levels in
relatively high ambients,” IEEE Transactions on Power Apparatus and Systems, vol. PAS-89, pp. 1657–
1662, Sept./Oct. 1970.
30
The NESC is available from the Institute of Electrical and Electronics Engineers, 445 Hoes Lane, Piscataway, NJ 08854, USA
(http://standards.ieee.org/).
31
National Electrical Safety Code and NESC are both registered trademarks and service marks of the Institute of Electrical and
Electronics Engineers, Inc.
32
ANSI publications are available from the Sales Department, American National Standards Institute, 25 West 43rd Street, 4th Floor,
New York, NY 10036, USA (http://www.ansi.org/).
33
ASTM publications are available from the American Society for Testing and Materials, 100 Barr Harbor Drive, West
Conshohocken, PA 19428-2959, USA (http://www.astm.org/).
95
[B14] IEC 60076-2:1997, Power transformers—Part 2: Temperature rise.34
[B15] IEC 60076-3:2000, Power transformers—Part 3: Insulation levels, dielectric tests and external
clearances in air.
[B16] IEC 60076-4:2002, Power Transformers—Part 4: Guide to lightning impulse and switching impulse
testing—Power transformers and reactors.
[B17] IEC 60076-4:2002, Power transformers—Part 4: Guide to the lightning impulse and switching-
impulse testing—Power transformers and reactors, Appendix A: Principles for Wave Shape Control.
[B18] IEC-60076-6:2007, Power Transformers—Part 6: Reactors.
[B19] IEEE 100™, The Authoritative Dictionary of IEEE Standard Terms, Seventh Edition. New York:
Institute of Electrical and Electronics Engineers, Inc.35, 36
[B20] IEEE Std C62.1™-1989 (Reaff 1994), IEEE Standard for Gapped Silicon-Carbide Surge Arresters
for AC Power Circuits.37
[B21] Kusters, N. L., and Petersons, O. A., “Transformer-ratio-arm bridge for high voltage capacitance
measurements,” IEEE Transactions on Communications and Electronics, vol. 69, pp. 606–611, Nov. 1963.
[B22] Moore, W. J. M., Love, G., and Raftis, F. A., “Measurement of short-circuit load losses in large
three-phase power transformers, using an alternating current comparator bridge,” IEEE Transactions on
Power Apparatus and Systems, vol. PAS-94, no. 6, pp. 2074–2076, Nov./Dec. 1975.
[B23] Moore, W. J. M., and Raftis, F. A., “Measurement of shunt reactor loss at high-voltage with an
alternating current comparator bridge,” IEEE Transactions on Power Apparatus and Systems, vol. PAS-92,
no. 2, pp. 662–667, Mar./Apr. 1973.
[B24] Petersons, O. A., “Transformer-ratioarm bridge for measuring large capacitors above 100 volts,” IEEE
Transactions on Power Apparatus and Systems, vol. PAS-87, pp. 1354–1361, May 1968.
[B25] Uniform Building code (UBC) 1997.38
34
USA.
35
36
The IEEE standards or products referred to in Annex D are trademarks owned by the Institute of Electrical and Electronics
Engineers, Incorporated.
37
IEEE Std C62.1-1989 has been withdrawn; however, copies can be obtained from Global Engineering, 15 Inverness Way East,
Englewood, CO 80112-5704, USA, tel. (303) 792-2181 (http://global.ihs.com/).
38
The Uniform Building Code is available from the International Code Council, Country Club Hills, IL (http://iccsafe.org).
96

IEEE STD C57.21

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

IEEE STD C57.21

Uploaded by

Copyright:

Available Formats

IEEE Standard Requirements,

Terminology, and Test Code for

IEEE Power Engineering Society

IEEE Standard Requirements,

Approved 27 March 2008

The Institute of Electrical and Electronics Engineers, Inc.

Copyright © 2008 by the Institute of Electrical and Electronics Engineers, Inc.

PDF: ISBN 978-0-7381-5412-1 STD95779

Secretary, IEEE-SA Standards Board

Laws and regulations

Richard F. Dudley, Chair

Sten Andersson Derek Foster Les Recksiedler

William J. Ackerman Kenneth Hanus Jeffrey Merryman

Robert M. Grow, Chair

Victor Berman Jim Hughes Ron Petersen

Satish K. Aggarwal, NRC Representative

1.1 Scope ................................................................................................................................................... 1

3.1 Shunt reactor ........................................................................................................................................ 3

4.1 Service conditions................................................................................................................................ 6

5. Basis for rating shunt reactors .................................................................................................................... 9

5.1 kVA ratings ......................................................................................................................................... 9

6.1 General .............................................................................................................................................. 11

7. Losses and impedance .............................................................................................................................. 11

7.1 Total losses ........................................................................................................................................ 11

8. Temperature rise ....................................................................................................................................... 14

8.1 Life of insulating materials ................................................................................................................ 14

9. Dielectric tests and insulation levels......................................................................................................... 15

9.1 Dielectric tests ................................................................................................................................... 15

10.1 General ............................................................................................................................................ 21

11. Construction for oil-immersed shunt reactors ........................................................................................ 58

11.1 Bushings .......................................................................................................................................... 58

12. Construction and installation of dry-type shunt reactors ........................................................................ 67

12.1 General description .......................................................................................................................... 67

Annex D (informative) Bibliography ........................................................................................................... 95

a) Shunt reactors, 500 kVA and smaller

ANSI/ASME B1.20.1-1983 (Reaff 2001), Pipe Threads, General Purpose (Inch).

IEEE Std 4™-1995, IEEE Standard Techniques for High-Voltage Testing.

3.1 Shunt reactor

3.2 Rating of a shunt reactor

3.4 Losses and impedance

3.5.2 outdoor shunt reactor: A reactor of weather-resistant design and construction.

3.6 Methods of cooling

⎯ O mineral oil or synthetic insulating liquid with fire point ≤ 300 °C

⎯ N natural convective flow through cooling equipment and in windings

Third letter: External cooling medium:

Fourth letter: Circulation mechanism for external cooling medium:

NOTE—IEEE Std C57.12.00-2006 can be consulted as a reference.

4.1 Service conditions

4.1.1 Usual temperature and altitude service conditions

4.1.2 Usual voltage conditions

4.1.3 Usual installation conditions

4.1.4 Unusual temperature and altitude service conditions

4.1.5 Other unusual service conditions

c) Excessively high or low ambient temperatures

4.1.6 Environmental aspects

4.2 Effect of air density on flashover voltage

Table 1 —Dielectric strength correction factors for altitudes

≤1000 ≤3 300 1.00

4.4 Effect of altitude on temperature rise

4.4.2 Operation at rated kVA

Table 2 —Maximum allowable 24 h average temperature of

4.5 Electrical properties of insulating oil

4.6 Magnetic characteristics