You are on page 1of 58

IEEE Guide for Accelerated Aging

Tests for Medium-Voltage (5 kV35 kV)


Extruded Electric Power Cables Using
Water-Filled Tanks

IEEE Power Engineering Society


Sponsored by the
Insulated Conductors Committee
TM

1407

m the p
fro

REDLINE
re
es
vi
ng

ou
xact cha

E
s from

IN
s version!

hange

D L
c t c

E
e x a n!

R
Shows evious versio
the p r
s e
w

o
IEEE Sh IEEE Std 1407-2007
3 Park Avenue
New York, NY 10016-5997, USA (Revision of
IEEE Std 1407-1998)
15 February 2008

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
This is a Redline Document produced by Techstreet, a business of
Thomson Reuters.

This document is intended to provide users with an indication of changes from one edition
to the next. It includes a full-text version of the new document, plus an indication of
changes from the previous version.

Redlines are designed to save time and improve efficiencies by using the latest software
technology to find and highlight document changes. More professionals are using valuable
new technologies like redlines, to help improve outcomes in a fastpaced global business
world.

Because it may not be technically possible to capture all changes accurately, it is


recommended that users consult previous editions as appropriate. In all cases, only the
current base version of this publication is to be considered the official document.

445 Hoes Lane, Piscataway, NJ 08854 USA | http://standards.ieee.org | Tel. +1 732-981-0060 Fax +1 732-562-1571

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.

Redline Processing Notes:

1. Red Text - Red strikethrough text denotes deletions.

2. Blue Text - Blue underlined text denotes modifications and additions.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
IEEE Std 1407-2007
(Revision of
IEEE Std 1407-1998)

IEEE Guide for Accelerated Aging


Tests for Medium-Voltage (5 kV35 kV)
Extruded Electric Power Cables Using
Water-Filled Tanks

Sponsor
Insulated Conductors Committee
of the
IEEE Power Engineering Society

Approved 6 December 2007


IEEE SA-Standards Board

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Abstract: Accelerated aging tests on extruded medium-voltage cables using water-filled tanks are
addressed. Information on the equipment, cable samples, test conditions, and measurements to
perform the aging tests is provided. Techniques on how to analyze the test data are also included.
The implementation of this guide will allow a better description of the test data obtained by different
laboratories.
Keywords: accelerated aging test, accelerated cable life test (ACLT), cross-linked polyethylene
(XLPE), ethylene propylene rubber (EPR), medium-voltage cable, tank structure, temperature
measurement and profile, test conditions, water-filled tanks

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 15 February 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.

The trademark SuperSMITH belongs to Fulton Findings LLC.

Weibull++ is a registered trademark of ReliaSoft Corporation.

PDF: ISBN 978-0-7381-5352-0 STD95749


Print: ISBN 978-0-7381-5353-7 STDPD95749
No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior
written permission of the publisher.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
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.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Introduction
This introduction is not part of IEEE Std 1407-2007, IEEE Guide for Accelerated Aging Tests for Medium-Voltage
(5 kV35 kV) Extruded Electric Power Cables Using Water-Filled Tanks.

This guide addresses accelerated aging tests of extruded medium-voltage cables using water-filled tanks.
Information is provided on the equipment, cable samples, test conditions, and measurements to perform the
accelerated aging tests. The guide identifies the critical test parameters and describes techniques for their
measurement and control. The implementation of the techniques will allow test data obtained by different
laboratories at different times to be better compared. Specific test values (e.g., voltage) are given in this
guide, and the range in test values (e.g., maximum temperature) are more limited than stated in the previous
version of this guide.

Notice to users

Laws and regulations

Users of these documents should consult all applicable laws and regulations. Compliance with the
provisions of this guide does not imply compliance to any applicable regulatory requirements. Implementers
of the guide 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.

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.

iv Copyright 2008 IEEE. All rights reserved.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
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 guide may require use of subject matter
covered by patent rights. By publication of this guide, 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 guide 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

The work of preparing this guide was carried out by Task Force A13W of Subcommittee A (Cable
Construction and Design) of the Insulated Conductors Committee of the IEEE Power Engineering Society.

At the time this guide was submitted to the IEEE-SA Standards Board for approval, the Task Force had the
following members:
John R. Densley, Chair
Mark D. Walton, Vice Chair
Bruce S. Bernstein Carlos Katz Serge Plissou
Jan-Ove Bostrom Frank Krajick Timothy J. Person
Ray O. Bristol Frank Kutchta Haridoss Sarma
Laurence H. Gross Carl C. Landinger John T. Smith III
Richard A. Hartlein Eric P. Marsden Steve Szaniszlo
John L. Hinkle William M. McDermid William S. Temple
Stanley R. Howell Harry E. Orton Dennis Wedam

The following members of the individual balloting committee voted on this guide. Balloters may have voted
for approval, disapproval, or abstention.

Stan Arnot Randall C. Groves Gary L. Michel


Michael P. Baldwin Richard Harp Jerry R. Murphy
William G. Bloethe Jeffrey L. Hartenberger Michael S. Newman
A. James Braun Wolfgang B. Haverkamp Serge Plissou
Kent W. Brown Gary A. Heuston Iulian E. Profir
Nissan M. Burstein Lauri J. Hiivala Bartien Sayogo
William A. Byrd David A. Horvath Michael J. Smalley
Tommy P. Cooper Gael Kennedy Jerry W. Smith
John R. Densley Jim Kulchisky John A. Vergis
Donald G. Dunn Chung-Yiu Lam Martin J. Von Herrmann
Mark D. Walton
Gary Engmann Benjamin T. Lanz
William D. Wilkens
Rabiz N. Foda Eric P. Marsden
Timmy S. Wright
Marcel Fortin William M. McDermid Ahmed F. Zobaa
Steven N. Graham John E. Merando

Copyright 2008 IEEE. All rights reserved. v

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
When the IEEE-SA Standards Board approved this guide on 6 December 2007, it had the following
membership:
Steve M. Mills, Chair
Robert M. Grow, Vice Chair
Don Wright, Past Chair
Judith Gorman, Secretary

Richard DeBlasio Richard H. Hulett Narayanan Ramachandran


Alex Gelman Hermann Koch Greg Ratta
William R. Goldbach Joseph L. Koepfinger* Robby Robson
Arnold M. Greenspan John Kulick Anne-Marie Sahazizia
Joanna N. Guenin David J. Law Virginia C. Sulzberger
Julian Forster* Glenn Parsons Malcolm V. Thaden
Kenneth S. Hanus Ronald C. Petersen Richard L. Townsend
William B. Hopf Tom A. Prevost Howard L. Wolfman

*Member Emeritus

Also included are the following nonvoting IEEE-SA Standards Board liaisons:

Satish K. Aggarwal, NRC Representative


Alan H. Cookson, NIST Representative

Don Messina
IEEE Standards Program Manager, Document Development

Matthew J. Ceglia
IEEE Standards Program Manager, Technical Program Development

vi Copyright 2008 IEEE. All rights reserved.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Contents
1. Overview.............................................................................................................................................. 1

1.1 Scope............................................................................................................................................ 2

2. Normative references ........................................................................................................................... 3

3. Definitions............................................................................................................................................ 3

3.1 Definitions.................................................................................................................................... 3
3.2 Abbreviations and acronyms........................................................................................................ 3

4. Cable materials and design considerations .......................................................................................... 4

4.1 Cable core test.............................................................................................................................. 5


4.2 Cable design test .......................................................................................................................... 6

5. Test specimens ..................................................................................................................................... 6

6. Prior tests and thermal preconditioning of specimens ......................................................................... 7

7. Test structure...................................................................................................................................... 10

8. Temperature ....................................................................................................................................... 13

8.1 Test specimen conductor temperature ....................................................................................... 13


8.2 Water temperature...................................................................................................................... 16
8.3 Ambient temperature in the test area ......................................................................................... 17
8.4 Establishing a temperature profile ............................................................................................. 17
8.5 Monitoring/controlling temperatures during the ACLT ............................................................ 18
8.6 Temperature measurement techniques....................................................................................... 18
8.7 Recommendations...................................................................................................................... 18

9. Water.................................................................................................................................................. 18

9.1 Water nature............................................................................................................................... 18


9.2 Conductor strand blockage ........................................................................................................ 20

10. Voltage............................................................................................................................................... 20

11. Time ................................................................................................................................................... 23

12. Test matrix ......................................................................................................................................... 23

13. Failure ................................................................................................................................................ 24

14. Abnormalities..................................................................................................................................... 25

15. Diagnostics......................................................................................................................................... 27

16. Data analysis ...................................................................................................................................... 28

Copyright 2008 IEEE. All rights reserved. vii

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
17. Final report......................................................................................................................................... 29

18. Conclusions........................................................................................................................................ 30

Annex A (informative) Preconditioning ........................................................................................................ 31

Annex B (informative) Water tanksHistorical information ....................................................................... 34

Annex C (informative) Temperature measurement ....................................................................................... 38

Annex D (informative) Final report............................................................................................................... 40

Annex E (informative) Bibliography ............................................................................................................. 42

viii Copyright 2008 IEEE. All rights reserved.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
IEEE Guide for Accelerated Aging Tests
for Medium-Voltage (5 kV35 kV)
Extruded Electric Power Cables Using
Water-Filled Tanks

IMPORTANT NOTICE: This guide is not intended to assure safety, security, health, or environmental
protection in all circumstances. Implementers of the guide 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

The insulations commonly used in extruded medium-voltage power cables are cross-linked polyethylene
(XLPE), tree- retardant XLPE (TRXLPE), and ethylene propylene rubber (EPR). These materials are known to
gradually deteriorate with time due to the synergistic action of moisture and electrical stress. This degradation,
often referred to as water treeing, causes a reduction in the breakdown voltage and an increase in the dielectric
loss. The degradation can be so severe that failures can occur in service at operating stresses. The mechanisms
of moisture-induced degradation have been studied extensively and it is believed known that several mechanisms
are occurring. A thorough knowledge of the degradation mechanisms allows optimum choices of test parameters
to be made; this results in an accelerated aging test that evaluates the long-term performance of a cable in the
absence of scattered defects. Although there is not a complete understanding of the mechanisms of moisture-
induced degradation, the urgency to develop an aging test has resulted in the proposal of several accelerated
aging tests for materials and cables (Banks, et al., [B4],1 Bartnikas, Densley, and Eichhorn [B5], Bartnikas, et al.,
[B6], Mashikian, et al., [B21], Schroth, et al., [B24], Steenish and Faremo [B26]).Several accelerated aging tests
have been developed to evaluate the long term performance of cables (Banks, Faremo, Steennis [B4],1 Bartnikas,
Densley, Eichhorn [B5], Bartnikas et al. [B6], Mashikian et al., [B20], Schroth, Kalkner, Fredrich [B25],
Steennis and Faremo [B27]). The tests may be either fixed time, where the cables are aged for a fixed time and
then subjected to one or more diagnostic tests to assess the cable condition, or a time-to-failure test. A problem
inherent in all accelerated aging tests is the choice of test parameters in the test itself. These should contain the
deterioration factors causing service aging and not introduce aging mechanisms that do not occur in service. The
diagnostic test must measure a property representative of service aging. Research has shown that the most
important parameters in tests to evaluate moisture-induced degradation are electrical stress, temperature, water
characteristics, and time.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Two accelerated aging tests have evolved in North America. One accelerated aging test that has evolved in North
America into, a Qualification test, is the accelerated water treeing test (AWTT) in which lengths of cables are
aged in water-Filled pipes for 4, 6, and 12 months and then subjected to a step ac breakdown (ACBD) test, the
diagnostic test (AEIC CS5ANSI/ICEA S-94, AEIC S6-96). An alternative -649 and ANSI/ICEA S-97-6822).
The other is a test has emerged where cables are aged in water-filled tanks, either for a Fixed time or to failure in
water-Filled tanks, i.e., an accelerated cable life test (ACLT) (Lyle [ B18], Walton, Smith, Thue [B32]). The
time-to-failure test has the advantage of involving the whole aging process, initiation, growth, and Final failure.
The main disadvantage is that the test duration is unknown and there can be considerable time between the First
and the last failure. Several laboratories have performed and are now performing tank-type testsaccelerated aging
tests in tanks (both time-to-failure tests in which the samples are aged to failure and aging tests in which the
samples are aged for fixed time periods followed by a diagnostic test such as a step ACBD test) using different
aging conditions of voltage, temperature, water characteristics, etc. This makes a direct comparison of the data
between laboratories very Difficult. The situation is further complicated because the methods of measuring some
of the critical parameters differ Significantly. As a result, apparently similar test conditions can, in fact, be
entirely different in two separate test arrangements. An example of this ambiguity is the temperature
measurement. Although a nominal temperature of 90 C is sometimes used as the maximum temperature during
temperature cycling, the precise location where this temperature is measured may vary. It may be measured in
the length of the cable outside the water or in the length of cable immersed in the water. If the water temperature
is not controlled, however, the temperature of the immersed cable will depend upon the following:

Tank size
Tank shape and material
Volume of water
Number of cables in the tank
Whether the tank is thermally insulated, etc.

