IEEE Guide for Sunchronizations, Calibration, Testing, And Installation of PMUs for Power System Protection and Control
Content
IEEE Guide for Synchronization,
Calibration, Testing, and Installation of
Phasor Measurement Units (PMUs) for
Power System Protection and Control
Sponsored by the
Power System Relaying Committee
IEEE
3 Park Avenue
New York, NY 10016-5997
USA
6 March 2013
IEEE Power and Energy Society
IEEE Std C37.242™-2013
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242™-2013
IEEE Guide for Synchronization,
Calibration, Testing, and Installation of
Phasor Measurement Units (PMUs) for
Power System Protection and Control
Sponsor
Power System Relaying Committee
of the
IEEE Power and Energy Society
Approved 6 February 2013
IEEE-SA Standards Board
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Abstract: Guidance for synchronization, calibration, testing, and installation of phasor
measurement units (PMUs) applied in power systems is provided. The following are addressed in
this guide: (a) Considerations for the installation of PMU devices based on application
requirements and typical substation electrical bus configurations; (b) Techniques focusing on the
overall accuracy and availability of the time synchronization system; (c) Test and calibration
procedures for PMUs for laboratory and field applications; (d) Communications testing for
connecting PMUs to other devices including Phasor Data Concentrators (PDCs).
Keywords: calibration, GPS, IEEE C37.242™, PMU, synchrophasor, testing
•
The Institute of Electrical and Electronics Engineers, Inc.
3 Park Avenue, New York, NY 10016-5997, USA
Copyright © 2013 by The Institute of Electrical and Electronics Engineers, Inc.
All rights reserved. Published 6 March 2013. 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.
PDF: ISBN 978-0-7381-8295-7 STD98172
Print: ISBN 978-0-7381-8296-4 STDPD98172
IEEE prohibits discrimination, harassment, and bullying. For more information, visit http://www.ieee.org/web/aboutus/whatis/policies/p9-26.html.
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: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Notice and Disclaimer of Liability Concerning the Use of IEEE Documents: IEEE Standards documents are developed
within the IEEE Societies and the Standards Coordinating Committees of the IEEE Standards Association (IEEE-SA)
Standards Board. 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 IEEE administers the process
and establishes rules to promote fairness in the consensus development process, 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. 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 any IEEE Standard document.
IEEE does not warrant or represent the accuracy or content of the material contained in its standards, 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 in its standards 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 ten years. When a document is
more than ten years old and has not undergone a revision process, 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 its standards available, IEEE is not suggesting or rendering professional or other services for, or on
behalf of, any person or entity. Nor is IEEE undertaking to perform any duty owed by any other person or entity to another.
Any person utilizing any IEEE Standards document, should rely upon his or her own 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.
Translations: The IEEE consensus development process involves the review of documents in English only. In the event that
an IEEE standard is translated, only the English version published by IEEE should be considered the approved IEEE standard.
Official Statements: 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 position of 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 of IEEE.
Comments on Standards: Comments for revision of IEEE Standards documents are welcome from any interested party,
regardless of membership affiliation with IEEE. However, IEEE does not provide consulting information or advice pertaining
to IEEE Standards documents. Suggestions for changes in documents should be in the form of a proposed change of text,
together with appropriate supporting comments. Since IEEE standards represent a consensus of concerned interests, it is
important to ensure that any responses to comments and questions also receive 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 comments or questions except in those cases where the matter has previously been addressed. Any person who
would like to participate in evaluating comments or revisions to an IEEE standard is welcome to join the relevant IEEE
working group at http://standards.ieee.org/develop/wg/.
Comments on standards should be submitted to the following address:
Secretary, IEEE-SA Standards Board
445 Hoes Lane
Piscataway, NJ 08854
USA
Photocopies: 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: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Copyright © 2013 IEEE. All rights reserved.
iv
Notice to users
Laws and regulations
Users of IEEE Standards documents should consult all applicable laws and regulations. Compliance with
the provisions of any IEEE Standards document does not imply compliance to any applicable regulatory
requirements. Implementers of the standard are responsible for observing or referring to the applicable
regulatory requirements. IEEE does not, by the publication of its standards, intend to urge action that is not
in compliance with applicable laws, and these documents may not be construed as doing so.
Copyrights
This document is copyrighted by the IEEE. It is made available for a wide variety of both public and
private uses. These include both use, by reference, in laws and regulations, and use in private self-
regulation, standardization, and the promotion of engineering practices and methods. By making this
document available for use and adoption by public authorities and private users, the IEEE does not waive
any rights in copyright to this document.
Updating of IEEE documents
Users of IEEE Standards documents 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-SA Website at http://standards.ieee.org/index.html or
contact the IEEE at the address listed previously. For more information about the IEEE Standards
Association or the IEEE standards development process, visit IEEE-SA Website at
http://standards.ieee.org/index.html.
Errata
Errata, if any, for this and all other standards can be accessed at the following URL:
http://standards.ieee.org/findstds/errata/index.html. Users are encouraged to check this URL for errata
periodically.
Patents
Attention is called to the possibility that implementation of this standard may require use of subject matter
covered by patent rights. By publication of this standard, no position is taken by the IEEE with respect to
the existence or validity of any patent rights in connection therewith. If a patent holder or patent applicant
has filed a statement of assurance via an Accepted Letter of Assurance, then the statement is listed on the
IEEE-SA Website at http://standards.ieee.org/about/sasb/patcom/patents.html. Letters of Assurance may
indicate whether the Submitter is willing or unwilling to grant licenses under patent rights without
compensation or under reasonable rates, with reasonable terms and conditions that are demonstrably free of
any unfair discrimination to applicants desiring to obtain such licenses.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Copyright © 2013 IEEE. All rights reserved.
v
Essential Patent Claims may exist for which a Letter of Assurance has not been received. The IEEE is not
responsible for identifying Essential Patent Claims for which a license may be required, for conducting
inquiries into the legal validity or scope of Patents Claims, or determining whether any licensing terms or
conditions provided in connection with submission of a Letter of Assurance, if any, or in any licensing
agreements are reasonable or non-discriminatory. Users of this standard are expressly advised that
determination of the validity of any patent rights, and the risk of infringement of such rights, is entirely
their own responsibility. Further information may be obtained from the IEEE Standards Association.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Copyright © 2013 IEEE. All rights reserved.
vi
Participants
At the time this IEEE guide was completed, the C5 Working Group had the following membership:
Farnoosh Rahmatian, Chair
Paul Myrda, Vice Chair
Mark Adamiak
Galina Antonova
Alexander Apostolov
James Ariza
Bill Dickerson
Vasudev Gharpure
Allen Goldstein
James Hackett
Yi Hu
Mladen Kezunovic
Harold Kirkham
Vahid Madani
Kenneth Martin
A.P. (Sakis) Meliopoulos
Jay Murphy
Krish Narendra
Damir Novosel
Manu Parashar
Mahendra Patel
S. Richards
Veselin Skendzik
Gerard Stenbakken
A. Vaccaro
Benton Vandiver
The following members of the individual balloting committee voted on this guide. Balloters may have
voted for approval, disapproval, or abstention.
Mohamed Abdel Khalek
William Ackerman
Mark Adamiak
Satish Aggarwal
Ali Al Awazi
Mihaela Albu
Saleman Alibhay
Galina Antonova
James Ariza
David Bassett
Martin Baur
Philip Beaumont
Kenneth Behrendt
Robert Beresh
Richard Bingham
Gustavo Brunello
Paul Cardinal
Arvind K. Chaudhary
Stephen Conrad
Luis Coronado
Andrew Dettloff
Michael Dood
Gary Engmann
Dan Evans
Ronald Farquharson
Fredric Friend
Doaa Galal
John Galanos
Vasudev Gharpure
David Gilmer
Jalal Gohari
Allen Goldstein
Stephen Grier
Randall C. Groves
Erich Gunther
James Hackett
Donald Hall
Dennis Hansen
Roger Hedding
Werner Hoelzl
Yi Hu
Gerald Johnson
Innocent Kamwa
Yuri Khersonsky
Morteza Khodaie
Harold Kirkham
Joseph L. Koepfinger
Jim Kulchisky
Chung-Yiu Lam
Raluca Lascu
Greg Luri
Vahid Madani
Wayne Manges
Kenneth Martin
William McBride
John McDonald
William Moncrief
Jay Murphy
Jerry Murphy
Bruce Muschlitz
Michael S. Newman
Damir Novosel
James O’Brien
Lorraine Padden
Donald Parker
Bansi Patel
Craig Preuss
Iulian Profir
Farnoosh Rahmatian
Reynaldo Ramos
Michael Roberts
Charles Rogers
Thomas Rozek
Sergio Santos
Bartien Sayogo
Thomas Schossig
Devki Sharma
Gil Shultz
Veselin Skendzic
James Smith
Jerry Smith
Aaron Snyder
John Spare
Gerard Stenbakken
Gary Stoedter
Charles Sufana
Richard Taylor
William Taylor
John Tengdin
Maria Tomica
Eric Udren
John Vergis
Jane Verner
Quintin Verzosa
John Wang
Solveig Ward
Karl Weber
Philip Winston
Jian Yu
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Copyright © 2013 IEEE. All rights reserved.
vii
When the IEEE-SA Standards Board approved this guide on 6 February 2013, it had the following
membership:
John Kulick, Chair
Richard H. Hulett, Past Chair
Konstantinos Karachalios, Secretary
Masayuki Ariyoshi
Peter Balma
Farooq Bari
Ted Burse
Wael William Diab
Stephen Dukes
Jean-Philippe Faure
Alexander Gelman
Mark Halpin
Gary Hoffman
Paul Houzé
Jim Hughes
Michael Janezic
Joseph L. Koepfinger*
David J. Law
Oleg Logvinov
Ron Petersen
Gary Robinson
Jon Walter Rosdahl
Adrian Stephens
Peter Sutherland
Yatin Trivedi
Phil Winston
Yu Yuan
*Member Emeritus
Also included are the following nonvoting IEEE-SA Standards Board liaisons:
Richard DeBlasio, DOE Representative
Michael Janezic, NIST Representative
Michelle Turner
IEEE Standards Program Manager, Document Development
Matthew J. Ceglia
IEEE Standards Program Manager, Technical Program Development
Soo H. Kim
Client Service Manager, Professional Services
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Copyright © 2013 IEEE. All rights reserved.
viii
Introduction
This introduction is not part of IEEE Std C37.242-2013, IEEE Guide for Synchronization, Calibration, Testing, and
Installation of Phasor Measurement Units (PMUs) for Power System Protection and Control.
Use of synchrophasor technology in the electric power industry is rapidly growing, moving from research
and pilot projects into system-wide production level deployment. Accordingly, a practical guide for
installing and testing phasor measurement units (PMUs) is expected to be very beneficial to field
practitioners, sharing and leveraging the early experience that the pioneers in this area have accumulated.
This document was developed by IEEE PES Power System Relaying Committee to guide and educate
various professionals interested in deploying PMUs and using the associated synchrophasor data.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Copyright © 2013 IEEE. All rights reserved.
ix
Contents
1. Overview.................................................................................................................................................... 1
1.1 Scope ................................................................................................................................................... 1
1.2 Purpose ................................................................................................................................................ 2
2. Normative references.................................................................................................................................. 2
3. Definitions, special terms, acronyms, and abbreviations............................................................................ 2
3.1 Definitions ........................................................................................................................................... 3
3.2 Special terms........................................................................................................................................ 3
3.3 Acronyms and abbreviations ............................................................................................................... 3
4. Synchronization techniques, accuracy, and availability ............................................................................. 4
4.1 Introduction ......................................................................................................................................... 4
4.2 Role of time synchronization in PMUs................................................................................................ 4
4.3 Satellite-based synchronizing sources ................................................................................................. 5
4.4 Terrestrial systems............................................................................................................................... 9
4.5 Synchronization distribution methods ............................................................................................... 10
5. Synchrophasor measurement accuracy characterization .......................................................................... 10
5.1 Introduction ....................................................................................................................................... 10
5.2 Data accuracy characterization.......................................................................................................... 11
5.3 Data accuracy .................................................................................................................................... 12
5.4 Characterization of instrumentation channels.................................................................................... 13
5.5 Characterization of GPS-synchronized measurement devices (PMUs)............................................. 14
5.6 GPS-synchronized equipment reliability........................................................................................... 14
6. PMU installation, commissioning, and maintenance................................................................................ 15
6.1 Preface ............................................................................................................................................... 15
6.2 Overview........................................................................................................................................... 15
6.3 Pre-installation procedures ................................................................................................................ 16
6.4 Analog and digital input .................................................................................................................... 20
6.5 Power input........................................................................................................................................ 21
6.6 Communications................................................................................................................................ 21
6.7 Summary of design considerations.................................................................................................... 24
6.8 Pre-installation tests........................................................................................................................... 25
6.9 Verification of end-to-end calibration ............................................................................................... 25
6.10 Communications operation.............................................................................................................. 28
6.11 Record keeping................................................................................................................................ 29
7. Testing and calibration ............................................................................................................................. 29
7.1 Overview........................................................................................................................................... 29
7.2 Objective of testing............................................................................................................................ 29
7.3 Types of tests..................................................................................................................................... 30
7.4 Test equipment .................................................................................................................................. 34
7.5 Methods for performing the tests....................................................................................................... 40
7.6 Synchrophasor message format ......................................................................................................... 54
7.7 Final comments.................................................................................................................................. 55
Annex A (informative) Bibliography ........................................................................................................... 56
Annex B (informative) Responses of the reference signal processing model to test signals........................ 60
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Copyright © 2013 IEEE. All rights reserved.
x
Annex C (informative) Effects of signal channels........................................................................................ 80
Annex D (informative) Examples of instrumentation channel impact on accuracy
using approximate models............................................................................................................................ 85
Annex E (informative) Example of commissioning tests and measurements............................................... 95
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
Copyright © 2013 IEEE. All rights reserved.