This guide directly addresses tank-type accelerated aging tests. Accelerated aging test parameters are presented
which that, if reported, will allow test data obtained by different laboratories at different times to be better
compared.

This guide will:

Identify the critical test parameters and techniques for their measurement and control.
Review the levels of the test parameters, such as voltage (V), temperature (T), etc. It is not the intention of
this guide to recommend Specific test values.
Identify the test parameters to be included in reports, etc.
The following will be discussed:
Test specimens
Prior tests and preconditioning of specimens
Test structure
Temperature
Water
Voltage
Failures
Abnormalities
Diagnostics
Data analysis
Final report

It must be pointed out that some parameters are interrelated so that a change in one will have an effect on others
(i.e., as already mentioned, the size of the tank could influence the temperature of the cable immersed in a tank
without water temperature control).

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
1.1 Scope

This guide provides information on the equipment, cable specimens, test conditions, and measurements to
perform accelerated aging tests on medium-voltage cables using water-Filled tanks, whether the test be atime-to-
failure test or a test in which samples are aged for fixed times followed by a diagnostic test such as a step ACBD
test. The guide Identifies the critical test parameters and describes techniques for their measurement and
control. The implementation of the techniques will facilitate a comparison of data obtained by different
laboratories.

2. Normative references

This guide shall be used in conjunction with the following publications. When the following publications are
superseded by an approved revision, the revision shall apply

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 the 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.

AEIC CS5-94, CS8-07, Specification for Cross-Linked Polyethylene Insulated Extruded Dielectric Shielded
Power Cables Rated 5 through 46 kV. 3

AEIC CS6-96, Specifications for Ethylene Propylene Rubber Insulated Shielded Power Cables Rated
ANSI/ICEA S-94-649, Standard for Concentric Neutral Cables Rated 5 through 46 kV.4, 5

ANSI/ICEA S-97-682, Standard for Utility Shielded Power Cables Rated 5 through 69 46 kV. ICEA T-24-380,

Guide for Partial-Discharge Test Procedure.6

3. Definitions abbreviations, and acronyms

For the purposes of this document, the following terms and definitions apply. The Authoritative Dictionary of
IEEE Standards Terms [B13] should be referenced for terms not defined in this clause.

3.1 Definitions

3.1.1 abnormality: Any deviation from the preestablished test conditions, including the tolerance limits, that
may affect the outcome of the test.

3.1.2 air failure: A failure in the cable above the waterline but below the termination.

3.1.3 termination failure: A failure in the portion of the cable that does not have a metallic shield covering.

3.1.4 water failure: A failure in the active, shielded cable length that is below the waterline and that did not
occur as a result of mechanical damage.

3.1.5 waterline failure: A failure at the interface between air and the tank water to include the distance of the
total water line variation.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
3.2 Abbreviations and acronyms

ACLT accelerated cable life test

AWTT accelerated water treeing test

DSC differential scanning calorimetry

EPR ethylene propylene rubber

GMTF geometric mean time to failure

I conductor current

Mle maximum likelihood estimators

PD partial discharge

PL polymer or polymer-lined

R conductor resistance

SS stainless steel

tan loss tangent (dissipation factor) of insulation

T/C thermocouple

THD total harmonic distortion

TRXLPE tree retardant cross-linked polyethylene

V0 line to ground voltage

VLF very low frequency

2f where f is the frequency in Hz of applied voltage

XLPE cross-linked polyethylene

ACBD ac breakdown

3 AEIC publications are available from the Association of Edison Illuminating Companies, 600 N. 18th Street, P. O. Box 2641, Birmingham,
AL 35291-0992, USA (http://www.aeic.org/). AEIC publications are also available from Global Engineering Documents, 15 Inverness Way
East, Englewood, Colorado 80112-5704, USA (http://global.ihs.com/).
4 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/).

5 ICEA publications are available from ICEA, P.O. Box 20048, Minneapolis, MN 55420, USA (http://www.icea.org/).
6 See Footnote 5.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
4. Cable materials and design considerations
An accelerated aging test in a water-filled tank, whether a time-to-failure test or a fixed aging time test followed
by diagnostic testing such as a step ACBD test, imposes a complex combination of electrical, thermal, and
mechanical stresses on a cable during the aging test program. Since these stresses are applied to a manufactured
cable that, at a minimum, must be a cable core composed of an unfilled conductor (conductor without strand-
filling material to block the ingress of moisture or water), conductor shield, insulation, insulation shield (and
metallic shield), the test inherently evaluates these individual materials as well as the overall cable structure. The
synergistic effects between the various materials are important. Insulation/shield interface characteristics,
variable thermal expansion characteristics, volatile diffusion rates, ionic contamination, manufacturing process,
and many other factors provide for a complex, dynamic system that is more than a simple sum of the individual
parts.

An accelerated aging test in a water-filled tank, whether a time-to-failure test or a fixed aging time test followed
by diagnostic testing, can also be used to evaluate more than just the cable core. Since utilities now commonly
employ moisture blocked conductors, overall jackets, and a variety of different radial and longitudinal moisture
barriers, it is sometimes useful to test complete cable designs. Under these conditions the entire cable design
structure is evaluated, including the cable core materials, all moisture barriers jackets, etc. Once again, the many
synergistic interactions between all the cable components ultimately dictate the performance of the overall cable
design.

In this respect, an accelerated aging test in a water-filled tank can be considered both a cable core performance
test and a cable design performance test. For purposes of this discussion, when a cable core test is performed, the
test cable consists only of the minimum components required to make up a shielded cable design. When a cable
design test is performed, the test cable contains all of the components that the cable would have if placed in
service on an electric utility system.

4.1 Cable core test

This is by far the most common use of the accelerated aging test in a water-filled tank to date. When conducting
the cable core performance test, four different aspects of a cable design are addressed.

The first component evaluated is the insulation. It is the component that has to withstand the applied voltage, and
for this test, it is not protected from moisture ingress, a key aspect of the aging process for many extruded
dielectric insulation materials.

The second component considered to have an important effect on cable performance is the conductor shield. The
conductor shield primarily controls the smoothness of the conductor shield/insulation interface, a high stress
region of the design that is critical to achieving acceptable cable performance. Conductor shield cleanliness is
also very important. Some ionic contaminants, which come primarily from the carbon black used to make the
conductor shield conductive, can migrate into the insulation, enhancing sites for water tree initiation and growth.

The third aspect is the insulation shield. Like the conductor shield, the insulation shield controls the smoothness
of the insulation/insulation shield interface. It also contains carbon black. However, because the voltage stress at
the insulation/insulation shield interface is lower than that at the conductor shield/insulation interface, the impact
of the insulation shield on cable performance is considered to be less significant than the conductor shield.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
The fourth factor in a cable core evaluation test is the manufacturing process used to make the cable. This
includes a multitude of factors including the curing process, extruder and die design and their condition, line
speed, and the cooling process. Among other things, these processes affect void size and density, insulation
morphological characteristics, and residual mechanical stresses in the insulation.

In summary, a cable core material test is generally considered to be a test of the materials used to make up the
core. However, it is also a test of the process used to manufacture the cable and an evaluation of the synergistic
interplay between all of these items. Items evaluated in a cable core accelerated aging test in a water-filled tank
are listed as follows:
a) Insulation
b) Conductor shield
c) Insulation shield
d) Manufacturing processes
e) Synergistic effects between all preceding items

The relative ranking of the listed items is arguable, but the cable core performance test, by its very nature,
evaluates all of these components.

4.2 Cable design test

While the accelerated aging test in a water-filled tank is primarily used to evaluate cable cores, it is occasionally
used to evaluate complete cable designs. Since complete cable designs often employ some type of moisture
blocking component(s), they generally age more slowly than cables in a cable core test. The accelerated aging
test in a water-filled tank is ideally suited for evaluating these cables because the test generally runs for extended
periods of time, often several years.

A complete cable design normally incorporates a cable core plus a moisture-blocked conductor and a jacket. It
may also employ the use of additional moisture barriers including water swellable tapes, hydrophilic or
hygroscopic powders, and/or metal laminates. How all of these design characteristics combine to influence
performance is extraordinarily complex. The cable core continues to be the most important portion of the cable
structure. However, once moisture ingress to the core structure is reduced or eliminated by one or more of the
components that make up the cable design, then the dynamics of the aging process can change significantly. For
example, if moisture is no longer present at the conductor, the role of the conductor shield material may not be as
pronounced as it would be if moisture were present. The same is true for the insulation material. For cable
designs that provide for a complete hermetic seal, insulation characteristics such as water tree resistance may no
longer be important.

Because it could take several years before specimens in a cable design test fail, it may be appropriate to remove
specimens from the aging test periodically and conduct diagnostic tests such as ACBD and water tree counts.
This is particularly useful for test programs conducted for the purpose of comparing different cable designs.
Once a comparative performance trend is established, the test can be terminated without having to age the test
specimens to failure.

Whether conducting a cable core test or a cable design test, it is important to know that the accelerated aging test
in a water-filled tank does much more than evaluate the performance of one aspect of a cable construction. All of
the cable componentsfrom the conductor shield to the jacketare inexorably linked. This means that
performance data should not be associated with one component, but rather with the entire cable construction.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
This document does not make specific test protocol recommendations for a cable design test. However, test
specimens in a cable design test are likely to incorporate components such as jackets and blocked conductors that
impede the ingress of moisture into the cable structure. As a result, specimens in a cable design test will likely
age more slowly than designs in a cable core test. For this reason, it may be desirable to use more severe test
conditions in a cable design test than for a cable core test.

5. Test specimens
Table 1 lists a format for documenting the cable under test and provides example information. The values listed
in Table 1 are compatible with tanks shown in Figure 1, Figure 2, and Figure 3. The cable under test should
conform to AEIC CS8-07.

Table 1Example of cable documentation

Voltage class 15 kV,a 4.4 mm (175 mil) insulation thickness b


Conductor 53.5 mm2 (1/0 AWG) 1350-H19 stranded compressed aluminum a
Conductor shield 0.38 mm (0.015 in) state grade and manufacturer, blend number, and
date of manufacture

Insulation 4.4 mm (0.175 in) state grade and manufacturer, blend number, and date
of manufacture
Insulation shield 0.76 mm (0.030 in) state grade and manufacturer, blend number, and
date
of manufacture
Metallic shield 6 #14 bare copper wires a
Jacket None or identify material, thickness a
Cable extrusion True triple

Cable extrusion date To be included

Cure process Dry cure


Extrusion run identification Job number, reel number

Cable manufacturer To be included


Cable quality control tests To latest relevant AEIC spec* ICEA standard a
Number of specimens in population 10 a
Specimen total length 5 m (16 ft) c
Specimen length under water 3.7 m (12 ft)

Number of cable coils 1

Inner diameter cable coil 610 mm (24 in) a


Date water added to tank To be included

Date initial water added to conductor To be included

Start voltage/heat cycles date To be included

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Replacement specimen Same type and construction of cable as test specimen if cable heating is
necessary to maintain consistent water temperature a
a
Values or tests in common use according to a survey conducted by the task force.
b Insulation thickness at point of failure is recommended for reporting in Clause 13, Clause 14, and Clause 15.
c Specimen total length needs to be increased to a minimum of 9.5 m (32 ft) if the residual breakdown strength will be
measured after cable aging.

6. Prior tests and thermal preconditioning of specimens

Prior tests and preconditioning of cable specimens are carried out to assess whether the cable specimens have
defects that cause partial discharges (PD) and to limit the levels of some cross-linking by-products, which can
occur in some cross-linking processes and which are known to affect the electrical properties of XLPE.

Partial discharge tests should be performed on every test cable to ensure that the cable and terminations are
discharge-free prior to the aging test. The partial discharge test should be carried out according to AEIC CS5-94
and AEIC CS6-96.

PD tests should be performed on the entire reel of cable prior to cutting into test specimens. If there is any doubt
as to whether or not the reel was subjected to a PD test or there is concern about the integrity of the terminations,
it is suggested that PD testing be performed on individual test specimens. The PD test should be carried out
according to ICEA T24-380.

It is well known that the most commonly used chemical cross-linking agent, dicumyl peroxide, produces a
number of cross-linking agent by-products that include acetophenone, cumyl alcohol, -methyl styrene and
smaller quantities of other by-products. Their formation cannot be avoided, but they are migratory and can
theoretically volatilize from the cable. Methods for by-product analysis and data for new and aged cables have
been reported (Damon, et al., [B7], EPRI Report EL-7076 [B12]) Residual cross-linking agent by-products may
influence the electrical properties of polymeric insulating materials such as XLPE, TRXLPE, and EPR. The
residuals will depend on the specific chemical reactions, which occur during manufacture, and the rate at which
the by-products subsequently diffuse through the extruded layers. Although the same cross-linking by-products
occur in XLPE, TRXLPE and EPR, the rate of diffusion will differ for XLPE, TRXLPE, and EPR and may vary
for different TRXLPE and EPR formulations. There have been a number of studies of the effects of the by
products on the short- and long-term electrical behavior of XLPE insulation, and these are summarized in Annex
A.