1
IEEE Guide for Synchronization,
Calibration, Testing, and Installation of
Phasor Measurement Units (PMUs) for
Power System Protection and Control
IMPORTANT NOTICE: IEEE Standards documents are not intended to ensure safety, health, or
environmental protection, or ensure against interference with or from other devices or networks.
Implementers of IEEE Standards documents are responsible for determining and complying with all
appropriate safety, security, environmental, health, and interference protection practices and all
applicable laws and regulations.
This IEEE document is made available for use subject to important notices and legal disclaimers.
These notices and disclaimers appear in all publications containing this document and may
be found under the heading “Important Notice” or “Important Notices and Disclaimers
Concerning IEEE Documents.” They can also be obtained on request from IEEE or viewed at
http://standards.ieee.org/IPR/disclaimers.html.
1. Overview
1.1 Scope
The document provides guidance for synchronization, calibration, testing, and installation of phasor
measurement units (PMUs) applied in power system protection and control. The following are addressed in
this guide:
a) Considerations for the installation of PMU devices based on application requirements and typical
bus configurations.
b) Techniques focusing on the overall accuracy and availability of the time synchronization system.
c) Test and calibration procedures for PMUs for laboratory and field applications.
d) Communications testing for connecting PMUs to other devices including Phasor Data Concentrators
(PDCs).
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
2
1.2 Purpose
This guide is intended to be used by power system protection professionals for PMU installation and covers
the requirements for synchronization of field devices and connection to other devices including PDCs.
2. Normative references
The following referenced documents are indispensable for the application of this document (i.e., they must
be understood and used, so each referenced document is cited in text and its relationship to this document is
explained). For dated references, only the edition cited applies. For undated references, the latest edition of
the referenced document (including any amendments or corrigenda) applies.
IEC 61850-3, Communication Networks and Systems in Substations—Part 3: General Requirements.
1
IEC/TR 61850-90-5, Ed. 1.0, Communication networks and systems for power utility automation—
Part 90-5: Use of IEC 61850 to transmit synchrophasor information according to IEEE C37.118.
IEEE Std 1588™-2008, IEEE Standard for a Precision Clock Synchronization Protocol for Networked
Measurement and Control Systems.
2,
3
IEEE Std C37.118™-2005, IEEE Standard for Synchrophasors for Power Systems.
IEEE Std C37.118.1™-2011, IEEE Standard for Synchrophasor Measurements for Power Systems.
IEEE Std C37.118.2™-2011, IEEE Standard for Synchrophasor Data Transfer for Power Systems.
IEEE Std C37.233™-2009, IEEE Guide for Power System Protection Testing.
IEEE Std C37.238™-2011, IEEE Standard Profile for Use of IEEE 1588™ Precision Time Protocol in
Power System Applications.
IEEE Std C57.13™, IEEE Standard Requirements for Instrument Transformers.
IRIG Standard 200-04 (2004), IRIG Serial Time Code Formats, Telecommunications and Timing Group,
Range Commanders Council, U.S. Army White Sands Missile Range, NM, USA.
4
3. Definitions, special terms, acronyms, and abbreviations
These definitions, acronyms, and abbreviations are especially pertinent to GPS-synchronized devices,
communications protocols, and communications media.
1
IEC publications are available from the International Electrotechnical Commission (http://www.iec.ch/). IEC publications are also
available in the United States from the American National Standards Institute (http://www.ansi.org/).
2
IEEE publications are available from The Institute of Electrical and Electronics Engineers (http://standards.ieee.org/).
3
The IEEE standards or products referred to in this clause are trademarks of The Institute of Electrical and Electronics Engineers, Inc.
4
This document available for download at http://www.irigb.com/pdf/wp-irig-200-04.pdf.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
3
3.1 Definitions
For the purposes of this document, the following terms and definitions apply. The IEEE Standards
Dictionary Online should be consulted for terms not defined in this clause.
5
Global Positioning System (GPS): A satellite-based system for providing position and time. The accuracy
of GPS-based clocks can be better than 1 μs.
Intelligent Electronic Device (IED): A general term indicating a multipurpose electronic device typically
associated with substation control and protection.
Phasor Data Concentrator (PDC): A function that collects phasor data and discrete event data from
PMUs and possibly from other PDCs, and transmits data to other applications. PDCs may buffer data for a
short time period but do not store the data.
phasor measurement unit (PMU): A device or a function in a multifuncation device that produces
synchronized phasor, frequency, and rate of change of frequency (ROCOF) estimates from voltage and/or
current signals and a time synchronizing signal.
virtual private network (VPN): VPN is an end-to-end communications method that employs encryption
and key exchange as the security mechanism. VPNs can be established over either public or private
networks.
Wide Area Measurement System (WAMS): One or more networks of measuring devices that may
include phasor measurement unit (PMUs), local recorders, legacy equipment, or advanced technologies that
are Global Positioning System (GPS)-synchronized over a geographically diverse area.
3.2 Special terms
error: In this guide, error is the difference between the result of a measurement and the value of what is
generally called the true value (of the measurand).
uncertainty: A parameter, associated with the result of a measurement, that characterizes the dispersion of
the values that could reasonably be attributed to the measurand. Uncertainty of measurement is
conventionally divided into two components. Type A uncertainty may be evaluated from the statistical
distribution of the results of a series of measurements. Statistical processing can be used to reduce the value
of this kind of uncertainty. Type B uncertainty can be evaluated from assumed probability distributions
based on knowledge from other than that from the measurements, such as the manufacturer’s
specifications, reference data from handbooks or calibration certificates.
3.3 Acronyms and abbreviations
CT current transformer
CVT capacitive voltage transformer
DFR digital fault recorder
DUT device under test
5
IEEE Standards Dictionary Online subscription is available at:
http://www.ieee.org/portal/innovate/products/standard/standards_dictionary.html.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
4
FPS frames per second
GNSS Global navigation satellite system
PMU phasor measurement unit
PT potential transformer
ROCOF rate of change of frequency
TVE total vector error
VT voltage transformer
Refer to IEEE Std C37.118.1-2011 and IEEE Std C37.118.2-2011 for additional acronyms and
abbreviations.
6
4. Synchronization techniques, accuracy, and availability
4.1 Introduction
This clause reviews the main technologies available to synchronize geographically distributed PMUs, and
the basic principles of clock synchronization and its impact on the phasor measurement accuracy. This
clause also examines common synchronization sources for time referencing, including both satellite (e.g.,
GPS) and terrestrial [e.g., Precision Time Protocol (PTP)] based technologies. The main advantages (e.g.,
timing accuracy) and potential vulnerabilities (e.g., susceptibility to intentional and unintentional
interference) of these techniques are also reported.
This clause also presents testing procedures (e.g., periodic timing signals measurement, measurement of
two consecutive timing signals) aimed at assessing the main performance characteristics of synchronization
sources (e.g., short-term stability, “bad” data management, hand-off algorithm).
Lastly, the synchronization distribution infrastructure is also discussed within this clause.
4.2 Role of time synchronization in PMUs
The clocks used for time synchronization in PMUs are required to be very accurate. However, their
accuracy may vary over time due to manufacturing defects, changes in temperature, electric and magnetic
interference, oscillator age, and altitude. Additionally, even small errors in timekeeping can add up
significantly over a long period.
Careful work and analysis is required to describe and quantify performance of timing in PMU systems, to
meet the accuracy requirements of IEEE Std C37.118.1-2011. Some clock variations are random, caused by
environmental or electronic variations; others are systematic, caused by a miscalibrated or misconfigured
clock.
Correct operation of a PMU requires a common and accurate timing reference. The timing reference is
described in IEEE Std C37.118.1-2011, which establishes the relationship between the Coordinated
6
Information on references can be found in Clause 2.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
5
Universal Time (UTC) time scale and the phase of the reference cosine wave. The required performance
could be realized by either synchronizing the samples directly to the timing reference or by software-based
post processing of the acquired samples. To achieve a common timing reference for the PMU acquisition
process, it is essential to have a source of accurate timing signals (i.e., synchronizing source) that may be
internal or external to the PMU. In the first case the synchronization source is integrated (built-in) into the
PMU (external GPS antenna still required). In the latter case, the timing signal is provided to the PMU by
means of an external source, which may be local or global, and a distribution infrastructure (based on
broadcast or direct connections).
The timing signal generated by the synchronizing source must be referenced to UTC and provide enough
information to determine that the time is in agreement with UTC. The synchronization signal must also be
available without interruption at all measurement locations throughout the interconnected grid. The timing
signal should be characterized by the availability, reliability, and accuracy suitable for power system
requirements.
The timing signal should be accurate enough to allow the PMUs to maintain synchronism with an accuracy
sufficient to keep the total vector error (TVE) within the limits defined in IEEE Std C37.118.1-2011 (see
4.3 of that standard for discussion).
The PMU is required to detect a loss of time synchronization that would cause the TVE to exceed the
allowable limit, or within 1 min of an actual loss of synchronization, whichever is less
(IEEE Std C37.118.2-2011, 4.5). In this case a flag in the PMU data output (STAT word Bit 13) should be
asserted until the data acquisition is resynchronized to the required accuracy level.
In addition to the STAT word Bit 13, IEEE Std C37.118.1-2011 specifies further signals intended to
describe the time quality of the synchronization source. Each of the PMU output messages defined
(Configurations 1, 2, and 3, Header, and Data) have a time quality field of 4 bits. This field allows the PMU
to state the quality of the time source from clock locked, 1 ns to 10 s uncertainty (estimated worst-case
error), or clock failure. Also, the Data message STAT has two bits to indicate the length of time the clock
has been unlocked. This varies from locked to unlocked for more than 10 s, 100 s, or more than 1000 s.
Even though a clock may be unlocked for over 1000 s, a quality oscillator is able to maintain better that
1 μs accuracy over this period. Consult the clock manufacturer's documentation for its drift specification.
IEEE Std C37.118.2-2011 adds a 3-bit PMU Time quality field to the status word in place of a previously
unused security bit field. When used, this field indicates the uncertainty in the measurement time at the
time of measurement and indicates time quality at all times, both when locked and unlocked, and unknown
when the clock is starting up.
4.3 Satellite-based synchronizing sources
This subclause summarizes the main technologies that could be adopted for PMU synchronization.
The following figures of merit have been considered in assessing the performances of the synchronization
source technologies (Lilley, Church, Harrison) [B37])
7
:
⎯ Accuracy: The degree of conformance between the measured synchronization signal and its true
value.
⎯ Availability: The capability of the synchronization system to provide usable timing services within
the specified coverage area.
⎯ Continuity: The probability that the synchronization system will be available for the duration of a
phase of operation, presuming that the system was available at the beginning of that phase of
operation. The factors that affect availability also affect continuity.
7
The numbers in brackets correspond to those of the bibliography in Annex A.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
6
⎯ Reliability: The probability that a synchronization system will perform its function within defined
performance limits for a specified period of time under given operating conditions. It is a function
of the frequency with which failures occur within the synchronization system.
⎯ Integrity: The ability of the synchronization system to detect the timing signals’ degradation and
provide timely warnings to users.
⎯ Coverage: The geographical area in which the application-specific synchronization system
requirements for accuracy, availability, continuity, reliability, integrity, and coverage parameters
are satisfied at the same time. System geometry, signal power levels, receiver sensitivity,
atmospheric noise conditions, and other factors that affect signal availability influence coverage.
⎯ Ride-through capability: The clock accuracy that is maintained and for how long upon loss of
satellite synchronization
4.3.1 Satellite navigation systems
The carrier signals transmitted by Global Navigation Satellite Systems (GNSS) disseminate precise time,
time intervals, and frequency over wide geographic areas. GNSS is a generic term. It may be noted that not
all systems are equivalent in satellite orbits or time accuracy.
Satellite-based timing signals are particularly suitable for PMU applications, since they make possible
accurate synchronization without requiring the PMU user to deploy the user’s own primary time and time
dissemination systems. At the same time, GNSS systems provide intrinsic advantages such as wide area
coverage, easy access to remote sites, and adaptability to changing network patterns...
Global Positioning System (GPS) is a U.S. Department of Defense satellite-based radio navigation system.
It consists of 24 satellites arrayed to provide a minimum worldwide visibility of four satellites at all times.
GPS derives its timing from a ground-based clock ensemble that itself is referenced to UTC. Each satellite
provides a correction to UTC time that the receiver automatically applies to the outputs. The GPS satellites
broadcast on two carrier frequencies: the L1 at 1575.42 MHz, and the L2 at 1227.60 MHz. Each satellite
broadcasts a spread-spectrum waveform called a pseudorandom noise (PRN) code on L1 and L2, and each
satellite is identified by the PRN code it transmits (Lombardi et al. [B38]). At this time, GPS is the most
common source of time synchronization for synchrophasor systems.
Timing accuracy is limited by short-term signal reception whose basic accuracy is 0.2 μs. This accuracy
can be improved by advanced decoding and processing techniques, giving actual performance orders of
magnitude better than required for PMU application. The inherent availability, redundancy, reliability, and
accuracy make it a system well suited for synchronized phasor measurement systems (Holbert, Heydt, Ni
[B25] and IEEE Std C37.118.1-2011).
The Russian Global Navigation Satellite System (GLONASS) provides similar capabilities to GPS.
Sporadic funding, and the resulting inconsistent satellite coverage, have hampered widespread acceptance
of the GLONASS system, although it is in some ways superior to GPS with respect to accuracy (Dickerson
[B13]).
These systems, and others to come on line in the future, provide timing accuracy that easily exceeds the
needs of the power industry. Future development in receiver technology is expected to provide the ability to
receive signals from two or more GNSS systems, though existing receivers generally are limited to a single
system. Specifically, international cooperation efforts have led to new GPS and GLONASS signals for new
satellite deployments.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
7
4.3.2 Timing errors
Timing errors in systems that use satellite-based synchronization may be caused by various sources,
including uncompensated antenna cable delays and distribution delays (delay of clock output signals going
to PMUs). Uncorrected delays cause errors in the received timing signals at the PMU. The magnitude of the
errors can be estimated by dividing the electrical length of the cable with the propagation velocity of the
signal along the cable (approximately 3 ns/m—it may take slightly longer time for high-dielectric cables,
and typically 5 ns/m for signals through optical fibers). The errors must be compensated in power-system
applications if they introduce uncertainties not consistent for the desired level of performance.
Loss of signal from one or more satellites (perhaps due to antenna problems or even birds) can also reduce
the timing accuracy, but since the PMU receiver has a known location, it may be possible to “lock” this
position so that an accurate time may be provided even with only one satellite visible.
4.3.3 Systems vulnerabilities
Satellite-based synchronization systems rely on information transfer over the airwaves. The wireless nature
of satellite communications links and the weak power levels of received GNSS signals make them
vulnerable to radio-frequency interference (RFI). Any electromagnetic radiation source can act as an
interference source, if it can potentially emit radio signals in the GNSS frequency bands.
The disruption mechanisms that could limit the GNSS performance can be classified as the following:
a) Ionospheric effects: Sunspot activity causes an increase in the solar flux—charged particles and
electromagnetic rays emitted from the Sun. This solar flux affects the ionosphere and influences the
transit time of satellite signals through the ionosphere. Consequently the receiver equipment may
experience degraded performance in tracking of the satellites due to scintillations, rapidly varying
amplitude and phase of the satellite signal. The equatorial and high latitude regions are most
severely affected by this increased ionospheric activity (Orpen, Zwaan [B57], Volpe [B60]).
b) Unintentional interference: If the line of sight to satellites is restricted (e.g., in urban areas, near or
under foliage), the synchronization signal quality could deteriorate for short or long periods of time.
It is important to have realistic expectations of GNSS availability under conditions where there is
not a clear view of the sky.
c) Radio-frequency interference: RFI is caused by electronic equipment radiating in the GNSS
frequency band (e.g., television/radio broadcast transmitters, mobile phones). Although transmission
is designed to not interfere with GNSS signals, it can radiate at the same frequency as the GNSS
signals if it is faulty or badly operated. This interference, if powerful enough, can lead to
degradation of the GNSS signal received.
d) Intentional interference: Received GNSS signals are extremely weak and can therefore be
deliberately jammed by radio interference. The levels of interference needed to jam a typical
consumer GNSS receiver are quite low, and jamming equipment can be small. Further intentional
interference could be induced by:
1) Spoofing—Counterfeit signals
2) Meaconing—Delay, interception, and/or rebroadcast of navigational signals
3) System damage
4.3.4 Countermeasures
The main strategies that could be adopted to protect GNSS receivers from RFI attacks are based on the
principle of raising the power levels required by the jammers to disrupt the receivers. This makes attacks
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
8
too expensive, unsustainable in terms of the power required to run, or easily detectable and therefore
readily intercepted.
Mitigation strategies, such as physically shielding the GNSS receiver’s antenna from interference sources,
rely on prior knowledge of the location of the interference but may be useful under some circumstances.
Intentional spoofing is much harder to mitigate. There are methods to test and mitigate spoofing, which are
telecommunications technology related and are not addressed in this guide (Wesson, Shepard, Humphreys
[B61]).
The use of modular and flexible synchronization systems, including multiple external timing signals and
local oscillators, can provide a high degree of redundancy to increase reliability and accuracy of the overall
synchronization system.
4.3.5 Performance testing
The performance of commercially available GNSS receivers may vary depending on the receiver hardware
and software architecture. For example:
⎯ Satellite selection: Receivers could adopt different algorithms to automatically select the satellites
used in the timing solution (e.g., the satellites providing the best geometric dilution; see Mills
[B51]). Moreover, each algorithm could be characterized by a different set of thresholds defining
the condition for keeping, dropping or acquiring a satellite. Therefore, different receivers can
obtain different results even when connected to the same antenna in the same location.
⎯ Short-term stability: This is influenced by the hardware architecture of the receiver. In particular,
receivers integrating a satellite disciplined oscillator (e.g., an oven-controlled quartz oscillator or a
rubidium oscillator) exhibit improved short-term stability. To avoid perfect lineup between
sampling frequency and satellite spreading code, an alternative technique currently adopted in
commercial receivers is based on the employment of a temperature controlled crystal oscillator for
down-sampling of the satellite signals. This type of receiver accumulates time errors until the total
error reaches a maximum value (i.e., a multiple of the half period of the oscillator), and then
generates a phase step that reduces the time error to a minimum. Some receivers step phase in
increments of 100 ns (or less) or 1 μs (or larger) (Lombardi et al. [B38], Mills [B51]).
Consequently, the short-term stability of these receivers could vary significantly (although their
long term performance may be equivalent to models integrating a disciplined oscillator).
⎯ Hand-off algorithm: Since the satellite position within the range of the GNSS receiver changes with
time as the satellite movements changes the position, various hand-off strategies may be
implemented in the receivers. The application will determine the level of knowledge and testing
needed by the user.
⎯ “Bad” data management: Receivers manage satellite broadcast errors in different ways. Although
some receivers are equipped by specific software routines able to remove bad data, they might fail
under certain critical conditions (Lombardi et al. [B38], Mills [B51]).
Therefore in order to assess the performances of GNSS receivers applied in PMU time synchronization,
detailed experimental testing is necessary.
4.3.6 Experimental tests
The minimum set of measures may include tests of the Time to First Fix (TTFF), drift tests (time drift rate
after loss of satellite signals), and position accuracy/repeatability. Other tests that may be performed are
more suited in a laboratory environment.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
9
TTFF accuracy measurements are most important in the design validation stage of a GPS receiver.
The most common TTFF conditions are as follows:
⎯ Cold start: The receiver must download almanac ephemeris information to achieve a position fix.
⎯ Warm start: The receiver has some almanac information that is less than one week old, but does not
have any ephemeris information.
⎯ Hot start: The receiver has up-to-date almanac and ephemeris information. In this scenario, the
receiver only needs to obtain timing information from each satellite to return its position fix
location.
In most cases, TTFF and position accuracy are specified at a specific power level. It is worth noting that it
is valuable to verify the accuracy of both of these specifications under a variety of circumstances.
To perform both TTFF and position accuracy measurements, three different sources of data could be
adopted:
a) Live data where the receiver is set up in its deployment environment with an antenna.
b) Recorded data where a receiver is tested with an RF signal that was recorded off of the air.
c) Simulated data where an RF generator is used to simulate the exact time-of-week when live data was
recorded. When testing a receiver with three different sources of data, it is necessary to verify that
the measurements from each source are both repeatable and correlated with other data sources.
4.4 Terrestrial systems
Synchronizing signals may also be disseminated using terrestrial systems (e.g., radio broadcasts,
microwave, and fiber-optic systems).
Network Time Protocol (NTP) is a robust and mature technology for synchronizing a set of network clocks;
however, its performance is inadequate for typical PMU timing by several orders of magnitude.
A time distribution protocol called Precision Time Protocol (PTP) is also available now. PTP Version 2 is
specified in IEEE Std 1588-2008, and its profile for power system applications is specified in
IEEE Std C37.238-2011.
IEEE Std C37.238-2011 requires time distribution with ± 1 μs time accuracy over 16 network hops
(Annex B). At the top of time distribution chain, there is a grandmaster clock that synchronizes the clocks
in the entire system to UTC. Each device in the time distribution chain (including Ethernet switches) is
required to support IEEE C37.238-2011 to achieve 1 μs time accuracy. Ethernet switches supporting
IEEE C37.238-2011 should perform measurements and corrections for cable delay and residence time (e.g.,
variable time in which a synchronization message spends inside an Ethernet switch due to queuing and
other processing delays).
IEEE Std C37.238-2011 and IEEE Std 1588 do not operate over wireless networks at this time.
IEEE C37.238 distribution technology offers 1 μs time accuracy for PMU applications, the use of the same
communications infrastructure (Ethernet) for PMU/PDC data and time distribution, and reduced use of
GPS connectivity whenever possible. Due to these benefits a rapid adoption of this technology is expected.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
10
4.5 Synchronization distribution methods
IEEE Std C37.118.1-2011 requires a PMU to be synchronized to UTC time. UTC time synchronization
may be delivered to the PMU by using, for example:
⎯ IRIG-B: The IRIG-B time code is fully described in IRIG Standard 200-04. It repeats each second
and has a total of 100 bits per second. Some of these are framing (sync) bits, some are assigned for
time, and some are available for control functions. IRIG-B code may be used in either logic-level
(unmodulated) format or as an amplitude-modulated signal with a 1 kHz carrier (Dickerson [B13]).
The modulated IRIG signal is generally capable of accuracy exceeding
1 ms (one period of 1 kHz), but not usually better than 10 μs. The unmodulated IRIG-B code can
deliver accuracy limited only by the slew rate of the digital signal, usually better than 1 μs.
IEEE Std C37.118.1-2011 defines use of the control bits in the IRIG message to provide extensions
for real-time applications: leap second and daylight savings or summer time status; year of century;
and time quality. These extensions are generally required to meet the requirements of
IEEE Std C37.118.1-2011.
⎯ 1 PPS: a one pulse per second positive pulse with the rising edge on time with the second change
provides precise time synchronization (IEEE Std C37.118-2005). However, since each pulse is
identical there is no way of knowing which second a pulse is associated with. Resolving this
ambiguity requires a simultaneous data channel.
⎯ IEEE 1588: IEEE Std 1588-2008 specifies a PTP and IEEE Std C37.238-2011 specifies an
IEEE 1588 profile for power system applications, such as PMU. PTP distributes precise time over
Ethernet-based networks over multiple network hops and requires special hardware support at each
Ethernet port to achieve high time accuracy. Messages containing precise actual time are
transmitted once per second. By adding dedicated timing hardware to each port in a data network,
the time of transmission and reception of certain messages can be determined with accuracy
sufficient to transfer time with performance comparable to that of an IRIG-B or 1 PPS signal. The
protocol supports corrections for variable cable and processing delays in intermediate devices (e.g.,
Ethernet switches). The protocol needs to be supported by all devices in time distribution chain to
achieve 1 μs time accuracy.
IEEE Std C37.238-2011 specifies extensions for real-time applications: leap second and daylight
savings, local time (if needed), and time quality. In addition a flag that indicates if provided time is
traceable to UTC is supported. These extensions are generally required to meet the requirements of
IEEE Std C37.118.1-2011.
5. Synchrophasor measurement accuracy characterization
5.1 Introduction
The overall objective of this clause is to provide a method by which users can assess the overall accuracy of
the instrumentation channel including their selected instrument transformers and GPS-synchronized phasor
measurement equipment. To do this, users should define accuracy characterization tests to be performed on
GPS-synchronized equipment, which will provide the necessary information to make informed decisions as
to the quality of data obtained with these units. Users should also determine the level of inaccuracy injected
into the measurements from instrumentation channels and provide methodologies to quantify this
inaccuracy.
Sources of error include instrumentation channel characteristics, GPS-equipment characteristics, and
system asymmetries.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
11
The characterization process is separated into two parts, as follows:
⎯ Accuracy for power frequency data (fundamental frequency phasors)
⎯ Accuracy during dynamic changes
5.2 Data accuracy characterization
The purpose of the instrumentation channel is to provide isolation from the high-voltage power system and
to reduce the voltages and currents to standard instrumentation level. Figure 1 illustrates the devices
forming voltage and current channels typically found in electric power substations. Ideally, it is expected
that the instrumentation channel will produce at the output a waveform that will be an exact replica of the
high voltage or current and scaled by a constant factor. In reality, the instrumentation channel introduces an
error. Specifically, each device in this chain, namely, instrument transformers, control cables, burdens,
filters, and analog-to-digital (A/D) converters, may contribute to some degree to signal degradation.
Furthermore, the error introduced by one device may be affected by interactions with other devices of the
channel. It may thus be important to characterize the overall channel error.
Figure 1 —Typical voltage/potential and current instrumentation channels
The first link in the instrumentation channel equipment chain consists of voltage and current transformers
(CTs), collectively called instrument transformers. These devices transform power system voltages and
currents to levels appropriate for driving relays, fault recorders and other monitoring equipment, isolated
from the original quantities. Several instrument transformer technologies are presently in use. The most
common devices are potential or voltage transformers (PTs or VTs) and CTs based on magnetic core
transformer technology. Another type of commonly used voltage transducer is the coupling capacitor
voltage transformers (CCVT), based on a combination of a capacitive voltage divider and a magnetic core
transformer. Voltage and current instrument transformers have also been constructed based on the electro-
optical and magneto-optical phenomena. These devices are known as optical voltage transformers (OVT)
and optical current transformers (OCT). There are also other sensor and transducer technologies available
for voltage and current measurement (e.g., Rogowski coils and Hall-effect devices). While reference is
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
12
made to some of these newer types of instrument transformers, this guide mainly focuses on VT, CT, and
CCVT devices.
These varieties of primary sensors allow us to classify instrumentation channels into five common
categories, depending on the instrument transformer used, as follows:
a) CT-based instrumentation channel
b) Wound-type VT-based instrumentation channel
c) CCVT-based instrumentation channel
d) OVT-based instrumentation channel
e) OCT-based instrumentation channel
Although this guide provides information of possible sources of error for each one of the five generic
categories listed above, focusing on the top three as the most commonly used instrument transformers, it is
important to realize that each of these categories may have several options (e.g., a CT-based
instrumentation channel may be implemented with different accuracy class CTs, see IEEE Std C57.13).
5.3 Data accuracy
GPS-synchronized equipment has the capability to provide a data acquisition system with the following
accuracy:
a) Time tagging with accuracy better than 1 μs (or equivalently 0.02 degrees of phase at 60 Hz).
b) Magnitude accuracy of 0.1% or better.