Silane-cured XLPE does not use dicumyl peroxide as the curing agent and thus contains different cross linking
by-products. No data are available at this time on the effects of the by-products for silane-cured insulation and
the procedures in this clause may not be applicable to silane-cured XLPE. Also no data are available in the
published literature on the effect of cross-linking by-products on EPR-insulated cables.

Although their effects on the performance of extruded cable insulation are not precisely known, there is
sufficient published data, at least for steam-cured XLPE, to show that an increased concentration of byproducts
improves the resistance of the insulation system to water treeing and that the concentration of byproducts in
XLPE gradually decreases with time, e.g., the concentration of cross-linking by-products in 15 kV XLPE cable
may be less than 100 ppm after 10 years in service and, as a result, may not contribute to the tree redundancy.
However, in accelerated aging tests lasting a year or even longer, if the accelerated temperature does not drive
off the cross-linking by-products, an apparent improved resistance of the insulation to water treeing could result.
Thus the long-term behavior, for the insulation with a low concentration of by-products, may not be as good as
that predicted by the results of the accelerated aging test. To reduce this possibility the cables are preconditioned
in order to reduce the amount of by-products in the insulation before being subjected to the aging test. The most
common form of preconditioning is to subject the cables to elevated temperatures, either by conductor heating or

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
oven heating, to increase the rate of diffusion of the by-products from the insulation. (AEIC CS5-94, AEIC CS6-
96).

In addition to removing the by-products, elevated-temperature preconditioning can:

Cause chemical changes in the insulation depending on the temperature employed.


Cause morphological changes in the insulation due to partial melting and recrystallization.
(EPRI Report EL-5921 [B11])

These changes can also affect the water treeing and long-term behavior and contribute to making preconditioning
and its influence a very complex issue. Although the effects on XLPE are only partially understood, and the
effects on TRXLPE and EPR are understood to a lesser extent, some form of preconditioning is preferred to
reduce the level of by-products as long as the other changes mentioned here do not alter the mechanism of aging
and loss of life. The preconditioning described here is based on limited measurements on 15kV XLPE cables
with no strand blocking (Abdolall [B1], Martin and Hartlein [B19], Plissou [B23]) and its aim is to reduce the
level of the cross-linking by-products to less than 100 ppm. The choice of 100 ppm is based upon a limited
number of measurements on cables that had been in service for up to 10 years and may have to be modified as
more information becomes available. The procedure may not be suitable for silane-cured XLPE, TRXLPE, or
EPR cables, as no measurements have been made to determine the level of cross-linking by-products in service-
aged cables with these insulations, nor tests performed to determine the rates at which the by-products diffuse
from these types of cables at elevated temperatures. In view of the unknowns, users should be cautious in
interpreting the data.

Table 2 lists recommended preconditioning protocols based on test experience at various laboratories. The
consensus opinion of the task force A13W for a preconditioning protocol is 360 h at 90 C in free air or in a
circulating-air oven (for cable core testing of a common design as well as cable design testing).

Table 2 Preconditioning protocols

List of preconditioning protocols used in different laboratories

1 360 h at 90 C conductor in free air a (recommended)


2 360 h at 90 C in circulating-air oven b (recommended)

The results are shown in Table 2.

Results of tests on 15 kV XLPE cables aged in tanks

Geometric mean time to failure (GMTF) in days


(90, +90% Confidence Limits)
Type of preconditioning Age of cable
treatment after extrusion
(months) Aged at 4 V0 and Aged at 3 V0 and
90 C* (4/90) 75 C*(3/75)
None (i.e., fresh cable) 2 64 (56, 74) 386 (366, 406)

90 C on conductor 60 45 (40, 50) 264 (214, 325)


cycled
8 h on 16 h off for 120 h
60 C on conductor con- 60 46 (37, 57) 321 (309, 333)
tinuous for 120 h

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Insulation shield to 110
54 103 (85, 124) 325 (284, 371)
C
then immediate cool
*Maximum temperature during load cycling

a When cable specimens are preconditioned by circulating current in the conductor, the temperature from specimen to specimen
and/or along the cable length will likely vary as much as 10 C. This variation is a function of several factors including the
orientation of the specimens, the distance between the specimens and/or coils and adjacent structures, and the location of the
specimen and/or coil relative to the group of specimens being preconditioned.
b Heating cables in a circulating-air oven should reduce the variations in temperature from specimen to specimen and along the
cable length that can occur when specimens are heated by circulating current in open air. It must be noted that heating in a
circulating-air oven subjects the complete cable core to the oven temperature, unlike conductor heating when the insulation
shield will be about 10 C lower than the conductor temperature. The higher temperature of the insulation shield could
increase the diffusion of volatile products from the shield into the insulation.

The following procedures have been used to analyze the cross-linking by-products and characterize the
insulation:

a) Heat the conductor by circulating current to 90 C 2 C for a time to be determined such as 500 h in an
open atmosphere. For steam-cured 15 kV cables, after 500 h at 90 C the amounts acetophe- none and -methyl
styrene were reduced to less than 10 ppm and cumyl alcohol to less than 100 ppm (Abdolall [B1]).

b) Heat the conductor by circulating current to 90 C 2 C for 72 h in an open atmosphere. However,


significantly larger quantities of by-products may remain in the XLPE insulation after only 72 h.

Analysis of the by-products in the insulation and conductor shield before and after preconditioning and also after
the aging: The residual volatiles are generally liberated from a specimen and subsequently analyzed either using
1) Solid head-space techniques, e.g., gas chromatography, or
2) A solution (e.g., liquid chromatography), or
3) Thermal desorption techniques or
4) Weight loss during preconditioning.

Of items 1), 2), 3) and 4), the most widely used is the solution technique item 2), wherein the by-products from a
specimen taken at 1/3 of the insulation thickness from the conductor are extracted using methylene chloride and
analyzed using liquid chromatography. A more Efficient procedure is to use the analytical techniques described
in item 5).
5) Analytical techniques, which rely on the direct introduction of volatile components into the chosen analytical
equipment as they are extracted from the solid specimen thereby keeping specimen handling and errors to a
minimum (EPRI Report EL-7076 [B12]). Groeger et al. [B10]). The same procedures have been used for
TRXLPE and EPR but less information is available. However, the suggested analytical protocols for by-product
detection are not expected to introduce any concerns or uncertainties that have not already been expressed for
XLPE insulation.
b) Characterization of the morphology or structure of the insulation before and after preconditioning and also
after the aging. The morphology may be measured on specimens from the inner 25% of the insulation wall
thickness using differential scanning calorimetry (DSC). Prolonged exposure of XLPE to elevated temperatures

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
below its melting point may introduce an additional peak(s) on the DSC thermogram.This additional peak(s) is
related to the indicative of prior temperature(s) to which the insulation specimen may have been subjected earlier
and which signifies that and is indicative of polymer recrystallization (Groeger et al. [B10]). This additional
peak(s) in the DSC thermogram can be eliminated by heating the insu-lation to a higher temperature or tolation
to temperatures above its melting point (~106 C), for example to 120 C or 130 C, and then allowing it to cool
to ambient temperature. More sophisticated methods of measuring the morphology are available (EPRI Report
EL-5921 [B11]). Phillips and Vatansever [B23]).
c) Measurement of the percent gel or extractable at the inner 25% of the insulation wall before and after pre-
conditioning as Defined in AEIC CS5 ANSI/ICEA S-94-649 and AEIC CS6-96 ANSI/ICEA S-97-682.

The purposes of the above preceding analyses are to address one of the main concerns about the use of thermal
preconditioning to remove the cross-linking by-products, i.e., the effects of the changes in insulation morphology
that can occur during the preconditioning or cable aging that may influence the mechanism of aging. It should be
noted that changes in morphology and the migration of cross-linking by-products will also occur during the
aging in the tank and that these will depend upon the aging temperature. Thus, until more information becomes
available, it is suggested might be useful that, for information purposes only, measurements be made of the
cross-linking by-products and also the morphology of the insulation before and after the preconditioning, and
also after the aging.

Any interpretation of cable test data will be incomplete without the actual details of the location of the sampling
for the cross-linking by-products with reference to the test cable and of the analytical methods used to measure
the by-products. It is therefore suggested to report these two parameters while giving quan- titative information
on the concentration of by-products so as to track down the effect of these on cable aging precisely.

Monitoring of the amounts of by-products in cables before the aging tests may lead to a Refinement of the
preconditioning procedure later. For example, levels of one or more by-products could be Specified or the
conditions for preconditioning changed in future editions of this guide.

7. Test structure

It should be recognized that tank shape, size, material of construction, water volume and characteristics, and
number of cables in the tank will have a direct influence on important parameters, the cable and water
temperatures, and ultimately on the results of the test. As of the 2007 revision of the guide, recommended tank
material and tank sizes are given. The details of other tank sizes and constructions, which were included in the
previous edition of the guide, are given in Annex B

Figure 1, Figure 2, and Figure 3 show the recommended tank layout schematics. The tank shape is rectangular
(Figure 1 and Figure 2) or cylindrical (Figure 3.). The rectangular tank sizes allow 10 or 20 specimens plus any
control specimens such as a dummy loop used for temperature monitoring. The cylindrical tank configuration
allows up to 10 specimens each 65 m (~210 ft) long to be aged. It also can accommodate successive failures and
re-terminating of the non-failed remnant lengths. There should be sufficient distance between cable specimens
and tank sides, bottom, and other cable specimens to prevent uneven heating of any individual specimens and to
allow adequate circulation of the water.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
It is recommended that stainless steel (SS) tanks be used with the dimensions listed in Table 3.

When data from various tests are compared, the number of cable specimens plus the following tank size
variables should be considered: length, width, height (in the case of generally rectangular shapes or the
appropriate dimensions for other shapes), and depth of water. If a nonrecommended tank construction is used,
the ratio of water volume to submerged cable volume should also be considered. These parameters will have
significant effects on the temperature of the water if the water temperature is not controlled. However, if the
water temperature is controlled, less attention to tank dimensions and structure can be tolerated.

Table 3Recommended dimensions of SS tanks

Rectangular tanks

10 specimens 20 specimens

Length 1590 60 mm (62.5 in 2.5 in) 3175 125 mm (125 in 5 in)


Width 1220 150 mm (48 in 6 in) 1220 150 mm (48 in 6 in)

Height 1220 150 mm (48 in 6 in) 1220 150 mm (48 in 6 in)

Cylindrical

10 specimens each 65 m (213 ft) long

Diameter 1220 150 mm (48 in 6 in)


Height 1220 150 mm (48 in 6 in)

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
The tank is made from an inert material (SS). In addition, the specimen mounts, and any other materials making
contact with the water in the tanks should be constructed from suitable inert materials that will not corrode in the
presence of water and will not form an electrolytic cell with any component in the cable specimen. The tanks are
usually grounded. In particular, the tank should not cause the introduction of a significant level of ion
contamination, which can lead to erroneous results.

The type of material contacting the water should be reported. Also, any materials used to insulate the tanks or
used to insulate or retard surface evaporation of the water and their thicknesses should be reported to allow
determination of their impact on water temperature and quality.

8. Temperature
Temperature has been shown in previous studies to play a very significant role as an aging parameter for some
cables in tank-type testing (Lyle [B18]). In order to elevate the conductor temperature to the levels required for
accelerated aging of the cable, current is usually induced into the conductor using a current transformer. Cable
heating occurs due to resistive losses in the conductor. This method of heating is the common practice and
preferred method since cables are heated in this way in actual field use.
Conductor heating may be cyclic or continuous. The usual practice is cyclic heating because this type of heating
more closely simulates the type of heating that a cable would experience in actual service. Current heating cycles
of 8 h on, 16 h off (7 days a week) is the most common practice. It should also be noted that tests without
conductor heating are also performed.

8.1 Test specimen conductor temperature

Since part of each test specimen length is in water and part is in air, the thermodynamics of heating and cooling
the test specimens can be quite complex. For this reason, it is important that the conductor temperature be
monitored in both the water and air portions of the test specimen, preferably at the hottest spot in both places.
Locating these hot spots, should they exist, may take some experimentation. During ACLT, temperature
monitoring is typically accomplished by placing thermocouples (T/Cs) on the conductor of a dummy cable. The
concentric neutral of the dummy cable should only be grounded at one end to avoid circulating currents. A
dummy cable experiences approximately the same heating conditions as the active cables but has no voltage on
the conductor except the voltage necessary to drive current in the conductor. This condition is complicated if the
insulation material of the active specimens has high dielectric losses and experiences heating due to the
application of voltage. To accurately determine what the actual temperature differences are between the dummy
cable and the active cables, a temperature profile must be established (see 8.4). Figure 4 shows a typical dummy
specimen with T/Cs located on the conductor in the air portion of the specimen, on the conductor at the midpoint
of the water section, and on the cable surface in the water.

Figure 5, Figure 6, and Figure 7 show progressively more complex heating control schematics using a dummy
cable. These heating control schematics are only examples. Other control schemes may be acceptable as long as
they provide for reproducible heating of the test specimens.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Table 5 Commonly used construction materials for tanks

Inner tank surface Tank insulation


Specimen Surface cover
Thickness, Thickness, mounts
Material mm Material mm
(in) (in)
1 SS* 2.2 (0.087) Foam 25 (1.0) Acrylic PL spheres
sheets
2 SS 0.8 (0.03) Fiberglass 75 (3.0) PVC pipe Plexiglass
racks

3 Vinyl liner Concrete struc- 150 (6) Fiberglass Styrofoam


ture with sand concrete sides racks
on bottom and and bottom
foam on sides 150 (6) sand
bottom only

4 SS 3.2 (0.125) Redwood 41 (1.63) Acrylic PL spheres


sheets
5 SS 3.2 (0.125) Redwood 41 (1.63) Acrylic PL spheres
sheets
6 SS 1.6 (0.06) Redwood 41 (1.63) Acrylic PL spheres
sheets
7 SS 1.6 (0.06) Redwood 41 (1.63) Acrylic PL spheres
sheets
8 SS 5 (0.19) Foam 51 (2.0) Polypropylen Polycarbonat
e e
9 Polyethylene 8 (0.32) Fiberglass 38 (1.5) None

1 Polyethylene Fiberglass None


0
1 Galvanized 1 (0.04) Foam 38 (1.5) Fiberglass
1 steel racks
with vinyl liner
*
Stainless steel
Polymer

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Maximum conductor temperatures of ambient to 90 C have been used in tank tests. In some cases, this
conductor temperature is measured on the section of the cable sample in air, and in other cases, it is measured on
the section of cable sample in water. The conductor temperature of the active test sample in water is the critical
temperature; the conductor temperature of the active test sample in air is of secondary importance. How- ever,
care should be taken to prevent overheating in this portion of the test sample. With water in the conductor, no
portion of the active test sample conductor should exceed 100 C.

The time-temperature profile of the active test specimen in water is a critical parameter in ACLT protocols.
Examples of some time-temperature profiles are shown in Figure 8 and Figure 9. It is not the purpose of this
guide to recommend a particular profile. The conductor temperature profiles, which are used in ACLT protocols,
should not be strictly Defined as there is still much to learn in developing a reliable ACLT. Different types of
cable insulations behave differently in the ACLT for the same test conditions. Typically used test temperatures
are listed in Table 5 and Table 6 in Clause 12.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
8.2 Water temperature

Throughout the tank, the water temperature should be uniform at any given time. This is most often achieved by
circulating the water in the tank and making provisions for evenly distributing the circulated water. If some form
of water circulation is provided and adequate precautions have been taken multiple temperature measurements
have been made at different locations in the tank to insure a uniform distribution of the circulated water, only
one temperature probe in the water tank should be necessary for monitoring purposes. Hollow plastic spheres or
styrofoam expanded polystyrene sheets are often used on the water surface to insulate it from the surrounding air
and to minimize water evaporation, and aid in water temperature control.

Water temperature may or may not be controlled. In order to maintain a consistent water temperature Profile,
cooling or heating may be required during the test cycle. Where water temperature is not controlled, the water
temperature simply rises and falls with the conductor heat cycles. In this case the water temperature will depend
on tank size, shape, water volume, number of cable specimens, etc.

8.3 Ambient temperature in the test area

Ambient temperature in the test area should be monitored and recorded. If the tank water temperature is not
controlled, it is desirable to control this variable as a means of achieving a more repeatable time-temperature
Profile.

8.4 Establishing a temperature Profile

On cable constructions that use insulation materials with low dielectric losses, it is possible to establish accurate
temperature Profiles by setting up a tank with cables of the same construction (not the actual ACLT specimens)
and applying heating load cycles without the application of high voltage. This allows T/Cs to be placed on the
conductors of the active specimens, which would normally be energized during the ACLT. It is important that
temperature Profiles be established for the following locations:
a) Active test cables
1) Conductor temperature in water and air

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
2) Cable surface temperature in water and/or air
b) Dummy cable, if used for temperature monitoring and control
1) Conductor temperature in water and air
2) Cable surface temperature in water and/or air
c) Water temperature
d) Ambient temperatureair

The temperature of the conductor will vary between the water level and the stress cone and could be higher
under the stress cone. Dummy cables used for temperature monitoring and control may not experience the same
heating currents as active test cables due to losses in the iron core used to isolate the dummy cable from the high-
voltage source. These losses can result in slightly different conductor temperatures on the dummy cable and the
active test cables. These temperature differences are believed known to be very small and can be determined
while establishing the temperature Profiles.

Conductor heating currents in both the active test cables and the dummy test cable (if used) should be mea- sured
and recorded while establishing the temperature Profiles since these currents have a direct relationship to
conductor temperatures. These currents should be periodically compared to the conductor heating currents
measured during the ACLT as a means of verifying that no Significant changes have occurred since the
temperature Profiles were established. All conductor heating currents should be measured with a true RMS
meter.

For cables exhibiting high dielectric loss (V2C tan ) characteristics (measured and calculated under test
conditions), the energized cables will run hotter than an unenergized dummy cable in the same tank.
Dielectric heating can also occur in tests conducted at ambient temperature. For these cases, an additional
Correction factor or adjustment is required to ensure the energized cable is at the Specified test temperature. One
simple and practical method to achieve temperature control is to lower conductor loss (I2R) by an amountequal
to the dielectric loss calculated from the actual measured cable characteristics. If dielectric heating is Significant
during tests at ambient temperature, cooling of the water may be necessary.

8.5 Monitoring/controlling temperatures during the ACLT

Once the temperature Profiles have been established and the actual ACLT cable specimens have been placed into
the tank and energized with the application of high voltage, it is important that the temperatures at the following
locations be monitored and/or controlled:

a) Active test cables


1) Cable surface temperature in water and/or air
b) Dummy cable if used for temperature monitoring and control
1) Conductor temperature in water and/or air
2) Cable surface temperature in water and/or air
c) Water temperature
d) Ambient temperatureair

It is intended that there should be excellent correlation between the values measured during the ACLT testing
and those values obtained from the temperature Profile measurements.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
8.6 Temperature measurement techniques

Accurate measurements of conductor temperature or cable surface temperatures with T/Cs require that proper
techniques for preparation and mounting of the T/Cs be used. The Several techniques given in Annex C have
been developed for the preparation and mounting of T/Cs. These techniques are a compilation of those used in
use at several laboratories throughout North America.

8.7 Recommendations

The common practices and recommended test temperatures are listed in Table 4.

9. Water

Water is one of the primary test parameters in ACLT tank-type testing. Since moisture in the insulation and
shield materials, and its ability to transport ions, is linked to cable aging as manifested through water tree
formation (Plissou [B22], UNIPEDE-DISCAB [B30]), careful control and monitoring of the water quality are
essential during the test.

9.1 Water nature

In tank-type aging tests, water is used to fill the tank and the interstitial spaces of the specimens inner conductor
and to make up evaporative losses from both the tank and the conductor. Both deionized water and tap water
have been used by different laboratories but no systematic comparison has been carried out on identical cable
specimens in the same laboratory. Tap water contains different ions and impurities depending on its source and
even the time of year so that it would be expected that it would have a significant influence on the test results.
Deionized water contains less ions and impurities but tends to become contaminated with time during a test
(see Figure 10). Thus, even with tests using deionized water, the nature of the water does not remain constant
with time, and it is not known how much these variations affect test data. Equipment to deionize the large
volumes of water, particularly if several tanks are used, adds to the cost of testing and has to be maintained
regularly. Until the results of comparative studies of the effects of deionized versus tap water are known, it is
preferred that the water for the tank and the conductor be tap water.

sample inner conductor and, also, to make up evaporative losses from both the tank and the conductor. It is
preferred that the water for the tank be initially deionized (charcoal preFilter, using cation and anion resin beds)
with a resistivity not less than 1000 m. Since the water quality can be affected by electrical breakdown,
insertion of new cable samples and the mixing of different cable types, the water in the tanks should be changed
totally or partially when the resistivity is less than 250 m. Based on the data shown in Figure 10 for six tanks
over a period of more than 36 months (EPRI Report TR-108405-V2 [B14]) it is recommended that the water be
checked by monitoring the pH and resistivity monthly. Ion nature and concentration may play a role in the life of
cables in the ACLT (EPRI Report TR-108405-V2 [B14]). These records will allow variations within a tank and
between tanks to be tracked as a function of time, and ultimately could help to explain particular sample fail-
ures. Furthermore, an adequate water/cable volume ratio should be Fixed and maintained during the aging
period, as it infuences the impurity concentration, hence the frequency of water changes, and, also, the time/
conductor temperature characteristics. Two ratios have already been suggested (120/1 in Katz, et al., [B16] and
47/1 in Walton, et al., [B29]) and at least 40/1 is recommended. To keep the water volume constant, refilling can
be performed automatically by means of a floating valve to a control level of 13 mm (0.5 in), using deionized
water similar to that initially used. The water should be circulated continuously to avoid hot spots, within a
temperature variation of 2C, to ensure uniform aging.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
9.2 Conductor strand blockage

To make sure all interstices are filled, a filling water pressure of 70 kPa to 200 kPa (10 psi to 30 psi) should be
applied at one end of each cable specimen. Pressurization for filling or checking for strand blockage should only
be done when the water temperature is 45 C. Since water corrodes the strands over time, particularly those
made from aluminum (Luzzi [B17]), they should be checked for blockage when the cable is at room temperature
using water at a sufficient pressure. Pressures up to 200 kPa (30 psi) may be used. If no water flow is observed
within 5 min the corrective actions listed in Clause 14 should be investigated. If frequent blockages are
occurring, more regular checks should be considered.

It is cautioned that if the water quality changes, be it tap water, deionized water, or controlled ion salt water (e.g.,
0.1 normal NaCl), the ACLT results can change. Different aqueous solutions can yield different ACLT results,
even though all other conditions are equal. This is because different ions can penetrate into the insulation wall at
different rates and can induce different aging and failure results.
Similar recommendations apply to the water in the cable strands. For the sake of uniformity and for practi- cal
reasons, it is usually similar to the surrounding water (charcoal preFilter, using cation and anion resin beds) with
a resistivity not less than 1000 m.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
10. Voltage

Voltage stress is a critical parameter for the aging process and is required, along with water, for water tree
growth. Its level, stability, and quality are all important parameters that must be regulated and recorded. It should
be noted that water tree initiation and growth are a function of localized electric stresses that cannot be directly
measured.

The values of the test voltages arrived at by consensus of the working group based on testing experience are as
follows:

a) As a materials test on typical cable designs using time-to-failure ACLT, the prevailing test voltage is
4 times rated voltage to ground (4V0).

b) As a materials test on typical cable designs using fixed aging time(s) in the tanks followed by a diagnostic
test such as a step ACBD test, the prevailing test voltage is 3 times rated voltage to ground (3V0). By reducing
the test voltage, the cable samples are less likely to fail during the designated aging time, thus making more
cable samples available for the diagnostic measurement at the completion of the designated aging time.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
These recommendations are not unanimous. There was concern that testing at 4V0 resulted in electrical stresses
that were too high and had the potential of introducing failure mechanisms that may not be encountered in the
field. However there were similarities in the Weibull shape parameter for time-to-failure test results generated at
voltage stresses of 4V0, 3V0, and 2V0. This indicated a similar failure mechanism at the three test voltages.

With regard to testing reduced wall cables, the limit on the maximum average test stress for any cable design
should be 12 kV/mm. In Europe the typical accelerated stress for medium-voltage cables is 5 kV/mm.

The voltage is applied between the cable conductor and the metal shield or neutral wires. One end of the neutral
of each cable is connected, either directly or through a fuse, to the high-voltage circuit ground. The voltage level
should be controlled to within at least 5% and be free from substantial harmonic content. Harmonic content of
the test voltage can be specified by total harmonic distortion (THD), but the overall effects of harmonics on the
aging of the specimens have not been investigated. The average voltage stress level chosen is left to the
individual, but is usually between 2 kV/mm and 8 kV/mm. The relationship to rated line to ground voltage may
be preferred, between 2V0 and 4V0. It must be noted that while aging may be accelerated using above normal
voltage stresses, too high a stress might introduce a mechanism of aging that would not occur at normal stress. It
is important that the voltage measuring equipment be calibrated regularly, preferably at least once a year. Also
the voltage should be continuously monitored regularly for several days to ensure the voltage variation is within
5%. The monitoring should include weekdays and a weekend. As an example, if the aging is assumed to
follow an inverse power law model (Vnt = constant) with an exponent n of 4, a voltage variation or error in
calibration of 5% will result in a time variation of 20%. If the voltage is controlled to 3%, the time
variation will be reduced to 12%. If the voltage exponent n is 6, voltage variations of 3% and 5% will
result in time variations of 18% and 30% respectively. Thus limiting the voltage variations to 3% is
preferred. Typically used test temperatures and voltages are shown in Table 5 and Table 6 in Clause 12.