This accuracy may not be available in all GPS-synchronized equipment. Even for the equipment that
conforms to IEEE Std C37.118-1-2011, this accuracy cannot be achieved for the overall system in any
practical application (e.g., in the substation environment). In addition, depending on the implementation
approach and equipment used, the accuracy of the collected data and the reliability of the data availability
may differ. Typical GPS-synchronized equipment (PMU) are very accurate devices. However, the inputs to
this equipment are scaled down voltages and current via instrument transformers, control cables,
attenuators, etc., collectively referred to as the instrumentation channel. The instrumentation channel
components are typically less accurate. Specifically, potential and current instrument transformers may
introduce magnitude and phase errors that can be orders of magnitude higher than the typical PMU
accuracy. Although high accuracy laboratory grade instrument transformers are available, their application
in the substation environment is practically and economically infeasible.
Note that for most of the CT, VT, CCVT, etc., in substations, the associated secondary circuit wiring
(significant component of the instrumentation channel) is not normally “instrumentation class” wiring. In
many cases, this wiring is control type cabling (non-twisted pairs) and is often unshielded. Often changes
are made to these secondary circuits that affect the overall secondary circuit burden (e.g., adding or
replacing relays or other devices when electromechanical or static equipment is used, which has a high
burden), without a detailed engineering analysis of the impact on high accuracy applications such as the
PMU installation. This problem does not exist when the modern microprocessor relays are used because
they have a very low burden. The use of isolating switches, the application of grounds on these secondary
circuits, and the presence of nonlinear burdens are a few of the items that can have a significant impact on
the accuracy of the instrumentation channel.
In some jurisdictions, utility regulators have mandated the use of dedicated instrument transformers for
revenue or tie line metering (including those located in high-voltage substations) as well as the application
of specific design and testing criteria for the associated secondary circuit wiring. In at least one jurisdiction,
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
13
this secondary wiring is “secured” to help ensure that other devices (burdens) are not inadvertently
connected, either permanently or temporarily. In other words, the instrument transformer secondary circuit
is carefully designed and tested (measuring actual burdens), and, then, access is controlled to help ensure
the ongoing accuracy of the overall revenue metering installation.
With many users in the long term, there is a risk that the presence of PMUs may be overlooked when
changes are made to the secondary circuits of “shared use” instrument transformers. Care should be taken
to avoid such inadvertent impact on the accuracy of individual PMU installations.
The previously described issues are some of the commonly practiced considerations for the purpose of
assessing the quality of data from PMUs.
5.4 Characterization of instrumentation channels
High-voltage instrumentation channels introduce errors to phasor measurements. The level of error is
dependent upon the type of instrument transformers, control cable type and length, and protection circuitry
at the input of the PMUs. Table 1 illustrates an example of the errors for a specific VT instrumentation
channel with 500 ft of cable between the VT secondary and a PMU. Note that the VT introduces a very
small error (0.01 degrees), while the 500 ft cable introduces an error of 0.54 degrees. The overall error is
more than an order of magnitude higher than the error of a typical PMU.
Table 1 —Representative instrumentation channel errors—example with 500 ft of cable
between VT secondary and PMU
Primary voltage
VT secondary
voltage
Error
PMU
input
voltage
Error
V
an
62.53 kV
27.52º
62.19 V
27.51 º
0.68%
0.01 º
61.63 V
27.11 º
1.44%
0.41 º
V
bn
62.96 kV
–92.68 º
62.61 V
–92.70 º
0.55%
0.02 º
63.09V
–92.85 º
0.2%
0.17 º
V
cn
62.33 kV
147.46 º
61.99 V
147.45 º
0.54%
0.01 º
61.72 V
148.00 º
0.98%
0.54 º
Depending on the instrumentation channel, characterization of these errors may be possible. Meliopoulos,
Cokkinides [B45] provides some additional information. In addition to actual field calibration, in many
cases these errors may be accounted for and corrected via software. The following two software approaches
are considered:
a) Modeling the instrumentation channel and providing model based correction algorithms
b) Using state estimation methods to correct the error
A combination of these two approaches has advantages. Addressing this issue is important for overall
accuracy.
Annex C and Annex D provide examples of instrumentation channel characterization and the effects on the
overall accuracy of the GPS-synchronized measurements. Further work is recommended to develop
methods for characterizing the instrumentation channel errors, and algorithms to correct for these errors.
GPS-synchronized equipment may also be connected to existing instrumentation in substations that may
serve other purposes (e.g., metering). For example, the instrument transformers may be connected in an
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
14
arrangement that generates a phase shift (e.g., delta connection). The resulting phase shift must be
accounted for.
Optical CTs and VTs present an opportunity for more accurate measurement because of their inherently
linear response. Nevertheless, in case of optical CTs and VTs having analog outputs, the analog interface
device, i.e., the electronic equipment associated with converting the optical CT/VT measurements to analog
output signals, must be qualified for PMU related applications. Refer to IEEE Std C37.233-2009 for
information related to optical sensor characterization and testing. The interface device introduces signal-
processing and conversion latency, typically in the order of few tens of microseconds (equivalent to TVE of
1% to 4% if not compensated for). There are methods to compensate for delays associated with this signal
conversion, either in the optical CT/VT analog interface device (preferred), or within the PMU. For
example, adjusted time tagging within the PMU (i.e., subtracting sensor latency before generating the PMU
time stamp) can yield very accurate synchrophasor data.
Ultimately, optical sensors providing digital output may be more suited for PMU applications than analog
outputs. IEC 61850-9-2 allows for suitable digital outputs from the merging unit, which provide digital
time-tagged measurements of voltage and current signal waveforms with the signal processing delays
already accounted for (within the accuracy specification of the digital instrument transformer). In these
cases, the PMU function will be a simple mathematical conversion from “sampled values” to
“synchrophasor data” through a purely digital algorithm (no analog-to-digital conversion errors), with very
consistent accuracy and repeatability.
5.5 Characterization of GPS-synchronized measurement devices (PMUs)
Equipment for synchronized measurements from various vendors may have different designs and,
therefore, different ways of data acquisition and processing and different accuracy characteristics. It is
expected that devices conforming to IEEE Std C37.118.1-2011 will have matching characteristics and
perform similarly within the allowed error limits. It is additionally possible for a user to identify specific
application requirements and coordinate expected performance with manufacturers.
Full characterization of a GPS-synchronized device should include error analysis of both timing accuracy
and magnitude accuracy over a generally accepted range of operating conditions defined in terms of
frequency, frequency rate of change, voltage magnitudes, current magnitudes, harmonics, and imbalances.
IEEE Std C37.118.1-2011 defines the standard and testing ranges and limits.
5.6 GPS-synchronized equipment reliability
Reliability data for GPS-synchronized equipment are scarce. The few in-service reliability data available
from first-generation equipment may not be representative of the present technology. Nevertheless, we
have included Table 2 to illustrate reliability data for some PMUs on the western system (North America)
over a two-month period in 2002. The synchrophasor system performance analysis typically includes
recording of data loss, signal loss, and PMU time-synchronization failures. In Table 2, signal reliability is
the percent of time the system continuously received data from the PMU. Sync reliability is the percent of
time the PMU is synchronized with a GPS. Note that the few available data indicate that most of the
unreliability is due to the GPS signal availability.
It should be emphasized that device quality might have improved significantly since 2002, and many new
devices and systems provide much better reliability. This example is only a reference. System reliability
additionally depends on system architecture, which, in turn, is influenced by system requirements and
purpose.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
15
Table 2 —Synchrophasor system performance for a random two-month period in 2002
Station Reliability (%)
PMU Signal Sync
Notes
GCoul 97.52 99.974 PMU failed after 2 days
JDay 99.929 99.996 Normal, modem
Malin 99.997 93.74 PMU clock failure
Colstrip 99.82 100
Communications system
problems
BigEddy 99.99 99.988 Normal, fiber, digital
MValley 99.983 99.74 PMU clock problems
Keeler 99.996 99.95 Normal, modem
6. PMU installation, commissioning, and maintenance
6.1 Preface
This clause discusses recommendations for PMU installation, and is based on general installation
requirements for PMUs and typical substation configurations. The information found in this clause is only
considered a starting point, and is expected to be expanded and modified as applications are expanded.
6.2 Overview
A PMU installation requires access to the power system signals to be measured and a time signal to time-
stamp the measurements (see Figure 2). It also requires a communications system to transmit the
measurements to a remote location in real time, at the PMU reporting rate and matching the format and the
interface.
In some cases, the PMU will have status inputs (Boolean 1 or 0) or other measured value (“analog”) inputs
such as temperature, wind speed, or power factor. PMU communication provides support for analog and
digital data as well as phasor, frequency, and rate of change of frequency (ROCOF) data. Access to these
signals will depend on the situation, but needs to be considered at the planning and specification stages. See
IEEE Std C37.118.1-2011 for more detail on synchrophasors.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
16
Figure 2 —Representative (example) PMU system showing major elements
6.3 Pre-installation procedures
6.3.1 Installation design
In most cases, a PMU installation is considered a permanent installation and requires a complete design.
Following are supplementary guidelines to help scope a project and proceed with the design. Typical
design includes selection of power system sources, inputs, outputs, alarms, absolute time source, and
communications interfaces. Some of the key components of a typical PMU are described in the following
subclauses.
6.3.2 Timing input
Timing accuracy is critical. The choice of hardware, antenna location, cabling, and routing of the cable are
therefore critical in the overall design of the PMU system. For a GPS system, an antenna open to GPS
satellites should have a clear sky view (free from obstructions) above about a 10 degree elevation, so that
some satellites are always in view. In cases where this is difficult to achieve, some compromise is possible
because the PMU internal clock should be able to ride through complete or intermittent loss of GPS
synchronization. When available, it is recommended that the GPS clock be configured with its position
LOCKED (a setting in some GPS clocks). With the position LOCKED, the GPS clock can deliver accurate
time with only one satellite visible / receivable.
Figure 3 shows recommended antenna mountings. In the northern hemisphere most of the signals will come
from the south (see Figure 4). It is expected that there may be obstructions that obscure the sky within ten
degrees of the horizon. Above that, good visibility of the sky should be achievable.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
17
Figure 3 —Recommended antenna mounting locations
Figure 4 —Example of GPS signal visibility pattern reflecting the orbit
for various satellites
Figure 4(a) shows what is called a sky plot or a polar plot of the satellite orbits. It indicates the trajectories
of the GPS satellites for a day, essentially the view from a fish-eye lens gazing upward. This particular plot
is based on a latitude of 45 degrees north (about the latitude of Minneapolis), so the satellites trajectories
across the sky are predominantly to the south. (The satellites would appear to rise in the east and set to the
south, or rise in the south and set to the east. At this assumed latitude, some do go directly overhead. If the
location were further north, the “hole” in the plot would move down the diagram, as no satellites would go
directly overhead.) Figure 4(b) shows the time of day that the various satellites would be visible. At any
latitude, at least six satellites should be in the sky at all times, and at this latitude, the number is often as
many as nine.
In the northern hemisphere, the antenna should be mounted with a clear view to the south. Small
obstructions more than half a meter away from the antenna will not cause a problem, but a large flat
obstruction within a few hundred meters could act as a reflector and cause multi-path problems. Check
around the mounting location for a structure, such as a flat metal roof, that is oriented so it can reflect a
satellite signal to the antenna (keeping in mind satellites will traverse most points in the sky). Also check
for obstructions that can block the signal, and high power signal sources that can saturate the GPS input.
The most reliable antenna mounting option is an open air sky-view installation.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
18
The satellite synchronization signal (center frequency) is transmitted in the 1 GHz to 2 GHz range, which
attenuates rapidly in a cable. Most device suppliers recommend limiting cable runs to less than 50 m. There
are alternatives for longer cable distances, such as high-gain antenna or in-line amplifiers and low-loss
cable. It is common practice to compensate the time signal to address delays and latency associated with
the length of the antenna cable. The cable length could be a straight run from the antenna to the clock
receiver and/or the distance within the control room to distribute the clock signal. Cable delay
compensation and/or clock offset are also available in some clocks.
When substations do not have adequate lightening protection that includes the antenna location, such as at
distribution substations or pole line locations, it is advisable to incorporate a lightning arrestor into the
antenna design.
It is common practice to use a signal source that will provide the required signals at the accuracy the PMU
requires for meeting timing requirements. Some PMUs input time from a local source, such as a GPS
receiver or clock server, using a local signal type such as IRIG-B, 1 PPS, or IEEE 1588. Some of these
signals degrade rapidly. Delays are associated with the dielectric constants (i.e., capacitance of cable) and
the length of the transmission medium. Therefore, excessive cable runs should be avoided. When using this
type of PMU, consult the vendors as to what signals they require and whether the delays are compensated.
For example, IRIG-B may be specified and can be used in any of its modulated forms, but the dc level-shift
or the modified-Manchester coding forms will allow the highest accuracy. The example shown in Figure 5
uses Manchester time coding.
Figure 5 —Example showing forms of IRIG-B comparing the unmodulated (level shift) B000,
1 kHz modulated B120, and modified Manchester B200
A GPS-fed clock can potentially serve a number of devices using one IRIG-B output port. The number of
clients (users) that a clock can serve depends on both the drive capacity of its IRIG-B port as well as the
amount of load (both capacitive and resistive) that the devices connected to the clock represent. The
cabling, associated routing, and termination methods also affect the clock loading. To design a highly
reliable timing circuit, product specifications should be studied and respective manufacturers may be
consulted. Some of the key factors to consider are the following:
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
19
⎯ Load and capacitance of the wire/cable—use ultra low capacitance cables. For example, twisted-
pair usually has lower capacitance than coaxial cable;
⎯ Cumulative sum of the loads represented by input impedances of all clock clients (PMU and non-
PMU) and the associated cabling should not be larger than the clock’s output port (IRIG-B port)
output serving capacity with some safety margin (e.g., more than 30% margin).
⎯ The impact or presence of redundant clock architecture should be considered. For example, if the
application has primary and alternate clock provisions into the respective device input, both clocks
need to have similar port output ratings.
The use of fiber-optic time distribution may overcome several of the issues previously referenced.