Interruptions to the test voltage are needed for the maintenance of water levels, etc., and also necessary when
specimens fail. The duration of an interruption must be minimized by having replacement specimens available
and methods to detect a failed cable in place before startup. The time and duration of the interruptions should be
recorded. Damage to the specimen during failure can be limited by providing a high source impedance power
supply, or alternatively, an overcurrent trip device in the circuit. For example, the neutral of each test cable may
be connected to the high-voltage circuit ground through a fuse (~2 A) to facilitate identification of a failed cable.

Voltage frequency is usually specified as either 50 Hz or 60 Hz. Extensive tests have been done up to 8 kHz, but
frequencies higher than 50 Hz or 60 Hz require expensive power supplies that must be protected against surges
when specimens fail as the output stages of the power supplies are vulnerable to damage. For these reasons, no
tank tests are currently performed at high frequency in North America. Different results may occur when
different frequencies are used.

In some test procedures, voltage surges are introduced intentionally during an aging program to simulate the
effects of lightning impulses and switching surges that occur during the service life of an underground cable
(Hartlein, Harper, Ng [B11], Katz, Seman, Bernstein [B15]). Surges can occur unintentionally when cable
specimens fail due to the test circuit setup. If the aging test area is located near a potential site of external surge
sources, the test setup should be checked for voltage transients. The magnitude, wave shape, number, and
frequency of intentionally applied impulses or surges should be accurately recorded.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
It should be noted that dc voltage has also been applied during aging studies and has been shown to influence
cable life (Srinivas, Bernstein, Decker [B26]). If dc voltage is used in the test procedure, the test method should
be noted, e.g., voltage level, time of application, and time of grounding. DC hipot testing is not recommended in
ACLT tests.

Termination of the cable specimens should receive special consideration as normal stress cones will be
overstressed during the aging and could lead to premature failure. An over-designed cable termination system
should be considered to provide adequate stress relief at the expected high-voltage level for the duration of the
test program. Premature stress cone failures are expensive and time-consuming.

Diagnostic tests (see Clause 15) using high voltages may be carried out both before and after the aging test.
These tests should be performed as soon as possible after the specimen is removed from the aging test program
as voltage relaxation can change the result obtained. The specimens should be stored in water to reduce drying.
Test voltage measurements should be made on the high voltage side with a suitable divider, continuously
monitored and displayed.

11. Time

Both timer days (timer hours that the test voltage was on divided by 24) and the total number of daily 8 h heat
cycles (only count heat cycle if the desired temperature was maintained for a minimum of 4 h) must be specified.

12. Test matrix

As mentioned earlier, two of the most significant cable aging parameters are conductor temperature in the water
and voltage stress. Because of the complex nature of accelerated aging tests on extruded dielectric cables, this
guide does not specify these or any other test parameters. However, in an effort to allow a comparison of test
results developed at two different laboratories, a matrix of commonly used test voltages and conductor
temperatures in the water is provided in Table 5 and Table 6. The values were selected to cover test parameters
typically used by cable researchers. They are also broad enough to allow for a wide variety of aging conditions.
It is not intended that a cable be evaluated by subjecting lengths to all test conditions but that laboratories select
one (or more) set(s) contained in the tables. In time, a specific set of test conditions may evolve that could then
be acceptable as a standard. This set of conditions might be different for each particular insulation system, i.e.,
the accelerated test conditions for XLPE may not be the same as those for TRXLPE or EPR.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Table 5Designation of common conductor temperatures in the water and voltage test
conditionsACLT (time-to-failure tests)

Designation of test conditions


Test voltage
30 C or ambient 45 C a 60 C a 75 C a

V0 b 1/30 1/45 1/60 1/75


2V0 2/30 2/45 2/60 2/75

3V0 3/30 3/45 3/60 3/75

4V0 4/30 4/45 4/60 4/75


a
Conductor temperatures in air may be higher.
b Line to ground voltage (8.7 kV for 15 kV cables).

The voltages are applied between the conductors and the neutrals of the cables. The temperature is generally the
maximum conductor temperature achieved during the 8 h current-on period when load cycling is used. To
reliably compare test results from different laboratories, the two laboratories should use the same test protocol.
This includes not only the same test voltage and the maximum test temperature, but also the same temperature
profile during a load cycle if load cycling is used. That is, the curve of time versus temperature should be the
same for both laboratories. The temperature gradient across the insulation should also be the same for both
laboratories. A reliable comparison of test results is only possible if both laboratories subject test specimens to
the same test parameters.

Table 6Designation of common conductor temperatures in the water and voltage test
conditionsFixed time tests followed by diagnostic testing such as an ACBD test

Designation of test conditions


Test voltage
30 C or ambient 45 C a 60 C a 75 C a

V0 b 1/30 1/45 1/60 1/75


2V0 2/30 2/45 2/60 2/75

3V0 3/30 3/45 3/60 3/75


a
Conductor temperatures in air may be higher.
b
Line to ground voltage (8.7 kV for 15 kV cables).

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
13. Failure
The following information is considered critical for a thorough test evaluation. Also, in Table 7 a list of codes is
provided to facilitate reporting the data. It is important that the locations of the failures be reported since the
cables may have seen different temperatures at different locations in the tank. The method of locating the failure
should also be reported.

Table 7Code designations

Failure codes Type

A In air
T Termination

WL At waterline

W In water

D Removed for diagnostics

C Censored from data analysis, but should be reported and used in


conjunction with one of the above

M Mechanical damage

Abnormality codes See Clause 14

Measurements
t Insulation wall thickness at failure
R1 Radius under insulation

R2 Radius over insulation

L Active cable length (shielded cable length)

Suggested information to report on each specimen in test evaluation: Position in tank


Time under voltage Number of current cycles Test days, i.e., time in water
Type failure (A, T, WL, W, D, C, M) Abnormalities
Measurements (t, R1, R2, L)

14. Abnormalities

A normal test consists of applying voltage of a specific magnitude and with temperature cycling, if used, applied
at the same time to specimens immersed in a tank. Although this type of testing is intended to accelerate cable
aging, a single evaluation can go on for months and even years before it is completed. Thus, it is expected that an
evaluation will have a number of abnormalities (see the definition in Clause 3). Historically, many abnormalities
have not been reported with the test results because their significance was not clearly known. At the same time,
not reporting abnormalities could be viewed as censoring of information by those who want the opportunity to
form independent conclusions on the meaning of the results. The user may choose to censor the abnormal
failures, but they must be reported. The user may provide data analysis with or without these failures.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Abnormalities may be classified in four groups with respect to the effect on the outcome of the test.

Class I. These result in immediate specimen failure (e.g., termination failure) [see item a) in following list].
Class II. These seriously affect the test outcome, but do not result in immediate specimen failure
(e.g, excessive temperature fluctuations) [see item b) in following list].

Class III. These have a minimal effect on the test outcomer (e.g., minor temperature fluctuations) [see item c)
in following list].

Class IV. These specimen failures are physical or statistical outliers (e.g., failures that occur due to a
different mechanism than expected, and failures that are not expected or explainable) [see item d) in following
list].

Experience shows that the following abnormalities and corrective actions have occurred during past tank tests:

a) Class I (result in immediate specimen failure) Abnormalities


Termination failure (includes stress cone failure)
High-voltage test specimen conductor connector failure
Waterline failure7
Air failure7
Failure at cable support point
Corrective action options
1) Log failure as abnormality; reterminate specimen (or splice); continue test; perform failure analysis;
analyze statistically to determine if it is an outlier.

2) Log failure as abnormality; reterminate (or splice); continue test.

3) Log as failure; treat as censored data value in analysis; replace with dummy specimen;8 continue test.
4) Log failure as abnormality; treat as censored data value in analysis; remove specimen and start new
specimen.
5) Analyze data to determine how long a hot connection may have been affecting aging conditions; then
follow the previous steps 1, 2, 3, or 4.

6) Log failure as abnormality; remove specimen and start new specimen; perform failure analysis; analyze to
see if outlier.
b) Class II (seriously affect the test outcome) Abnormalities
Extended periods (> 2 h) above or below temperature tolerance for the entire sample before any failures have
occurred, or for specific specimens remaining after failures have occurred.
Extended periods (> 24 h) above or below voltage tolerance for the entire sample before any failures have
occurred, or for specific specimens remaining after failures have occurred.

Blocked conductor strands for any specimen. Corrective action options


1) Terminate test.
2) Adjust number of completed load cycles, but do not change test days. It has been shown that if the timer days
and the load-cycling days do not differ by more than 10%, the analysis is not affected.
3) Adjust the number of test days but do not change the load-cycle days. It has been shown that if the timer days
and the load-cycling days do not differ by more than 10%, the analysis is not affected.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
4) Quantify abnormality and report with results.

5) Remove entire sample population or specimen and start new sample population or specimen.
6) Remove conductor blockage with pressure up to 2 atm (30 psi) with the cable at room temperature. If unable
to remove blockage, treat blocked specimens as deletions in statistical data analysis.
c) Class III (minimal effect on the aging outcome) Abnormalities
Tank water level outside tolerance.
Short periods (< 2 h) beyond temperature tolerance.
Short periods (< 24 h) beyond voltage tolerance
. Corrective action options

1) Log length and magnitude of abnormality.


8 This dummy specimen is not the same dummy cable that has T/Cs embedded in it for conductor temperature control and monitoring (see
8.1). A dummy specimen should be used to fill vacated slots in tanks that do not employ water temperature control, since each sample in the
test population is in fact a heating element during the conductor load cycle. In order to maintain the same temperature profile on each of the
remaining test samples in the population for the duration of the ACLT, a failed sample should be replaced with a dummy specimen of a
similar construction and length as the failed sample. This dummy is not considered a part of the original population since no water is injected
into the strands to reduce the probability of a failure. Replacement of failed cables with dummy specimens in water temperature controlled
tanks should not be necessary in order to maintain the same temperature profile on each of the remaining test samples in the population for
the duration of the ACLT.
7 If there are three or more failures of this type, they may be analyzed separately

d) Class IV (outliers) Abnormalities


Very early (obvious) outlier. Less than 10% of expected mean, geometric mean or charac- teristic time to
failure or mean breakdown strength.
Early outlier. Not obvious at First, but after data analysis, it is shown to be an outlier.
Late outlier (time to failure).

Corrective action options


1) Analyze for cause. If cause found, replace with new specimen.
2) Analyze for cause. If cause found, replace with dummy specimen. 9
3) Analyze for cause. If cause not found, replace with new specimen.
4) Analyze for cause. If cause not found, replace with dummy specimen. 9

4) Outliers should be deleted from data analysis (and should be noted), but should not be deleted from the
reported data that accompanies data analysis.

Data censoring consists of two possible actions; deleting outliers (statistical or physical), or suspending data
values. Deleting actually means not using that value in the statistical data analysis. Suspended values are used in
the data analysis. The effect of suspensions is to optimistically overestimate the scale (mean) parameter of the
distribution. The effect of a deletion is to accurately estimate the scale parameter, but widen the confidence
bounds of the data set distribution by reducing the population size, thereby increasing the possibility of passing
the hypothesis test in a comparative data analysis.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
15. Diagnostics

The following are commonly used evaluation or diagnostic techniques:


Time-to failure
Fixed time of aging followed by ac dielectric strength assessment
Truncated population time to failure with ac dielectric strength assessment on remaining specimens

Other diagnostic tests that may be used are impulse breakdown tests (Katz, Seman, and Bernstein [B15]) and
water tree counts (AEIC CS5 ANSI/ICEA S-94, AEIC CS6-96-649 and ANSI/ICEA S-97-682).

When ac dielectric strength is a part of the evaluation, the breakdown strength of unaged cables should also be
measured and the statistical parameters calculated. A F 5 min step-rise test is commonly used (AEIC C S 5-
ANSI/ICEA S-94, AEIC CS6-96-649 and ANSI/ICEA S-97-682) that starts at 4 kV/mm (100 V/mil) and
increases in 1.6 kV/mm (40 V/ mil) steps each5 min duration, until failure of the insulation occurs. The 1.6
kV/mm (40 V/mil) step should be based on the design insulation wall thickness [e.g., a 4.4 mm (0.175 in) design
wall equals 7 kV steps].

The ACBD test failures should be reported as having been removed from the tank for diagnostics (failure code
D), as previously described in Clause 13. This guide recommends the following information be reported on
ACBD test failures:
kV step in which failure occurred.

Time between removal of tank test voltage from specimen and start of voltage for ACBD test.
Time between removal of specimen from tank water and start of voltage for ACBD test.

9 See Clause 14 for explanation of dummy specimen.

Thickness of insulation at point of failure.


Location of failure along cable.