Signal propagation in cabling should be considered and corrected for longer cables (on the order of 1 ns/ft
or 3 ns/m of delay for electrical cables and 5 ns/m for glass optical fibers). Also, dispersion in longer cable
length (depending on the amount of effective impedance/capacitance) will be seen as slowing down the
rise-time of the clock pulse edge received by the PMU. Accordingly, minimizing cable lengths will be
beneficial. When daisy chaining of the clock signal is considered, the user should carefully examine the
devices to make sure the receiving device does not process the clock and then pass to other devices that are
part of the daisy chaining. Processing time by the first device will introduce delays for the other subsequent
devices that are part of the daisy chain.
The type of connector, e.g., screw-terminal versus BNC, usually has negligible impact—all other factors
being equal.
When driving several receiving devices from one time signal output, the signal level and termination
should be checked to verify adequate signal amplitude and quality are available for the receiving device to
lock into the timing signal.
When designing the timing circuit, the substation environment and routing of the cable should also be
considered. In larger substations, it may be more practical and economical to use two or more clocks and
GPS antennas to serve devices in different parts of the substation as opposed to running long cables to all
devices from one clock. This approach also helps with distribution of the load on a given clock.
Some PMUs (and other devices) have options to accept clock signals (through IRIG-B, 1 PPS, etc.) and/or
to use an internal GPS clock to provide internal timing information if connected to a GPS antenna. The
choice of timing circuit depends on various factors. Generally, sharing GPS antenna signals can become
complicated, especially when active antennas (antenna requiring power from the clock) are used.
Accordingly, use of a PMU’s internal GPS clock connected to the GPS antenna is more practical when the
PMU is the only device (or one of very few devices) using the time information. In cases where absolute
timing information is necessary for several devices, use of external clock connected via IRIG-B interface
(or 1 PPS signal) is more practical. Use of an external clock may also facilitate on-site troubleshooting
processes.
It should also be noted that the naming convention for identifying a clock when using 1 PPS signal is not
standardized; it is product specific. Consequently, the use of IRIG-B protocol may be more appropriate as
opposed to using the 1 PPS signal for sharing clock signals.
6.3.3 Voltage and current input
When possible, the PMU should be installed with test access to the PT and CT inputs so test signals can be
injected for performance tests and calibration. When multiple PMU devices are installed, test procedures
are required to account for the various sources of current and voltages signals.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
20
Good measurements require consideration of the input signal sources. PT and CT characteristics and cable
coupling and other loads may lead to distortion, both in amplitude and phase. PMUs require an accurate
time source to generate accurately time-stamped measurements (in accordance with standards such as
IEEE Std C37.118.1-2011). The most universally available time source for this purpose is the GPS satellite
system. The fundamental GPS time pulse extraction may be augmented by backup or flywheel oscillators
and may be distributed to PMUs internally, by direct connection, or by network.
The PMU measurement and reporting rates must be sufficient to capture the characteristics of interest.
Performance standards describe necessary data rate and corresponding bandwidths, but desired parameters
will dictate actual rates used. PMU full-scale ranges are a function, also, of the measurement being made
and the system used. For example in steady-state configurations monitoring disturbances, full-scale settings
will be different from a multifunction device performing protection or fault recording along with steady-
state PMU data.
User-defined scaling factors will impact resolution and dynamic range for measurement/monitoring phasor
quantities. Scaling factors are application driven. For example, under certain operating conditions, when
voltages up to 200% of nominal voltage are of interest, the scaling factor of 50% would be desirable and
would prevent the secondary voltage measurements from being clipped. This threshold set point gives some
overhead for large swings, but good resolution for nominal input. On the other hand, current values will
need a much larger dynamic range. It is very important to consider references to “nominal current” with
respect to normal currents on the primary system. When connecting current measurement inputs of the
PMU to the CTs, the rating and ratio of the CTs relative to normal primary current levels should be
considered. For example, at a wind farm with certain energy generation capability, the nominal primary
current may be 400 A while the short circuit duties near the station may be much higher (e.g., on the order
of 60 kA). In order to have adequate resolution (and accuracy) at 10% of nominal current (40 A primary),
using a protection CT output (operating near 1% of its 3000 A rated value) may not be appropriate.
Fault recording and relaying requires a scale that will accommodate very high currents; phasor
measurements are concerned with normal operational and system swing currents that range from somewhat
over nominal to light load (10% of nominal). Setting the maximum readable value at about
two times nominal will give good measurements at light load and normally will cover swings. However, it
may be considered more important to capture all swings and overloads at the expense of lighter loads. If
synchronized measurements are desired during faults, values up to 40 per unit (pu) may be considered.
Sometimes the PMU is a part of another device, such as a relay or digital fault recorder (DFR) and the
calibration will be dictated by another requirement. This is a user consideration where the user needs to be
aware of the tradeoffs.
The PMU will also derive frequency from the ac input signals. These are fully user selectable in some
PMUs and fixed in others. Some PMUs will allow selecting a backup channel when the primary one has a
low signal. To provide a good frequency measurement, it is best to select a voltage input channel that has
good signal strength, since these do not vary as much as currents.
6.4 Analog and digital input
Some PMUs allow connecting other measurements such as megawatt (MW) transducers, or sensors for
measuring temperature or wind speed. These values will be reported along with phasor values in the time
synchronized data packet, which opens up the use of phasor data for many real-time and analysis
applications. Generally these data signals are easier to transmit than high-current CT circuits, but they need
to be considered. When a PMU provides “status” inputs such a breaker status, these parameters are
represented as a Boolean logic (“0” or “1”) single bit in a digital word. Access to these signals is usually
not difficult, but again needs consideration during installation.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
21
6.5 Power input
Since PMUs need to operate continuously, and without interruption during disturbances, most applications
are designed to connect the PMU to the station dc power. When using station dc, the impact of the dc ripple
to the power supply should also be considered to make sure quality or accuracy of the measurements are
not impacted. PMU Manufacturers should specify dc ripple immunity.
6.6 Communications
Communications interfaces are critical parts of phasor measurement systems. It is important to carefully
plan the system and be sure all aspects are addressed. The first part will be cabling from the PMU to the
communications interface device. Next will be the interface medium. The third component is the interface
with the PMU. The overall system, including the application end (PDC or other devices), requires complete
planning of both the physical and logical networks, both local to the substation and remote, where this
Guide only discusses some of the physical design issues at the substation. IEEE Std 1615™ [B31] can be
consulted for logical network design issues for substation networks. Station architecture must be considered
in the design of the communications interfaces. For example, when an application includes an interface
with station bus, the PMU communicates with the station bus processor. Likewise, in process bus
applications, the PMU interfaces perhaps through a merging unit to the local data processor.
6.6.1 Cabling
Cabling is required between the PMU and the communications interface (e.g., modem, router). Cable
installation recommendations can be found in IEEE Std 525™ [B28] and IEEE Std 1615 [B31], where care
should be taken in routing communication cables in existing cable trays. The principal issues are
interference and cable length. An additional consideration is the impact of the secondary voltage during a
nearby fault. When possible, use fiber-optic (FO) cable for all signals that travel outside of a single rack
(grounding unit). Care should be taken with communications cable to ensure that the length does not
exceed maximum specified distances, as shown in Table 3. In addition, IEEE Std 1615 should be followed
regarding the application of FO and copper cabling for serial and Ethernet cables. If the signal is to run
between buildings (e.g., between a control building and a communications interface building), it is best to
use FO cables. In some cases, this requires signal converters at each end, but many vendors have FO
communication ports built into their equipment.
When metallic cables are used, considerations should be given to minimize the most likely sources of
interference. Coaxial cable has good electrostatic shielding but is subject to electromagnetic interference,
particularly where ground differences can occur (conduction through the ground). Shielded twisted pair
(CAT5) has good rejection for both electrostatic and electromagnetic interference, but all shielding has
limitations. High magnetic fields can penetrate most kinds of shielding and are best treated with appropriate
grounding and bonding practices. It is best to route cables with some thought as to what is in close
proximity. Also note that many substation cables are large and heavy, so they may pinch a poorly routed
communications cable, which may result in loss of shielding and transmission properties. Table 3
summarizes the signal types, the recommended and practical length, and some comments.
In addition to the physical installation considerations, the communication types need to match in type and
rates. Some, but not all, PMUs can report at higher rates than required by IEEE Std C37.118.1-2011, if
required by the application.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
22
Table 3 —Recommended cable lengths for signal transmission
Signal type Cable type Recommend max length Interference/comments
Asynchronous
serial (EIA-232)
Twisted
pairs
15 m @ 20 kbps Actual length depends on cable and data rate.
V.35
Twisted
pairs
600 m @ 100 kbps
90 m @ 10Mbps
Standard well specified and followed. Widely used
outside of U.S.
Synchronous serial
(EIA-422)
Twisted
pairs
1200 m @ 100 kbps
Bipolar low-level signal has good transmission
characteristics.
Ethernet thin-net Co-axial
Good static shielding, use tri-axial cable where
there are grounding issues. Actual length depends
on cable and data rate.
Ethernet 10BaseT CAT4 100 m Good shielding, move up to Cat5 for better signal
Ethernet
10/100BaseT
CAT5 100 m
Good shielding, use ruggedized version for better
mechanical protection.
a
Ethernet
100BaseFX
FO 2 km
Best for interference rejection and avoiding
grounding problems.
a
Standard CAT5 is unshielded twisted pair (UTP). This should be shielded twisted pair (STP).
6.6.2 Interface medium
PMUs generally will have asynchronous serial (EIA-232, previously known as RS-232) or Ethernet
communications. Both will handle data at the rate and block size used in most measurement systems. An
interface device between the PMU and the communications system is required. The user must identify the
communications medium, interface, and technology required for the particular application (e.g., whether
multi-cast communications is needed). Generally, a modem translates between a digital data system and an
analog communications system, and a service unit or router or bridge translates between a digital data
system and a digital communications system. In either case, the PMU output is a digital data type and needs
one of these devices to interface to the communications system. Most importantly, the appropriate sides of
the interface have to match the PMU and communications system, and must handle the required data rate.
If the PMU output is asynchronous serial, the interface must handle asynchronous serial at the given rate.
See Table 4.
8
PMU communications by Ethernet match directly with digital systems and interface well with digital
communications such as SONET. Ethernet-based digital systems are well developed and supported, so
building these systems is generally very straightforward. Care needs to be exercised when connecting the
local Ethernet system through a digital communications system (e.g., SONET), usually called the Wide
Area Network (WAN) connection. The WAN needs to support the minimum data rate with some room for
overhead, and the interface needs to have sufficient buffering to match WAN and Ethernet speed
differences. Most of these systems currently use Internet Protocol (IP) for data communications. Building
and managing these systems is centered on setting up IP hosts, subnets, and routing. Additional
considerations for communications interface include data exchange with neighboring systems, neighboring
countries, local independent (electricity) system operator (ISO), or at the regional levels. Depending on the
application, data reliability and accuracy may be critical factors.
Cyber security requirements are also important considerations when the application is considered critical.
Each utility will have its own recommendations, but the use of a virtual private network (VPN) form of
communications not connected to the internal Local Area Network (LAN) is one option for providing
isolation and security. Care should be used when applying only VPN technology for security, as whatever
traffic traverses the VPN may be inherently trusted and create a vulnerability to both connected systems.
8
Notes in text, tables, and figures of a standard are given for information only and do not contain requirements needed to implement
this standard.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
23
Table 4 —Examples of required data rates using IEEE C37.118 data format and protocol
30 FPS
All integer format
(bytes/s)
All floating-point format
(bytes/s)
5 phasors, no analog/digital 1260 1980
10 phasors, no analog/digital 1860 3180
5 phasors, 2 analog/ 1 digital 1440 2280
10 phasors, 2 analog/ 1 digital 2040 3480
NOTE 1— Asynchronous serial requires 10 bits/byte, so required bps rates are 10 times the above
figures.
NOTE 2— UDP/IP over Ethernet has a fixed size overhead of 54 bytes per packet, so actual required
rate is higher than that requirement above. It ranges from 23 040 bps for phasor/integer to 41 120 bps
for the 10 phasor, 2 analog/1 digital/floating point shown above.
NOTE 3— Since data is sent continuously at the rates shown in the table, the communications
channel should have a capacity at least that large, and should be at least 10% higher than the required
data rate to accommodate error correction and short dropouts. The PMU port speed should likewise
be equal or higher than the actual data rate. In many cases this requires high serial data rates, such as
38.4 kpbs or 57.6 kpbs.
NOTE 4— When TCP/IP is used, the spare bandwidth needs to be significantly larger in order to
accommodate retries. Additionally, when TCP is used, the transmission data buffering needs to be
large enough to accommodate the largest expected communication loss time.
6.6.3 Communication over serial port
IEEE Std C37.118.1-2011 and IEEE Std C37.118.2-2011 define the concept of “frames” for transmitting
data from a PMU to a PDC. The standard does not impose any restriction on the communications media
itself. Basically a Configuration frame, Data frame, Header frame and a Command frame are specified.
These frames have a particular structure and data type associated with them. Configuration frame, Data
frame, and Command frame are binary types, while Header frame is a ASCII type. The Data frame is the
most frequently transmitted message based on the PMU sample rate, and the typical size is of the order of
few hundreds of bytes. The variable size in the Data frame is the number of phasors and analog and digital
signals transmitted depending on the PMU capability. For example, if serial communications is chosen to
transmit the PMU data, the data transfer capability depends on the baud rate of the communication port.
Table 5 shows the typical data transfer capabilities, assuming that the PMU supports up to 12 phasor
channels.
Table 5 —Estimated number of phasor channels that can be transmitted at
various bit rates and PMU reporting rates over serial port, assuming that the PMU
supports only up to 12 phasor channels
Reporting rate (FPS)
Bit rate
(bps)
10 12 15 20 25 30 50 60
9 600 12 12
a
10
a
6
a
4
a
2 0 0
19 200 12 12 12 12 12 10 4
a
2
38 400 12 12 12 12 12 12 12 10
56 700 12 12 12 12 12 12 12 12
115 200 12 12 12 12 12 12 12 12
a
Calculated required bandwidth is numerically very close to the actual bit rate.
Actual number of phasors may be less than the estimate. The above estimates were
calculated for 1 start bit, 1 stop bit, 8 data bits and no parity bit, or a total of 10 bits to
transmit each byte.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
24
6.7 Summary of design considerations
Table 6 summarizes basic design considerations in the order of importance. Since many factors impact the
application, it is important to scope the entire project before deciding on equipment and design. For the
overall project consideration, spare parts, training, overall cost, and company policies are equally important
factors.
Table 6 —Summary of design considerations for PMU installation
Question Answers Additional comments
Primary source of current and
voltages?
Number of available voltage sources,
location of the voltage source (bus side or
line side). When line voltages are used,
contingencies are commonly considered in
case a line is out of service.
Number of feeder currents needed for
important power flow calculations?
Does PMU require direct GPS
antenna input?
The choice of IRIG vs. GPS antenna
connection is a user consideration based on
the device specification. The choice of
synchronization is based on the PMU
design—connect IRIG-B, IEEE 1588 over
Ethernet, or a GPS antenna as required by
the manufacturer. Note that some devices
may accept both IRIG-B and IEEE 1588 as
redundant clock sources. When a GPS
receiver is provided within the device, the
clock is internal to the device. The mounting
location for the antenna must provide for
best reception of the signal. Several GPS
receivers can be connected to one antenna.
When using an IRIG-B input, the clock is
naturally external to the PMU. It is
recommended to use unmodulated IRIG-B
signals as opposed to modulated IRIG-B.
IRIG-B/1PPS signals are generally easier
to route, but check wiring availability.
Long cable runs or serial line drivers may
cause unacceptable delays in IRIG B/1PPS
signals.
A suitable GPS antenna splitter is required
when connecting more than one GPS
receiver to a single antenna. Some
antennas require preamplifier power so the
receiver and splitter must be capable of
routing appropriate power to the antenna.
Does the PMU output match
the communications system
interface?
If system is Ethernet connected and PMU is
EIA-232, make sure the interface conversion
is available with type and speed required (see
Table 5).
What types of communications
are established within the
substation?
If FO, will it match the PMU interface?
If not FO, is the distance short enough for
metallic cabling (see Table 3)?
Other factors include grounding and
shielding when metallic mediums are used.
Design considerations should include
methods of grounding the shield and
whether the cable shield is grounded on
both ends or one end.
What is the impact of including
a PMU into an existing CT and
PT circuits, and the additional
burden impacting PMU
performance?
Most PMU devices have a very low burden.
A good design with engineering
calculations for the overall burden is
necessary to prevent installing a PMU in a
circuit that would impact performance of
the PMU.
When impact of the burden is high,
separate sources (CTs and PTs) may have
to be used or installed to achieve PMU
performance.
Does the application have
particular filtering needs (P
and M classes as defined in
IEEE Std C37.118.1-2011)?
Yes, be sure the PMU has settings that match
or use alterable settings to satisfy the
application requirements.
Filter in this context is in reference to P
and M class. In general, microprocessor
based devices may have filtering beyond
the intent of the reference to the P and M
classes.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
25
6.8 Pre-installation tests
The extent of pre-installation testing depends on power company practices. At a minimum, some quality
assurance testing should be done to verify that units work in the planned system and are reasonably
calibrated. Many of the tests can be performed using conventional test equipment. In addition, a risk
assessment is needed to investigate worst-case problem scenarios that may occur while commissioning the
PMU.
A complete set of test and performance results for each PMU type must be acquired to confirm that the
product meets the user specification. This testing is referred to as type test (see IEEE Std C37.233-2009 for
additional information regarding different types of tests).
As part of the pre-installation testing, wiring checks are performed to verify that the PMU connectivity
meets the user specifications and the manufacturer-specified installation requirements. Performing some
basic tests will confirm proper wiring, terminations, and grounding of the equipment mounted. Pre-
installation tests for the communications components, such as cabling, will be the next step in the pre-
installation test sequence. Typically, the biggest problem is data communications, so minimizing other
problems ahead of time is worth the effort.
Cyber security should be coordinated with testing, such that testing is done without cyber security first to
confirm the PMU data flow and then later with cyber security applied to confirm the application still works
when secured.
Communications links should be installed and successfully tested for continuity or transport prior to the day
of commissioning the PMU.
6.9 Verification of end-to-end calibration
Full calibration and characterization of the PMU and measurement system are generally performed in a lab
environment where the input values can be precisely controlled.
For installation, it is important to verify measurement calibration continuity from instrumentation to the end
device or application. The verification will be based on the actual system signals. When the PMU
installation includes test switches supervising the current and voltage inputs to the PMU, the inputs can be
accurately measured, but since the real system is constantly changing, it is difficult to achieve precision.
Calibrated test equipment can be used to provide test signal sources or reference measurements for
validating the field installations and PMU calibrations. Application considerations may affect the type of
installation testing. Other factors impacting the testing include the type of installation (e.g., Y or Delta
transformer connectivity), the PMU filtering, and any instrument transformer compensation (phase or
magnitude) to offset for current and VT inaccuracies.
The principal points covered include the following:
⎯ GPS connectivity is validated, clock is locked in, and the PMU and IRIG-B output modes are
coordinated (i.e., modulated or unmodulated), and the proper UTC offset is configured on both the
PMU and the clock.
⎯ Accuracy of the measured quantities, and whether the values are reasonably correct.
⎯ Whether the measured values are in reasonable agreement with other measurements.
⎯ Whether the phase angle is consistent with other local devices for basic connectivity verification.
⎯ Whether the phasing is in agreement with other devices capable of producing PMU-type values.
⎯ Whether communication with other devices, or data streaming, is working properly.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
26
6.9.1 Correctness of measured value
A simple method of performing a check on the measured values is a comparison of the values reported by
the PMU with the voltage and current measured with a calibrated meter. If the negative and zero sequence
components are small, the positive sequence phasor magnitude is approximately the average of the rms
values of all three phases.
For a PMU with three-phase voltage and current inputs, it is best to confirm the input values when the
system is in steady-state condition and large value swings are not anticipated. All three-phase voltage
magnitudes are measured, with a calibrated rms meter in quick succession, and are then averaged. The
averaged values are then compared against PMU secondary positive sequence values displayed by the
device software or local display. The same process is generally repeated for three-phase currents. For
current and voltages, the compared input and displayed values are anticipated to be within 1% to 2%. If the
currents are very steady and midrange, the comparison should also be within 2%.
It is a good practice to also verify the phase angle relative to a known reference and anticipate less than
2 degrees deviation between measured and displayed values. When the phase imbalance is significant
(magnitude differences > 2%), it is advisable to use phase angle as well as magnitude readings to estimate
the positive sequence value.
6.9.2 Comparison with other measured values
For a dedicated and newly installed PMU, the measurements can be compared with other Intelligent
Electronic Devices (IEDs) connected to the same PT and CT circuits. Additionally, the phasor
measurements can be compared with panel meters and transducer measurements, such as those available
through the Supervisory Control And Data Acquisition (SCADA) system. For angular measurements, some
IEDs may also provide similar measurements, but state estimator (SE) values can be used as an alternative
(since the angles in the SE output are only relative, angular differences between two locations in the system
need to be compared). Keep in mind that the EMS measurement may be a single-phase measurement. The
telemetered values may not agree very closely because of variances in technology or manufacturing, or
inaccurate calibration of existing equipment. Nevertheless the comparison provides a way to check on any
wiring-related problems. It is best to take the measurements when there is sufficient amount of current
flowing through the portion of the system where the PMU values are monitored and wiring is being
confirmed.
When single-phase voltage or current measurements are available, single-phase metering is used similarly
to the three-phase without averaging. It is possible that the single-phase measured values can be further off
than expected. When only power measurements are available, the user can either combine the phasor
current with the appropriate bus voltage and compare power, or divide a power measurement by (3 × line-
to-ground voltage) to determine the current.
Comparison with another certified (calibrated and traceable) PMU can provide a good check of measured
phase angles. It is important that the comparisons are made at the same time, since loading and angles
constantly change.
6.9.3 Phasing of local signals
The phasing of the signals relative to one another (as well as the polarity of the connections) should be
checked.
It is usually beneficial to convert the phasor values into MW and MVAR and compare with other
measurements. At low values, various measurement systems may differ significantly, but anything near full
scale may provide useful comparison within a few percent.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
27
6.9.4 System signal phasing
In general within a single power company, the phase naming convention is consistent throughout the
system. For example, phase-A is well defined and should be the same at all substations. One way to
validate that phase-A at a new PMU installation agrees with other installations is by comparing the phase
angle with the nearest installed PMU.
When the PMU installations are at two adjacent stations, then the angle between them can be
approximately calculated by Equation (1):
) arcsin(
2 1V V
PX
= φ
(1)
where
P = power flow on the line
X = line reactance
V
1
and V
2
= bus voltages
Power flows from buses with leading angle to those with lagging one. The measured angle will be
reasonably close to the value calculated from the power flow equation, Equation (1).
Even if the stations are not adjacent, or the parameters are not well known, the measurement phasing can be
confirmed using a heuristic approach and applying the following rules:
⎯ Power flows from buses with leading angles to those with lagging angles. By noting the power flow
in the area between the stations being compared, if it is notably flowing from one to the other, the
angle polarity should be consistent with power flow direction.
⎯ If there is little power flow, the distance is short, or the impedance is low, the angle will be small.
Higher voltage systems have lower impedance. Multiple lines or a meshed grid have lower
impedance.
⎯ If the stations are fairly close (within 300 km), expect an angle of 30 degrees or less. If they are
more distant, it could be higher (possibly more than 90 degrees in a meshed grid).
⎯ If the power flow between stations is not clearly one direction or the other, the angle could go
either way but the total will be smaller, perhaps in the area of 15 degrees.
⎯ Large errors of 30 degrees can usually be detected if the PMU is assumed to be connected for line-
to-neutral measurements but is actually line-to-line (or vice versa).
Since the main objective here is to detect phasing errors and phase selection errors resulting in a 120 degree
phase shift, such a large discrepancy will be distinguishable from the angles estimated by power flow using
the rules previously mentioned. However, certain factors such as Y-∆ and phase shifting transformers, as
well as series capacitors, can produce results that make phase checking difficult to determine.
A similar approach is recommended when the PMU data is compared with a neighboring system. The first
step is to confirm the phase naming convention and determine the relative phase angle relation between
comparable power system phases.
With these large errors removed, comparison to SCADA-based EMS load flows can be expected to be
within 2 or 3 degrees.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
28
6.10 Communications operation
Deployment of appropriate communications technology with well-defined monitoring and maintenance
system is important in meeting life-cycle requirements when installing PMU and associated interface
equipment and a data concentrator.
Communications from the PMU to the data concentrator is a critical link. Initial checkout details depend on
the particular system. A few general observations are offered here.
Data are sent at the selected reporting rate from the PMU in packets that are precisely time tagged and
terminated with a cyclic redundancy check (CRC) or checksum. In many cases, the communications
medium provides its own packetizing with a check word. The receiving device (e.g., PDC) should use the
check word to detect corrupted data, and time tags to keep track of missing data. This information is
essential for initial system checkout as well as ongoing maintenance and repair.
After installation, corrupt and missing data must be monitored closely for a few days. For example, the
communications system may not lose more than 3 packets per hour at a data rate of 30 data points per
second, and may vary based on application, design, and tolerance requirements among factors influencing
communications requirements. At the receiving end, it is not possible to differentiate between data that has
been lost in communication and data that was never sent. Pre-installation testing must include enough
monitoring to prove that the PMU outputs data reliably to meet the application requirements. Overall
performance is also impacted by levels of redundancy in the application.
Patterns of data loss indicate different problems depending on the type of communications. For example,
alarms are commonly set to identify/measure the acceptable number of lost packets over a given time
period.
For example, with EIA-232 serial systems using audio channel (analog) or digital communications, the user
may set parameters for monitoring lost packets after a few check-word errors and a few lost packets per
hour. Lost packets may have increased activities—particularly longer periods of outage—at certain times of
day if affected by microwave channel fading (weather related). Occasional frequent outages can indicate
communications system synchronization problems or modem resynchronization due to low signal levels.
Any more than a few lost samples per hour usually indicates a communications problem that should be
resolved.
Network-based systems over analog channel communications will behave somewhat like the EIA-232.
Network over digital communications can be divided into connected types like Transmission Control
Protocol (TCP/IP) and non-connected types like User Datagram Protocol (UDP/IP). TCP/IP provides
packet ordering and retransmission in the case of missing or corrupted packets. With a good
communications system, data will be delayed occasionally, but otherwise there will be no data loss
apparent to the user. With a degraded or overloaded system, data delay may become significant enough
such that the PMU’s internal buffers overflow and TCP cannot recover lost packets, which would result in
longer outages while the system holds data waiting for recovery, and perhaps resynchronizes the data
stream.
With UDP, data is sent to a destination, but there is no built-in ordering or recovery. Sometimes ordering
and retransmission can be built into the application. UDP is used commonly with phasor measurement
systems since UDP is a one-way communication to the receiver(s), and therefore it is undelayed, could use
on average lower bandwidth than TCP, and is simple.
With a good communications system, data will rarely be delayed and data loss will be less than one packet
per hour (in practical experience, these systems have been observed to run weeks without any data loss).
The overall performance is application dependent, and the user will need to define acceptable levels of lost
data packets (i.e., occasional packet loss may be tolerated based on the application). If a general purpose
network is being used, there may be greater loss during periods of heavy use. Router or switch
misconfiguration problems can cause high loss. A good commission or maintenance testing will often
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
29
identify key sources of configuration or data errors. A sound engineered system with detailed engineering
and design records provides means for guidance of the maintenance staff to troubleshoot and expedite
repairs, and helps in maintaining the system to meet overall performance requirements.
6.11 Record keeping
A permanent record of installation details and initial tests is important. The installation will normally have
formal drawings that include wiring and physical layout, and should also include details of scaling, ratios,
and settings. A clear record of settings is essential for checkout, verification, and ongoing maintenance.