Direction of failure

Along with the destructive tests previously noted, nondestructive tests may also be employed as a diagnostic
tool. However, these are not as common or widely used at present as diagnostic techniques during
accelerated aging tests, as are the destructive ACBD or impulse tests, or water tree analysis previously described.
Both global and non-global diagnostic tests can potentially be employed (i.e., determining the properties of the
entire cable length in the tank or determining the condition of the cable at specific locations). Global tests are
represented by capacitance and/or tan delta measurements using power frequency or very low frequency
(VLF, e.g., 0.1 Hz) (Kuschel et al. [B16], Thrning et al. [B29]), or a range of frequencies (dielectric
spectroscopy) (Hvidsten et al. [B12]). These tests require that care be Although not widely used at present as a
diagnostic technique in accelerated aging tests in the laboratory capacitance and/or tan delta measurements using
power frequency or very low frequency (VLF) e.g., 0.1 Hz, are being actively considered for measurements on
cables in the Field (Kuschel, et al., [B17], Tharning, et al., [B27]). Capacitance and/or tan delta measurements
could be employed as a diagnostic tool for aging tests in tanks although care should be taken to isolate the cable
being measured from other cables in the tank. If such tests are being considered they should be performed at the
start of the aging and at regular intervals during the aging. At least three voltage levels (e.g., 0.5V0) up to the
aging voltage, are recommended (Densley [B8]). In addition both off-line PD measurements and online signal
measurements including PD may also be employed.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
16. Data analysis

The reporting of failures and their codes has already been discussed. This clause will deal with how reported
failure data is to be analyzed. Failure data generated from the accelerated life test is often analyzed using a
probability distribution such as the Weibull or log normal (Abernathy [B2], Nelson [B21]). The analysis may be
done manually, but it is recommended that a computer program be used for the calculations (Abernathy [B2],
Nelson [B21]). Two packages are widely used.10 For data sets where the population size is < 20, the distribution
used should be the Weibull. However, a distribution analysis to determine the best-fit distribution should be
performed and is readily available in the recommended software packages. If the plot of the data is done
manually, then the analysis of the data with the distribution (Weibull or lognormal) providing the largest value of
the coefficient of determination (r2) is the proper distribution. Acceptable data fits to distributions are
characterized by r2 values > 0.90. However, when comparing two or more data sets, the distributions must all
be identical. The log normal distribution should be considered if the fit with the Weibull distribution is poor, i.e.,
if the r2 value of the regression of the Weibull distribution is < 0.9.

Typically, most failures occur in the section of cable submerged in water (failure code W). However, some
failures also occur at the water line (failure code WL), in the air (failure code A), or in the termination (failure
code T). There may also be failures that occur due to mechanical damage (M). Any combination of these failure
types may be included in the statistical analysis, but if the failure is not in the water, the infor- mation on
abnormality failure given in Clause 12 should be used to report it. For example, a waterline, air,. However, all
of these failure codes should be treated as suspensions, as they did not occur in the area of the specimen that has
water in the conductor strands and outside of the cable specimen, the desired and intended failure location. Also,
specimens removed for destructive diagnostic testing (D), such as ACBD or termination failure impulse
breakdown testing, should be reported when analyzing the data.

When reporting failure test results, it is very important to identify all failure types included in the data anal- ysis.
This is easily accomplished through the use of failure code(s) A, W, and WL.

It is also important to note that termination failures (failure code T) are usually not included in the statistical
analysis. This is because as suspensions, using the voltage stress is often much higher in the section of cable
inside the termina- tion. Additionally, the temperature gradient in the section of cable inside the termination is
significantly influenced by the termination aging times at which they were removed from the test.
The Weibull analysis may be done manually, but it is recommended that a computer program be used for the
calculations (Abernathy [B2], Nelson [B22]).

Failures that are determined to be physical outliers (failed due to some other reason than the intended failure
mechanism) or statistical outliers (early or late failures, also known as perpetual survivors) should be deleted
from the data analysis. These outliers should have also been reported in the data report and should be noted in
the data analysis.

When reporting failure test results, it is very important to identify all failure types included in the data analysis.
This is easily accomplished through the use of failure code(s) A, W, and WL.
10 Fulton Findings LLC SuperSMITH and ReliaSoft's Weibull++. This information is given for the convenience of users of this guide and
does not constitute an endorsement by the IEEE of these products. Equivalent products may be used if they can be shown to lead to the same
results.

It is important to include termination failures (failure code T) in the statistical analysis, because this implies that
the specimen section in the preferred failure location was capable of surviving beyond the time of the
termination failure and would have failed at a later time.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
The following information should be reported in the data analysis:
a) The characteristic life (or scale parameter), referred to as eta () or alpha () for the Weibull distribution
in the literature, and Defined as the time or value at which 63.2% of the specimens have failed. It is usually
calculated by a computer program, or can be read from the cumulative distribution function plot. Alternatively,
the geometric mean time to failure (GMTF), defined as the time or value at which 50% of the specimens have
failed, and referred to as MuAL (-mean antilog). Both scale parameters are usually calculated by a computer
program or can be read from their cumulative distribution function plots.
b) The shape factor of the Weibull distribution, usually given as , and defined as the slope of the plotted line.
The shape factor of the lognormal distribution, usually given as SigF, and also defined as the slope of the plotted
line. Both of these distribution shape parameters are usually calculated by a computer program, or can be
calculated from their cumulative distribution function plots
c) The number of specimens tested, and suspended, if any.
d) The number of deletions (outliers), if any.
e) The two-sided 90% Confidence limits for the entire set of failure times of the plotted line. Ft.
f) The goodness of fit, as calculated by the coefficient of determination, r2.
g) The plot of the data.

The analysis should be reported as complete if all the specimens of the test population have failed and are
included. The analysis should be reported as censored if some specimens have not yet failed, or if
abnormalities are included. In the latter case it is essential that a computer program be used, and that the method
of analysis giving the least bias in the results is used. Although the method of maximum likelihood estimators
(mle) is statistically correct and does handle the censored data case, it is best suited for specimen sizes > 50. The
method of linear regression (LR, X on Y) is the method that gives the least bias for data sets with small specimen
sizes (i.e., < 50). The general thinking is that LR is best for 80% of cases, and mle for 20%. Therefore it is
important not to rule out the mle method, but LR should be the initial default method.

Many computer simulations for techniques for statistically analyzing lifetime data with Weibull statistics exist.
Some of these may be found in Abernathy [B2] and Nelson [B21].

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
17. Final report

The Final report should contain all the important data about the test conditions and the results. An example of the
typical information to be included is given in Annex D.

Although it is preferred that the test conditions be reported in full (e.g., 4V0, 75 C and deionized water) they
sometimes are reported in an abbreviated form. As the most important parameters are voltage magnitude,
maximum temperature of the conductor in the water, and the water characteristics, these three parameters could
be arranged as follows:
X/Y where X is the ratio of the test to the rated voltages, Y is the temperature in C. For example 4/75 would
represent the test conditions of 4V0 at 75C. Alternatively, the temperatures could be assigned digits for Specific
values (e.g., 45 C could be represented by 1, 60 C by 2). If salts are added to the water, the concentration in
grams per liter could be included, for example 4/75/5 would represent 4V0, 75 C and 5 g/l water solution.

18. Conclusions
The guide discusses the typical test parameters to be controlled in a tank test to accelerate the aging of medium-
voltage extruded power cables due to water trees. It also describes techniques to control and mea- sure these
parameters.

In the first edition of this guide it was not possible to recommend specific test conditions. In this second edition
it has been possible to narrow the range of recommended test parameters such as tank type and size, voltage, and
temperature.

Following the recommendations ensures that aging tests performed in different laboratories according to a
Specific set of aging conditions will be conducted having well-controlled critical parameters and will result in
the cables in the laboratories being subjected to the same test conditions. In addition, following the suggestions
for the test specimens, tank type and size, preconditioning, and data analysis should also maintain a high degree
of consistency in the data between laboratories. Unusual aging conditions that can affect the test results have
been discussed along with relevant data to be included in a test report.

As more data are collected, it is anticipated that controversial issues discussed in this guide will be clarified in
future editions. In addition it is hoped that in future editions specific test conditions such as voltage and
temperature, can be recommended.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Annex A

(informative)

Preconditioning

Preconditioning of test specimens, see Clause 6, is carried out to limit the concentrations of cross-linking by-
products, which can affect the aging of XLPE cables. Although there will be similar cross-linking by- products
in TRXLPE and EPR cables, not enough is known about the effects of by-products on these types of cables.

The most commonly used chemical cross-linking agent, dicumyl peroxide, produces a number of cross- linking
agent by-products that include acetophenone, cumyl alcohol, -methyl styrene, and smaller quantities of other
by-products. Methods for by-product analysis and data for new and aged cables have been reported (Damon,
Huang, Johnson [B7], Groeger et al. [B10]).

Residual cross-linking agent by-products can influence the electrical properties of polymeric insulating materials
such as XLPE, TRXLPE, and EPR. The residuals will depend on the specific chemical reactions, which occur
during manufacture and the rate at which the by-products subsequently diffuse through the extruded layers. In
carefully controlled experiments on XLPE molded slabs and ribbons, removed from cables, the short-term 60 Hz
breakdown strength increased but the impulse breakdown strength decreased with increased concentration of by-
products (Damon, Huang, Johnson [B7], Eichhorn [B9], Wartusch and Wagner [B33], Yan et al. [B34]). The
ACBD strength of XLPE increased between 10% and 15% when the by-products were present up to 2% but
decreased when the by-product concentrations exceeded 2% (Yan et al. [B34]). This study also showed that
mixtures of the volatile by-products gave synergistic results. The electrical and water treeing resistance was
improved by the presence of by-products (Ashcraft [B3], Wartusch and Wagner [B33]). Szaniszlo [B28] has
shown that the ACBD strength of TRXLPE 25 kV cables decreased with increasing acetophenone content
(contrary to conventional XLPE).The ACBD strength did not show a good correlation with another by-product,
dimethylbenzyl alcohol. Although this discussion is not intended to be a comprehensive literature review, the
results show that the influence of the by-products on breakdown strength is quite complex.

Limited data are available on the effects of by-products on the long-term behavior of either molded slabs or
cables. Controlled tests to examine the long-term effects are difficult to carry out due to the already mentioned
migratory nature of the by-products. An examination of new and service-aged steam-cured XLPE cables showed
that the acetophenone level decreased from approximately 3000 ppm to 40 ppm during the first two years of
installation and that very little (< 15 ppm) remained after 8 to 10 years (Martin and Hartlein [B19]). The loss of
by-products also occurred when a cable was stored on the reel. It was concluded that the loss of acetophenone
would be expected to increase the water tree growth rate. In another study of steam-cured XLPE cables that had
failed after 6 to 13 years in service at six utilities (Plissou [B23], Katz et al. [B14]) organic volatiles and halos
were present in many of the cables. The total organic levels in the aged cables were < 100 ppm. Although the
levels of acetophenone were < 10 ppm, the levels of the cross-linking agent, dicumyl peroxide, were found to be
variable, typically < 20 ppm but up to 70 ppm in one cable after
11 years in service. The halo contained several thousand ppm of water. The ACBD strength was lower in the
cables with halos.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
The most common form of preconditioning is to subject the cables to elevated temperatures to increase the rate
of diffusion of the by-products from the insulation. For example, after heating the conductor of steam- cured 15
kV cables by circulating current to 90 C 2 C for 500 h in an open atmosphere, the amounts of acetophenone
and -methyl styrene were reduced to less than 10 ppm and cumyl alcohol to less than 100 ppm (Abdolall [B1]).
The effects of three types of thermal preconditioning on dry-cured 15 kV XLPE cables aged in tanks have been
studied (Walton, Smith, Thue [B30]). The preconditioning procedures were as follows:
a) Cyclically heat the conductor to 90 C for 8 h on and 16 h off for 120 h.
b) Heat the conductor to 60 C continuously for 120 h.
c) Heat the conductor continuously until the temperature of the insulation shield reaches 110 C and
immediately switch off the conductor heating current.

The results are shown in Table A.1.

Table A.1Results of preconditioning tests on 15 kV XLPE cables aged in tanks

GMTF in days
(90, +90% confidence limits)
Age of cable
Treatment of preconditioning treatment after
extrusion Aged at 4V0
(months) Aged at 3V0 and
And 90 C a 75 C a(3/75)
(4/90)

None (i.e., fresh cable) 2 64 (56, 74) 386 (366, 406)


A 90 C on conductor cycled 8 h 60 45 (40, 50) 264 (214, 325)
on 16 h off for 120 h

B 60 C on conductor 60 46 (37, 57) 321 (309, 333)


continuous for 120 h

C Insulation shield to 110 C 54 103 (85, 124) 325 (284, 371)


then immediate cool down
a Maximum temperature during load cycling.