This record must include all PT/CT scaling, method of synchronization and GPS clock accuracy, PMU
calibration factors, communications settings, triggers, limits, and recording parameters. It will be used and
confirmed during the checkout outlined above. Additional information such as cable type and length (signal
attenuation and delays), CT/PT/CCVT details, line or bus voltage source, etc., can help with ease in
troubleshooting. The record of initial tests can be used as a baseline during ongoing maintenance and
provides a baseline for problem resolution. Installation test records have also been used to identify whether
a particular issue/feature was tested initially, or if a problem is introduced as the result of a change in the
system, or some kind of component failure.
7. Testing and calibration
7.1 Overview
This clause describes test and calibration procedures for PMUs used in the electric power industry to
monitor the condition of the electric power grid. The tests include those specified in IEEE Std C37.118-
2005, IEEE Std C37.118.1-2011, IEEE Std C37.118.2-2011, and IEC 61850-90-5, as appropriate. This
clause contains detailed description of testing and calibration procedures together with quality conditions
and class accuracy that test and calibration equipment must satisfy.
The purpose of this clause is to provide a basis for comprehensive testing of PMUs. The testing should
provide characterization of PMU measurements under a wide variety of conditions suitable for most
deployments. There may be some specialized deployments that will require specialized tests that are not
covered by this clause. This clause covers all tests required for conformance to the standards and some tests
that are not required but are useful during PMU algorithm design.
7.2 Objective of testing
7.2.1 Compliance
Specific testing is required to claim conformance to IEEE Std C37.118-2005, IEEE Std C37.118.1-2011,
and IEEE Std C37.118.2-2011 as specified by the standard:
a) Documentation should be provided by any vendor claiming compliance with IEEE Std C37.118.1-
2011 that should include the following information:
1) Performance class
2) Measurements that meet this class of performance
3) Test results demonstrating performance
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
30
4) Equipment settings that were used in testing
5) Environmental conditions during the testing
6) Error analysis if the verification system is based on an error analysis as called for above
b) Compliance certification is by PMU and class of performance. All signals supplied under a
particular class of service must meet the compliance requirements under the specified class of
service to qualify the PMU as compliant.
7.2.2 Manufacturing
When a PMU is manufactured, functional testing of the PMU is conducted to verify that no manufacturing
defects have caused the device to deviate from the intended design.
7.2.3 Installation and commissioning
Testing a PMU following installation is needed to verify that the PMU is configured as intended and has
been installed correctly. See Clause 6 for details.
7.2.4 Periodic maintenance
Installed PMUs are periodically tested to verify that the PMU performance has not deviated beyond
required performance limits over the course of its service history. Periodic maintenance also verifies that
the PMU configuration has not changed from the intended and documented configuration. See Clause 6 for
details.
7.3 Types of tests
As with other power system protection and control devices, PMUs must undertake various types of tests to
verify reliable and suitably accurate operation in the installed system. These tests include a thorough
conformance test, which typically includes electrical, environmental, mechanical, and performance
conformance tests for each new model of PMU. Factory acceptance tests verify that a manufactured
product performs as designed. Field acceptance tests (commissioning tests) and periodic maintenance tests
verify that PMUs are configured and perform as desired in the particular installation. These tests should
follow the requirements and procedures of the relevant international, national and industrial standards, and
guidelines where applicable. The application of a PMU has some unique requirements, as follows:
⎯ High accuracy of time synchronization of the measurement results and time tagging
⎯ Unified performance of all PMU units across an installed system
⎯ Interoperability of PMUs and other PMU system components, such as PDCs
These unique requirements demand a comprehensive set of functional performance, conformance, and
interoperability tests to be conducted, employing a well-defined test approach to verify that the
performance of an installed PMU system can meet its desired objectives and allow products from various
vendors to be integrated into the PMU system. The descriptions of different types of tests in this document
focus on the conformance and functional performance tests that are unique to PMUs and synchrophasor
measurement devices. The conformance tests are designed to verify whether the PMU meets the
requirements of IEEE Std C37.118-2005 or IEEE Std C37.118.1-2011, under steady-state, transient, and
dynamic power system conditions, and the associated interface requirements as given in
IEEE Std C37.118.2-2011 or IEC 61850-90-5. For other standard tests, such as electrical, environmental,
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
31
and mechanical tests, please refer to the relevant international, national, and industrial standards and
guidelines as listed in Clause 2 and Annex A.
7.3.1 PMU conformance tests
PMU conformance tests will be included in a comprehensive type test program and performed on each new
type of PMU devices. A complete set or a subset of conformance tests should also be performed for PMU
devices after firmware/software and hardware revisions depending on the extent of the revision. A
complete PMU test program will also include electrical, environmental and mechanical tests. Please refer to
the relevant international, national, and industrial standards and guidelines, such as IEC 61850-3 or
IEEE Std 1613 [B31], for requirements of these tests.
The description of the PMU conformance tests in this guide includes performance and interoperability
conformance tests that are unique to PMU and synchrophasor measurement devices. The conformance tests
are performed to verify whether a PMU or a synchrophasor measurement device meets the requirements of
the relevant standards, e.g., IEEE Std C37.118-2005, IEEE Std C37.118.1-2011,
IEEE Std C37.118.2-2011, and IEC 61850-90-5.
7.3.1.1 PMU message protocol conformance test
To verify the interoperability of different PMU models, a message protocol conformance test can be
performed to confirm all configuration/command/data frame implementations conform to the relevant
standards, e.g., IEEE Std C37.118-2005, IEEE Std C37.118.2-2011, and/or IEC 61850-90-5.
If manufacturer/user custom control/data message frames are defined, they should be verified according to
agreements between manufacturers and users.
7.3.1.2 PMU performance conformance test
IEEE Std C37.118-2005 defines the TVE metric and establishes the level 0 and level 1 performance
compliance requirements under steady-state conditions. IEEE Std C37.118.1-2011 establishes M-class and
P-class performance compliance requirements under steady-state as well as transient and dynamic
conditions. Consistent performance among all PMUs in an interconnected system is of great importance for
interoperability.
This subclause describes the purpose of each performance conformance test and the requirements of the
test. Performance tests are divided into steady-state and dynamic tests, latency, and time quality tests.
Steady-state tests are defined as tests where the magnitude and frequency of the test signals do not change
during an individual subtest. Otherwise, a test will be considered as a dynamic test.
The steady-state and dynamic test requirements for test signals injected at the PMU’s input terminals are
defined by IEEE Std C37.118.1-2011 or IEEE Std C37.118-2005.
7.3.1.2.1 Initial functionality verification
Prior to the full spectrum steady-state tests and/or dynamic tests, a PMU must pass the following basic
functionality tests.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
32
7.3.1.2.2 Basic functionality test
The basic functionality of a PMU device must be verified at all reporting rates under the following
conditions:
⎯ Pure sinusoidal input signal (THD < 0.1%).
⎯ Input signal frequency at rated 50 Hz and/or 60 Hz system frequency.
⎯ Balanced three-phase input signal.
⎯ Single-phase input signals.
⎯ Input signal is synchronized to produce known phase angles for each reported phasor.
⎯ PMU must report correct phasor values in the preceding tests with TVE less than 1% as specified in
IEEE Std C37.118.1-2011. Additionally, the out-of-band interference test should be run to verify
the filters at each reporting rate.
Manufacturer-implemented non-standard reporting rates may be verified according to the manufacturer’s
own specifications and/or agreement between the manufacturer and users.
7.3.1.2.3 Steady-state performance conformance test
The steady-state performance conformance tests are conducted to confirm that the accuracy of a PMU is
within the specified limits when exposed to specified steady-state operating conditions.
The steady-state performance conformance test should be performed according to the signal range and test
conditions specified in IEEE Std C37.118.1-2011 Table 3. Additionally this guide presents a test plan for a
phase or magnitude unbalance test.
It is common practice to verify manufacturer-specified steady-state performance of a PMU in extended
signal ranges and test conditions according to manufacturer’s own specifications or agreement between the
manufacturer and users.
7.3.1.2.4 PMU performance conformance levels
In order for a PMU to conform to IEEE Std C37.118.1-2011, it must be classified by performance class (M
or P). The P or M performance level is to be specified for each data reporting rate that a PMU supports.
7.3.1.3 Multifuncation PMU device performance test
When phasor measurement is one of the functions in a multifunction device, the performance test of such
devices should also verify the following:
⎯ That operation of other functions will not cause interference, degradation, and disruption of the
performance of the PMU function under all operating conditions.
⎯ That the PMU function will not cause interference, degradation, and disruption to the performance
of other functions under all operating conditions.
⎯ That the actual range of input signals for a multifuncation PMU is determined by (1) the types of
transducers dictated by its primary functions, and (2) the IEEE C37.118 required signal range for
phasor measurement.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
33
7.3.1.4 Certification of conformance
Certification of conformance should be issued only to PMUs fully conformed to the requirements of
IEEE Std C37.118-2005, IEEE Std C37.118.1-2011, and/or IEEE Std C37.118.2-2011, accompanied by a
detailed test report. Both certification of conformance and the test report will declare that all tests were
conducted in full compliance with the test requirements and test procedures specified in the standard or
proposed standard, and that the tests were performed by qualified test laboratories. The test report must
contain a detailed description of each test conducted, test equipment used, test setup description, test
conditions, and test results.
7.3.2 Interoperability test
PMUs and all other system components that communicate with each other, such as PDCs from different
manufacturers, must conform to the same protocol specified by IEEE Std C37.118.2-2011 or IEC 61850-
90-5 through protocol conformance tests. Protocol conformance is the basis for all system components to
be interoperable with each other, but protocol conformance may not be sufficient. To further confirm that
PMUs, PDCs, and other system components from different manufacturers can interoperate and work with
each other as a system, a set of tests, including performance tests, should be conducted. The tests to show
interoperability are in addition to protocol conformance tests conducted for individual PMU, PDC, or other
system component. Interoperability tests should confirm that system components work properly together
and adhere to performance requirements as required by applications. Tests for interoperability should also
be conducted for PMUs, PDCs, and other system components with custom-defined protocols, if any.
The tests to demonstrate interoperability may be conducted by a qualified independent test laboratory, or
jointly by manufacturers and user representatives. The tests may be performed in a laboratory or field
environment.
7.3.3 Factory acceptance test
Each PMU unit passes a proper factory acceptance test program to detect any manufacturing defects,
similar to substation protection, control, and networking devices. In addition to electrical, environmental,
and mechanical tests, each PMU must be verified in the factory acceptance test that it can:
⎯ Correctly receive the synchronized timing signal according to the specifications (e.g., correct
reception of GPS signals under specified signal strength).
⎯ Send/receive configuration/command/data frames according to protocols defined (e.g., send/receive
configuration as given in IEEE Std C37.118.2-2011 or initiate an IEC 61850-90-5 frame through a
61850 client).
⎯ Meet phasor measurement performance requirements at a selected set of test conditions within the
specifications of IEEE Std C37.118.1-2011.
7.3.4 Field tests
Field tests are conducted to verify the proper functionality of a PMU at its installed location after it has
been installed. The field tests include field commissioning tests and periodic maintenance tests.
7.3.4.1 Field commissioning test
A field commissioning test will be conducted after a PMU is installed.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
34
As with any other digital IEDs installed in the substations, the PMU field commissioning test must include
routine visual inspection, insulation test, wiring check, basic functionality check, etc., as required by the
relevant standards, such as 2003 NETA Acceptance Testing Specifications for protection and control
equipment.
In addition, a PMU field commissioning test will verify the following:
⎯ Phase A designation verification and confirmation.
⎯ Correct phase sequence verification.
⎯ Correct phasor magnitude measurement verification.
⎯ Correct indication of time synchronization being normal or in a locked or synchronized condition.
Also, synchronized timing source verification (e.g., GPS signal strength and reception meet the
specification).
⎯ Data and control frames sending/receiving verification.
PMUs are not usually deployed as independent measurement devices. A PMU system typically consists of
many PMUs and one or more PDCs. Once a PMU passes the above basic tests and its proper
communication with PDCs are established, further system integration tests are conducted as a part of the
PMU field commissioning tests. The system integration tests verify the following:
⎯ Proper sending/receiving data/control frames to/from PDCs
⎯ Proper registration of each PMU in the PDC database
⎯ Proper logging of PMU activities, such as on-line/off-line time, setting change, and so on
⎯ PMU status monitoring and trouble reporting
⎯ Proper handling of communications channel problems and PMU malfunctions
⎯ Communications channel throughput and PDC loading levels
7.3.4.2 Periodic maintenance test
Performance of a PMU must be checked periodically to verify that it has not been changed and
deteriorated.
7.4 Test equipment
This subclause describes the test equipment used to perform the tests. General requirements for calibration
and test equipment may be found in ANSI/NCSL Z540.3-2006 [B1]. In that standard, a quantity called the
Test Uncertainty Ratio (TUR) describes how much better calibration equipment must be than the
equipment being calibrated. The ratio is defined as ratio of the span of the tolerance of a measurement
quantity subject to calibration, to twice the 95% expanded uncertainty of the measurement process used for
calibration. The standard also requires that in order to achieve a “false acceptance” rate of less than 2% in
acceptance testing, the TUR shall be at least 4:1. This figure has been used in the synchrophasor standards.
However, recognizing that the TVE of a synchrophasor is a quantity that combines uncertainties, this
subclause applies stricter requirements on the calibration equipment for the factors that could contribute to
TVE (ANSI/NCSL Z540.3-2006 [B1]).
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
35
Test equipment will generally be in one of the following categories:
⎯ Full-featured calibration laboratory equipment: Specialized test instruments developed and
adapted specifically for PMU testing. These may be systems made from general or specially
developed hardware, hand-tuned and controlled for testing. Design and documentation of the
collective test system must demonstrate traceability to national standards. Synchrophasor standards
require uncertainty less than 25% of the allowed error. However, it is generally recommended that
uncertainty of test equipment should be less than 10% of the allowed error (e.g., if a TVE of 1% is
allowed for a PMU, the test setup used to examine the accuracy of that PMU should have traceable
uncertainty of better than 0.1%).