Preconditioning treatment A subjected the cables to a total 40 h at 90 C, 120 h at 60 C for the cables in
treatment B, while the conductor temperature in treatment C was not measured but was probably significantly
greater than 110 C for a short time (on the order of minutes). Neither the amounts of cross- linking by-products
nor the morphology were measured before or after the preconditioning. Treatments A and B would be expected
to yield different insulation morphologies. Melting and recrystallization will occur through part of the cable
insulation wall thickness. Treatment C probably resulted in a temperature at the conductor in excess of 130 C
for a short time; this could lead to diffusion of species into or out of the shields into the insulation in addition to
morphological changes, the rearrangement of crystalline imperfections, and relaxation of the insulation.
Complete melting of the entire wall was probable as the temperature reached 110 C; hence re-melting and
recrystallization of the entire wall occurred, in contrast to procedures A and B.

Preconditioning treatments A and B yielded similar values of GMTF for the 4/90 conditions. Although there was
a 20% difference under 3/75 conditions, there was considerable scatter in the data, particularly for treatment A.
Without knowledge of the amounts of the cross-linking by-products or the morphologies, it is difficult to
interpret the results with respect to the effects of the cross-linking by-products or morphology. The results
demonstrate that there is a wide variation in GMTF as the preconditioning procedure is changed for XLPE, and
therefore great care is required in applying a preconditioning procedure. Good temperature measurement and
control are essential. The effect of such temperature variations on TRXLPE and EPR is not currently known.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
The recommended preconditioning protocols, listed in Table 2 in Clause 6, are 360 h at 90 C in free air or in a
circulating-air oven (for cable core testing on a common design as well as cable design testing).
Table A.2 lists a number of other preconditioning protocols that have been used in previous tank testing efforts at
various laboratories.

Table A.2Preconditioning protocols

List of preconditioning protocols used in different laboratories

1 No preconditioning (as received)


2 72 h at 75 C conductor in free air a
3 72 h at 90 C conductor in free air a
4 50 h at 75 C, followed by 150 h at 90 C conductor in free air a
5 50 h at 90 C, followed by 150 h at 130 C conductor in free air a
6 110 C instantaneous at insulation shield

7 120 h at 60 C conductor in free air a


8 120 h (8 on/16 off) at 90 C conductor in free air a
9 200 h at 90 C conductor in free air a
10 120 days AWTT

11 500 h at 90 C conductor in free air a


a When cable specimens are preconditioned by circulating current in the conductor, the temperature from specimen to specimen
and/or along the cable length will likely vary as much as 10 C. This variation is a function of several factors including the
orientation of the specimens, the distance between the specimens and/or coils and adjacent structures, and the location of the
specimen and/or coil relative to the group of specimens being preconditioned.

Heating cables in a circulating-air oven should reduce the variations in temperature from specimen to specimen
and along the cable length that can occur when specimens are heated by circulating current in open air. It must
be noted that heating in a circulating-air oven subjects the complete cable core to the oven temperature, unlike
conductor heating when the insulation shield will be about 10 C lower than the conductor temperature. The
higher temperature of the insulation shield could increase the diffusion of volatile products from the shield into
the insulation.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Annex B

(informative)

Water tanksHistorical information

When accelerated aging testing of medium-voltage cables began in the 1980s, several types and sizes of tanks
were used and described in the first edition of this guide. This annex gives a summary of the different tanks that
have been used and are still in use. The recommended tank material is stainless steel and the dimensions are
listed in Table 3 in Clause 7.

Figure 1, Figure 2, and Figure 3 in Clause 7 show several tank layout schematics that are in use. They provide a
reference for the following paragraphs on tank structure in this annex. It should be noted that not all elements
shown in the schematics will necessarily be used for all test setups. In Table B.1 and Table B.2, capitalized
letters in parentheses refer to elements in the schematics.

The tank structure is as follows:

Tank shape. The typical reservoir shape is rectangular, but others such as cylindrical, as shown in
Figure 3, are in use. Tank shape should be reported.

Tank size. The tank used should be large enough to accommodate at least 10 cable specimens or a single
cable equivalent in length to 10 specimens plus any control specimens such as a dummy loop used in induced
current heating/temperature monitoring. There should be sufficient distance between cable specimens and tank
sides, bottom, and other cable specimens to prevent uneven heating of any individual specimens and to allow
adequate circulation of the water.

When data from various tests are compared, the number of cable specimens plus the following tank size
variables should be considered: length, width, height (in the case of generally rectangular shapes or the
appropriate dimensions for other shapes), and depth of water. Where necessary because of unusual shape or
design, the ratio of water volume to submerged cable volume should also be considered. These parameters will
have significant effects on the temperature of the water if the water temperature is not controlled. However, if
the water temperature is controlled, less attention to tank dimensions and structure given in Table B.1, Table B.2,
and Table B.3 can be tolerated.

Table B.1 lists nominal size and shape parameters currently in use in North America.

Table B.2 lists the commonly used dimensions of tanks and materials to be used according to their shapes
(Figure 1, Figure 2, and Figure 3).

Materials of construction. In general, the inner tank surface, the specimen mounts, and any other materials
making contact with the water in the tanks should be constructed from suitable inert materials that will not
corrode in the presence of water and will not form an electrolytic cell with any component in the cable
specimen.
The type of material contacting the water should be reported. Also, any materials used to insulate the tanks
or used to insulate or retard surface evaporation of the water, and their thicknesses, should be reported to
allow determination of their impact on water temperature and quality.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Typical materials of construction in current use are shown in Table B.3.

Table B.1Nominal tank sizes

Typical tank shapes and sizes (metric units)

ayout tank type


1 1 1 2 2 2 2 2 1 3

Tank dimensions (mm)


Length 1500 1650 3000 1500 1550 1650 2100 2100 3250 1220
a
Width 1220 1350 1220 1220 890 1120 1070 1220 1220 1220
a
Height 1050 1500 1220 980 910 980 1220 1220 1220 1220

Specimen to tank spacing (mm)


Side (HS) 300 25 230 75 50 65 75 25 305 40

End (TS) 125 75 220, 38 65 50 75 130 100


430 b
Bottom (VS) 150 130 50 230 140 130 75 25 305 0

Specimen to specimen spacing (mm)


(SS1) 130 59 220 115 115 140 140 130 145 0

(SS2) 340 340 420 420 380

Water depth 900 1170 1170 915 810 915 1140 915 1120 1220

Typical tank shapes and sizes (English units)


ayout tank type 1 1 1 2 2 2 2 2 1 3

Tank dimensions (in)


Length 60 65 120 60 61 65 84 84 125 48 a
Width 48 53 48 48 36 44 42 48 48 48 a
Height 42 60 48 39 36 39 48 48 48 48

Specimen to tank spacing (in)


Side (HS) 12 1.0 9.0 3.0 2.0 2.5 3.0 1.0 12 1.5

End (TS) 5.0 3.0 8.5, 1.5 2.5 2.0 3.0 5.0 3.9
17 b
Bottom (VS) 6.0 5.0 2.0 9.0 5.5 5.0 3.0 1.0 12 0

Specimen to specimen spacing (in)


(SS1) 5 2.5 8.5 4.5 4.5 5.5 5.5 5 5.7 0

(SS2) 13.5 13.5 16 16 15

Water depth 36 46 46 36 32 36 45 36 44 48

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
a
Tank shape reported as cylindrical, length and width = diameter.
b Tank to specimen spacing reported as different for each end.

Table B.2Commonly used materials and tank dimensions

Tank details Type 1 Type 2 Type 3

Length (mm) [in] 1140 [ 45] 1500 [ 60] 840 [ 33]


Width (mm) [in] 800 [ 30] 800 [ 30]

Height (mm) [in] 910 [ 36] 910 [ 36] 910 [ 36]

Side spacing (HS) (mm) [in] 50 [ 2] 50 [ 2] 50 [ 2]

End spacing (TS) (mm) [in] 75 [ 3] 50 [ 2]

Bottom spacing (VS) (mm) [in] 50 [ 2] 50 [ 2] 0


Specimen spacing 1 (SS1) (mm) [in] 50 [ 2] 50 [ 2] 0

Specimen spacing 2 (SS2) (mm) [in] 75 [ 3]

Water depth (mm) [in] 800 [ 30] 800 [ 30] 800 [ 30]

Tank material SS or PL a SS or PL a SS or PL a
Tank insulation Water resistant Water resistant Water resistant

Insulation thickness (IT) (mm) [in] 25 [ 1.0] 25 [ 1.0] 25 [ 1.0]

Specimen mount Polymeric Polymeric Polymeric

Surface cover Polymeric Polymeric Polymeric


a SS = stainless steel; PL = polymer or polymer lined.

Table B.3Commonly used construction materials for tanks

Inner tank surface Tank insulation


Specimen Surface cover
Thickness Thickness mounts
Material Material
(mm) [in] (mm) [in]
1 Stainless steel 2.2 [0.087] Foam 25 [1.0] Acrylic sheets Polymer
spheres
2 Stainless steel 0.8 [0.03] Fiberglass 75 [3.0] PVC pipe Acrylic sheets
racks

3 Vinyl liner Concrete 150 [6] Fiberglass Expanded


structure with conc. sides racks polystyrene
sand on bottom and bottom
and foam on 150 [6] sand
sides bottom only

4 Stainless steel 3.2 [0.125] Redwood 41 [1.63] Acrylic sheets Polymer


spheres
5 Stainless steel 3.2 [0.125] Redwood 41 [1.63] Acrylic sheets Polymer
spheres

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
6 Stainless steel 1.6 [0.06] Redwood 41 [1.63] Acrylic sheets Polymer
spheres
7 Stainless steel 1.6 [0.06] Redwood 41 [1.63] Acrylic sheets Polymer
spheres
8 Stainless steel 5 [0.19] Foam 51 [2.0] Polypropylene Polycarbonate

9 Polyethylene 8 [0.32] Fiberglass 38 [1.5] None Expanded


polystyrene

10 Polyethylene Fiberglass None Polymer


spheres
11 Galvanized 1 [0.04] Foam 38 [1.5] Fiberglass Polymer
steel with racks spheres
vinyl liner

Annex C
(informative)

Temperature measurement
The techniques given in C.1 and C.2 for preparation and mounting of T/Cs are a compilation of those used at
several laboratories throughout North America.

C.1 Conductor temperature measurements using thermocouples

a) Lab A. To measure the temperature of the conductor of the dummy specimen, commercially available SS-
sheathed, Type T or Type K T/Cs are used. The sheaths have a 0.5 mm (20 mil) outside diameter and are
electrically isolated from the T/C junction located at the end of the SS sheath (ungrounded).
To attach the T/C to the conductor, a hole is drilled perpendicular to the cable axis down to the conductor. A
specimen piece of cable is often used to gauge the correct drilling depth. The drill bit diameter is approximately
0.53 mm (21 mil). The end of the T/C sheath is coated with a high- temperature heat sink compound, which is
commonly used as a heat sink compound for solid state electronics.
The T/C sheath is then placed into the hole down to the conductor. A small bend is put in the sheath such that the
sheath rises about 2.5 mm (0.10 in) above the cable surface at the drilled hole. It immediately bends back down
to the cable surface. Silicone tape is wrapped around the cable over this bend, placing a constant inward force on
the sheath, keeping it in good contact with the conductor. The silicone tape should be applied so that the T/C is
held securely in place but not so tight that the cable deforms during load cycling.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
b) Lab B. Cost effective unsheathed T/Cs can be fabricated from Type J (note that unsheathed Type J can
corrode in water) or Type T T/C wire. T/C wires with 20 AWG to 24 AWG solid conductors seem to be well-
suited for the fabrication of T/C junctions, which can be used to measure conductor temperature. The tips of the
T/C wires are usually joined together by a specially built T/C welder that utilizes a capacitive-discharge
technique. To locate the T/C junction next to the cable conductor, a 6.3 mm (0.25 in) diameter plug of insulation
shield + insulation + conductor shield material is removed using a sharp coring tool. The T/C wires are bent at a
90 angle near the junction so that the junction can be placed flat against the conductor surface that was exposed
by removing the plug. The same plug of insulation shield + insulation + conductor shield material is then
replaced and held in place by wrapping self-amalgamating tape around the power cable. A multi-purpose room-
temperature vulcanized (RTV) sealant may also be used to seal any gaps around the top of the plug prior to the
application of the tape. To insure that the T/C junction is making contact with the conductor, an ohmmeter is
used to check continuity between each T/C wire and the cable conductor.
Several T/Cs are mounted on the conductor surface in close proximity to the measurement area as a means of
determining when a particular T/C is no longer functional. It is also important to remember that the conductor
temperature will not always be the same along its length. To establish accurate temperature profiles, several T/Cs
should be installed along the length of the conductor.
c) Lab C. Prepare the T/C junctions by welding the tips of the T/C wire (usually Type J). Smaller T/C wires
at the point of temperature measurement will disturb the measurement area the least. Being careful not to drill
into the conductor, drill four small holes into the insulation separated by 90 along the circumference of the
cable. The T/C junctions are placed into these small holes. The T/C junctions should be firmly seated against the
conductor. On stranded conductors, two adjacent outer wires can be slightly separated so that the T/C junction
can be placed between the wires. Once the T/Cs are in place, machined plugs of polyethylene with the same
diameter as the drilled holes are placed over the T/Cs, and the plugs are then secured by taping them in place.
Another set of T/Cs located 60 cm to 90 cm (2 ft to 3 ft) away, and mounted in the same manner as the first set,
should also be monitored to insure the integrity of the temperature measurements.

C.2 Outside cable surface temperature measurements using thermocouples


a) Lab A. To measure the temperature of the insulation shield or jacket, a welded bead is made on a Type T or
a Type K T/C. The bead is then soldered to a small square of copper tape. The copper tape is typically 13 mm
13 mm 0.13 mm (0.5 in 0.5 in 0.005 in) thick. Heat sink compound is applied to the side of the copper tape
opposite the T/C and placed against the insulation shield or the jacket. It is then held in place with silicone tape.
The silicone tape should be applied so that the T/C is held securely in place but not so tight that the cable
deforms during load cycling.
b) Lab B. To measure the temperature of the insulation shield or jacket, commercially available SS- sheathed
Type T or Type J T/Cs are used. The sheaths are electrically isolated from the T/C junction located at the end of
the SS sheath (ungrounded). These SS-sheathed T/Cs are secured to the insulation shield or jacket with self-
amalgamating tape.
c) Lab C. To measure the temperature of the insulation shield, a welded bead is made on a Type T or a Type J
T/C. The bead is then soldered to the center of 25 mm 25 mm (1.0 in 1.0 in) square of copper tape. The
copper square is then laid flat on the insulation shield and secured with tape. To measure the insulation shield
surface temperature on a jacketed cable, a window is first cut in the jacket by folding back a small flap of the
jacket material. The copper tape square with the T/C is then laid flat on the insulation shield surface and the flap
of jacket material is replaced and secured with tape.
Several of these T/Cs are usually secured around the circumference of the cable in at least three different
locations along the length of the cable.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Annex D

(informative)

Final report
The following information should be included in the test report.

D.1 Test specimens

Table 1 of this guide gives information on the test specimens, cable construction, extrusion conditions, num- ber
of specimens to be tested, test specimen geometry, etc. The items in Table 1 should be included in the report, if
they are available.

D.2 Prior tests and preconditioning

The report should contain the results of cable quality control tests and the details of the preconditioning tests.
Any abnormalities, as described in Clause 14, should be included.

D.3 Test structure

A schematic of the tank layout should be given (see relevant clause of guide for examples). The details of the
tank should be tabulated. A typical example is shown as follows.

Tank details

Length (mm)
Width (mm)
Height (mm)

Side spacing (mm)

End spacing (mm)

Bottom spacing (mm)

Specimen spacing 1 (mm)

Specimen spacing 2 (mm)

Water depth (mm)

Tank material

Tank insulation
Insulation thickness (mm)

Specimen mount

Surface cover

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
D.4 Temperature
The report should contain a Figure indicating where the temperatures are measured in the test specimen and the
water and also how it is controlled. A Figure of the temperature Profile showing the variation of tempera- ture
with time for startup and at least one stable temperature cycle should also be included. The maximum and
minimum temperatures of the test specimen and the water should be tabulated, including tolerances. It is
important to list any abnormalities, as described in Clause 14.

D.5 Water
The water characteristics, in the tank or in the cable conductors, and any properties measured (e.g., resistively,
pH, ion content) should be included. The frequency of monitoring the water (e.g., weekly, monthly, or quarterly)
in each tank, the water/cable volume ratio and the tolerance on the water level, the water characteristics used to
maintain the levels in the tank and cable conductors, and the use of water circulation in the tank should be stated.
Any abnormalities in the condition of the water, either in the tank or in the cable conductor, should be listed.

D.6 Voltage
The amplitude and frequency of the voltage should be included along with the THD. Voltage interruptions, both
scheduled and unscheduled, should be reported. Details of transients, if measured, should also be included along
with any abnormalities, as described in Clause 14.

D.7 Results and data analysis


The data should be analyzed using the Weibull or lognormal distributions and presented either graphically or in
tabular form or both. The analysis should include the characteristic time to failure, the shape parameter, and the
90% Confidence limits. Data excluded from the analysis due to an abnormality such as termination failure, etc.,
should be mentioned.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Annex E

(informative)

Bibliography

[B1] Abdolall, K., Design of an Accelerated Aging Test for Polymer Insulated Distribution Cables,
Conference Record of 1990 IEEE International Symposium on Electrical Insulation, Toronto, Canada, 1990, pp.
343347.

[B2] Abernathy, R., The New Weibull Handbook, Dr. R. Abernathy, 536 Oyster Road, North Palm Beach, FL
33408-4328, 1993.

[B3] Ashcraft, A., Water Treeing in Polymer Dielectrics, Kabelitems, No. 152, 1979, pp. 511.

[B4] Banks, V. A. A., Faremo, H., and Steennis, E. F., An Accelerated Aging Test on the Basis of 500 Hz for
Water Treeing in Cables, Proceedings Fourth International Conference on Insulated Power Cables (Jicable),
June 2529, 1995, pp. 347356.

[B5] Bartnikas, R., Densley, R. J., and Eichhorn, R. M., Accelerated Aging Tests for Polymer Cables
UnderWet Conditions, IEEE Transactions on Power Delivery, Vol. PWD-6, pp. 929937, 1991.

[B6] Bartnikas, R., et al., Accelerated Life Testing of Wet Cable Specimens at Frequencies above 60 Hz,
IEEE Transactions on Power Apparatus and Systems, Vol. PAS-99, pp. 15751585, 1980.

[B7] Damon, D. H., Huang, S. J., and Johnson, J. F., Electric Strength of XLPE Cable Insulation Containing
Controlled Amounts of the Volatile Products of the Cross-linking Reaction, 1990 Annual Report, Conference
on Electrical Insulation and Dielectric Phenomenon, 1990, pp. 398404.

[B8] Densley, J., Aging and Diagnostics in Extruded Insulations for Power Cables, Fifth
InternationalConference on Conduction and Breakdown in Solid Dielectrics, 1995, pp. 115.

[B9] Eichhorn, R.M., Organic Residues in Polyethylene Insulation Cross-linked with Dicumyl Peroxide,
Conference Record 1972 IEEE Underground Transactions Conference, 1972, pp. 282285.

[B10] EPRI Report EL-5387,Groeger, J. H., et al., Surmounting Difficulties in the Application of Diagnostics
for Characterization of Failed Solid-Dielectric Cables: Phase 2, Sept. 1987. [B11] EPRI Report EL-5921,
Morphology of Extruded Dielectric Aged Polymeric Cable Insulation, 1988, 1990 CIGRE, Paper 15/33-09,
1990.

[B11] Hartlein, R., Harper, V. S., and Ng, H., Effects of Voltage Surges in Extruded Dielectric Cable Life
Project Update, IEEE Transactions on Power Delivery, Vol. PWD-8, No. 2, 1994, pp. 611619.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
[B12] Hvidsten, S., et al., Condition Assessment of Water Treed Service Aged XLPE Cables by Dielectric
Response Measurements, 2000 CIGRE, Paper 21-201, 2000.

[B12] EPRI Report EL-7076 (RP 7897-01), Evaluation of Sensitive Diagnostic Techniques for Cable Char-
acterization, Dec. 1990.

[B13] IEEE 100, The Authoritative Dictionary of IEEE Standards Terms, Seventh Edition. New York: Institute
of Electrical and Electronics Engineers, Inc.

[B13] EPRI Report TR-103908, Effects of the Volatile By-Products of the Cross-linking Reaction on the
Dielectric Strength of Cross-Linked Polyethylene Cables, July

[B14] EPRI Report TR-108405-V2 Aging of Distribution Cables in Controlled Temperature Tank Tests, Aug.
1997Katz, C., et al., Emergency Overload Characteristics of Extruded Dielectric Cable Operating and130 C
and Above, IEEE Transactions on Power Apparatus and Systems, Vol. PAS-103, No. 12, pp. 3454-3463,
December 1984.

[B15] Katz, C., Seman, G. W., and Bernstein, B. S., Low Temperature Aging of XLPE and EP Insulated
Cables with Voltage Transients, IEEE Transactions on Power Delivery, Vol. PWD-10, No. 1, 1995, pp.34
42.

[B16] Kuschel, Plath, R., Stepputat, I., and Kalkner, W., M.,et al., Diagnostic Techniques for Service-Aged
XLPE-Insulated Medium Voltage Cables, Fourth International Conference on Insulated Power Cables,
(Jicable),25-29 June, 1995, pp.504508, 504508; also published in REE, Special Issue, Vol.1, Aug.1996,
pp.6672.

[B17] Luzzi, G. J., Gas Pressure Build-up in Aluminum Conductor Cable, IEEE Transactions on Power
Appartus and Systems, Vol. 101, No. 6, 1982, pp. 16971699.

[B18] Lyle, R., Effect of Testing Parameters on the Outcome of Accelerated Cable Life Test, IEEE
Transactions on Power Delivery, Vol. 3(2), 1988, pp. 434439.

[B19] Martin, M. A., and Hartlein, R. A., Correlation of Electrochemical Treeing in Power Cables Removed
from Service and in Cables Tested in the Laboratory, IEEE Transactions on Power Apparatus and Systems,
Vol. PAS-99, 1980, pp. 15971605.

[B20] Mashikian, M. S., et al., Water-Treeing Propensity Tests of Materials and Cables in the
UnitedStates, CIGRE Paper 15/21-02, 1992.

[B21] Nelson, W., Accelerated Testing. John Wiley,1990.

[B22] Plissou, S., Effect of water on the performance of XLPE insulation, Proceedings of the 3rd ICSD
Conference, 1989, pp. 4326.

[B23] Plissou, S., Relation Between the Halo and Water Trees in XLPE Cables, 1988 Annual Report,
Conference on Electrical Insulation and Dielectric Phenomenon, pp. 101-108, 1988.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
[B24] Phillips, P. J., and Vatansever, A., Crystallinity in Chemically Crosslinked Low Density
Polyethylene, Polymer, Vol. 30, p. 710, 1989.

[B25] Schroth, R. G., Kalkner, W., and Fredrich, D., Test Methods for Evaluating the Water Tree Behaviour
of Extruded Cable Insulations, CIGRE Paper 15/2101, 1990, 11pp.

[B26] Srinivas, N. N., Bernstein, B. S., and Decker, R. A., Effects of DC Testing on AC Breakdown Strength of
XLPE Insulated Cables Subjected to Laboratory Accelerated Aging, IEEE Transactions on Power Delivery,
Vol. PWD-5, No. 4, 1990, pp. 16431651.

[B27] Steennis, E. F., and Faremo, H., State of the Art of Water Tree Testing on CablesThe Development of
an Accelerated Aging Test, 1992 CIGRE, Paper 15/2103, 1992

[B27] Tharning, P., et al., High Voltage Dielectric Response Analyser for Cable Diagnostics, 1993 Ann. Rpt.
Conf. on Elect. Insul. and Diel. Phenom., 1993, pp. 398404.

[B28] Szaniszlo, S., Breakdown Characteristics of TRXLPE, Minutes of 95th meeting of IEEE Insulated
Conductors Committee (ICC), Spring 1994, Reno, NV, pp. V-6, and V-C-1-V-C-4.

[B28] UNIPEDE-DISCAB, International Cable Fault Statistics on Synthetically Insulated MV Cables,


CIRED Conf., Vol. 1, 1989, pp. 200206.

[B29] Thrning, P., et al., High Voltage Dielectric Response Analyser for Cable Diagnostics, 1993
Annual Report, Conference on Electrical Insulation and Dielectric Phenomenon, 1993, pp. 745752.

[B30] UNIPEDE-DISCAB, International Cable Fault Statistics on Synthetically Insulated MV Cables,


CIRED Conference, Vol. 1, 1989, pp. 200206.

[B31] Walton, M. D. Aging of Distribution Cables in Controlled Temperature Tank Tests, EPRI Technical
Report TR 108405-V2, August 1997.

[B32] Walton, M. D., Smith, J. T., and Thue, W. A., Accelerated Aging of Extruded Dielectric
PowerCables, Part 1: Control and Monitoring Methodology, IEEE Transactions on Power Delivery, Vol.
7(2),1992, pp.596 602.

[B33] Wartusch, J., and Wagner, H., About the Significance of Peroxide Decomposition Products in XLPE
Cable Insulation, Proceedings, 1976 IEEE International Symposium on Electrical Insulation, Montreal, Canada,
1976, pp. 219223.

[B34] Yan, S., et al., Electric Strength of XLPE Containing Acetophenone and Other Volatile Substances,
Conference Record, 1990 IEEE International Symposium on Electrical Insulation, pp. 305-308, 1990.

Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.
Authorized licensed use limited to: Chulalongkorn University provided by UniNet. Downloaded on August 22,2017 at 11:53:27 UTC from IEEE Xplore. Restrictions apply.