⎯ Standard test equipment: General purpose test equipment that can be used to test PMU functions. It
should be 4 to 10 times more accurate than the test tolerance, but this will vary depending on the
circumstance. These test setups may be used in a laboratory or field location. They must be
calibrated against a standards laboratory, which will have the full-featured test equipment and
calibration method well defined and repeatable.
In general, any offset and delays associated with the test equipment should be taken into account, for
example if a test equipment has a settle-time delay (for example, 1 cycle), then this time should be
subtracted from PMU step response time. Similarly, possible digital to analog conversion errors should be
compensated for dynamic signal calibration.
7.4.1 Time reference
Test equipment must be accurately synchronized to UTC. PMUs make measurements relative to UTC, and
the test equipment needs to use that same reference to verify measurements.
7.4.2 Steady-state signal sources
This subclause describes the requirements for the voltage and current channels of the multiphase signal
generator for steady-state PMU tests. For steady-state tests, the signals have a constant amplitude and
frequency during the test.
7.4.2.1 Steady-state magnitude
The voltage and current sources should have an amplitude uncertainty of less than 0.1% of the specified
test level, with total harmonic and noise distortion of less than 0.1% of the specified test level.
7.4.2.2 Steady-state phase
The voltage and current phases must be known relative to a cosine wave of the nominal frequency (50 Hz
or 60 Hz) synchronized to the on time second of UTC. This phase is referred to as absolute phase angle.
The absolute phase angle of the test voltage must be known for each report time (i.e., the times the PMU
assigns to its phasor measurements). The absolute phase angle must have an uncertainty of less than 1 mrad
(0.057 degrees) relative to the specified test phase, or alternatively the absolute phase angle of the test
signal must be known with an uncertainty of less than 1 mrad (0.057 degrees).
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
36
7.4.2.3 Steady-state balance
The three-phase voltages and currents of the test signal source should be known to better than 0.1%. For
balanced tests, the magnitude of the zero sequence and negative sequence should each be less than 0.3%.
7.4.2.4 Steady-state frequency and rate of change of frequency
The frequency of the source should be known with an uncertainty of less than 0.1 mHz. The ROCOF
should be less than 1 mHz/s.
7.4.2.5 Steady-state harmonics
The test signal generator must be capable of generating multiple harmonics from the 2nd though the 50th
harmonic of the nominal frequency. The magnitude of the harmonics must be at least 10% of the magnitude
of the nominal input magnitude. These signals must be generated in conjunction with a signal at nominal
magnitude and frequency.
7.4.2.6 Steady-state out-of-band interfering signals
The test signal generator must generate out-of-band interfering signals (interharmonics) with frequency
from 0.1 Hz up to the Nyquist frequency of the highest reporting rate to be tested. The magnitude of the
interharmonics must be up to 10% of the magnitude of the nominal input frequency. These signals must be
generated in conjunction with a signal at nominal magnitude and frequency.
7.4.3 Dynamic signal sources
This subclause describes the requirements for the voltage and current channels of the multiphase signal
generator for dynamic PMU tests. For dynamic tests, the amplitude or frequency of the signals will vary
during the test.
7.4.3.1 Amplitude and phase modulation
During the dynamic compliance measurement bandwidth test, the magnitude of the fundamental phasor
should have an uncertainty of less than 0.2% of the nominal magnitude and the phase uncertainty should be
less than 2 mrad (0.12 degrees). The magnitude of the modulation should have an uncertainty of less than
0.2% of the nominal magnitude. The phase of the modulation should have an uncertainty less than the
equivalent of 1 ms. The frequency of the fundamental should have an uncertainty of less than 0.5 mHz and
the rate of change of the frequency should have an uncertainty of less than 10 mHz/s.
7.4.3.2 Frequency ramp
During the frequency ramp dynamic test, the magnitude of the fundamental phasor should have an
uncertainty of less than 0.2% of the nominal magnitude, and the phase uncertainty should be less than
2 mrad (0.12 degrees). The frequency of the fundamental should have an uncertainty of less than 0.5 mHz.
The rate of change of the frequency should have an uncertainty of less than 10 mHz/s.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
37
7.4.3.3 Magnitude step
During the dynamic magnitude step test, the magnitude of the source must settle to the new magnitude with
less than 0.2% variations within one cycle of the nominal frequency. The phase of the fundamental must
settle to the same phase as before the step within 1 cycle of the nominal frequency within 2 mrad
(0.12 degrees). The magnitude of the step should have uncertainty less than 0.1% of the nominal
magnitude. The time of the step must be within 1 ms variations.
7.4.3.4 Phase step
During the dynamic phase step test, the magnitude of the source must settle to the same magnitude as
before the step with less than 0.2% variations within one cycle of the nominal frequency. The phase of the
fundamental should settle to the new phase with less than 2 mrad (0.12 degrees) variations within 1 cycle of
the nominal frequency. The magnitude of the phase step must have uncertainty less than 2 mrad
(0.12 degrees). The time of the step should be within 1 ms variations.
7.4.4 Equipment calibration
Test equipment needs to be calibrated to reference standards. Established worldwide standards are
maintained as a comparison between first level physical standards maintained by designated standards
laboratories worldwide. First level standards are compared with each other using a suitable device that can
be exchanged between laboratories. Each laboratory makes measurements before and after exchanging the
device and compares measurements with the other laboratories. Analyses of results show how close the
reference at each laboratory is to the other. A statistical comparison of the results of measurements provides
a true standard and an offset that each laboratory can use to correct their measurement to the standard.
PMU calibration has not reached the point of developing a worldwide standard or even recognized first
level laboratories to maintain such a standard. Further, there is not yet consistent agreement on the
measurements and tests that need to be performed to verify these measurements. However, many of the
measurements are straightforward within the realm of current standard measurement. For verification of the
measurements particular to PMU technology, it is necessary to depend on a reference system suitably
linked to first principles for which an estimate of accuracy has been determined by a suitable suite of tests.
This then needs to be confirmed by comparison with tests done by other laboratories using the same or
other methods. Once an acceptable reference has been developed, it can be used for calibration of other test
systems using the transfer PMU as previously described.
The National Institute of Standards and Technology (NIST) in Gaithersburg, Maryland, and several other
laboratories have developed special tests for many of the PMU tests described in this guide. This is the start
of a test verification process for a worldwide standard compliance for PMUs.
7.4.4.1 Steady-state signal calibration
The test system output can be calibrated under steady-state output conditions using standard reference
instruments. Under steady state the test system must be capable of generating three-phase voltages and
currents with a set amount of balance and at varying frequency and amplitude. Voltage and current
magnitude and frequency are well-established measurements. Test system output should be calibrated at
suitable amplitude and frequency points using certified test instrumentation. PMU testing requires that
voltage and current signal phase angles be determined with respect to a UTC time reference. The following
paragraphs describe two possible methods to perform this calibration.
Both calibration methods use a high-speed/high-resolution oscilloscope to compare the phase of a
synchronized oscillator with a UTC synchronized clock signal, typically a 1 PPS signal. Since the rate of
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
38
change of a nominal frequency (50 Hz or 60 Hz) signal is not fast enough to easily align with the 1 PPS
signal, this first method uses a reference PMU that is capable of measuring the phase angle of signals with
frequencies significantly higher than power line frequencies. It also requires that the signal generator will
put out synchronized waveforms at much higher frequencies than the normal power signal. The second
method is not as precise, but does not require a reference high frequency PMU nor a specialized multi-
frequency signal source. These procedures take advantage of the fact it is easy to synchronize an
oscilloscope (or counter) with a 1 PPS signal, and the time of a zero crossing of a sine wave can be more
precisely determined than the time of a peak.
For the first method, assume the bandwidth of the reference PMU is 500 kHz. For each phase of the
reference PMU, set the signal generator to 30 kHz, then adjust the phase of the signal so the positive-going
zero crossing of the signal is aligned with the positive edge of the 1 PPS signal. The signal phase angle
measured by the reference PMU should be –90 degrees. Note the error, the difference between the result of
the measurement and the known value of –90 degrees. Readjust the phase of the signal so the negative-
going zero crossing of the signal is aligned with the positive edge of the 1 PPS signal. The signal phase
angle measured by the reference PMU should be 90 degrees. If the error observed is not the negative of the
previous error, the results indicate there is an offset in the signal, possibly due to dc in the signal. There
could also be another form of waveform distortion. Taking half the difference between these two test
results, one can convert the angle to a time delay of the 30 kHz signal.
Repeat the above time delay determinations at 10 kHz, 3 kHz, 1 kHz, and 500 Hz. Plot these time delays to
see if they fall on a straight line. Fit a line to the lower part of the data that is straight and project the time
delay above power line frequencies. Apply this as a correction to the signal phase angles measured. Repeat
the above procedure to demonstrate that with this correction the measured time delay at 30 kHz is now
negligible. Use this data to analyze the uncertainty of the phase angles measured by the reference PMU.
NOTE—The uncertainty in the time delay estimates for 10 kHz and 30 kHz will be the smallest. The lower frequency
measurements will show more scatter, but look to see that there is no significant slope to this data.
The second method is shown in Figure 6. The test signal output is set to a ±90º phase angle output at the
nominal system frequency. By the synchrophasor standard, A-phase will have a positive zero crossing at a
–90º phase angle and will have a negative zero crossing at a +90º phase angle. Use the synchronized 1 PPS
signal from a GPS source to trigger the trace on the oscilloscope. The signal source should cross 0 V at the
trigger point. At nominal system frequency (50 Hz or 60 Hz), the zero-crossing does not change. For off-
nominal frequencies, the angle changes constantly and may not align at the 1 PPS. However at exact integer
frequencies off nominal (e.g., 49, 51, 59, 61) the zero crossing will be aligned with the 1 PPS, making this
method of synchronization useable for checking phase over a band of frequencies. Use positive and
negative zero crossings to average out dc offset as previously described to improve accuracy.
Figure 6 —Phase calibration of reference PMU with the 1 PPS clock signal
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
39
The principal problem with this method is that the zero crossing needs to be measured within a
microsecond, and the slope of a 100 V
peak
60 Hz wave is 37 700 V/s or 0.037 V/μs. With the input set to
input a 100 V signal, determining a zero crossing to within 1 μs is not possible. If the oscilloscope limits
input overload without distortion, the input can be amplified to show the waveform at a millivolt level so
that the zero crossing can be clearly determined. Alternatively, a diode limiter can be used to reduce the
voltage signal to a low level so that the 0 crossings can be seen clearly without overdriving the front end.
For observing the 0 crossings of the current signals, use a low resistance non-inductive shunt. In all cases,
care is required to prevent ground loops from distorting the signal.
It is important to note that oscilloscopes that have a low-resolution digital front end (7 or 8 bits) do not
allow an accurate location of the positive zero crossing. Any dc offset will also change the phase angle
based on zero crossing. With 8 bits full scale, the zero location could be off by 1 count out of ±127. If the
full scale is ±100 V, this is about a 0.79 V error that is equivalent to about 10 μs or about 0.22 degrees at
60 Hz. (1% TVE = 0.57 deg if magnitude is perfect.) However, if the full scale is reduced to ±1 V, this is
about a 0.0079 V error that is equivalent to about 0.1 μs or about 0.0022 degrees, which is insignificant.
7.4.4.2 Dynamic signal calibration
A stable and accurate signal source is required for dynamic signal production. It may not be possible to
measure the signal with standard calibration instruments during dynamic signal production. It is necessary
to rely on other confirmation methods such as analysis of the signal production equipment or a reference
PMU with known analysis capability.
The first method relies on analysis of the test equipment being used and the test methodology. In most
cases, dynamic signals will be produced by a test set that converts signals stored in a digital file to analog
signals using D/A converters. In this case, the accuracy of these reproduced signals can be verified by the
following:
⎯ Calibration of the D/A converters at magnitudes that cover the range that will be required
⎯ Calibration of the D/A output clock relative to time and frequency references
⎯ Characterization of the output frequency response allowed by D/A output filters
⎯ Accounting for loading effects such as cabling and test unit burdens
This method depends on careful analysis of all these effects, and should provide sufficient accuracy for
performing dynamic tests. Test equipment designed for performing dynamic test signals should have
established performance characteristics and is a good choice for these measurements. A secondary method
of monitoring the test signals should also be used, such as by recording the waveforms or comparing
measurement results with a PMU with previously established performance.
A second method of dynamic signal calibration uses both a stable signals source and a reference PMU to
determine the actual values of the phasors, frequency, and ROCOF. This reference PMU can be made up of
a multichannel sampling system and appropriate software to calculate these quantities. This system must be
able to receive clock signals to determine the absolute time of each sample. Use of the 1 PPS signal and a 1
MHz or 10 MHz clock signal to synchronize the sampling to UTC is recommended. Alternatively the
system could read IRIG-B or similar timing codes to synchronize to UTC. The reference PMU has to be
itself confirmed using analytical methods to be sure it accurately determines the correct phasor value from
the analog signals, both for steady-state and dynamically changing signals.
Modulated waveforms can be confirmed to be accurate by fitting synchronized samples of the signals to an
appropriate model. Stenbakken et al. [B59] describe two analysis models that fit the data. The first is the
Taylor expansion model that fits modulation frequencies from 0.1 Hz to over 1 Hz. The second is the three
waveform model that fits modulation frequencies from 1 Hz to over 70 Hz.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
40
7.4.5 PMU data communication and analysis
PMU data communication includes transmitting PMU measurements and receiving PMU commands.
Various transmission methods, such as EIA-232, EIA-485, and Ethernet 10/100BaseT, and various
transmission protocols, such as TCP/IP and UDP/IP, are well-known methods. The test system needs to
have various communication capabilities available to communicate with different PMUs. At a minimum,
the test system should be able to send commands and receive data using the following:
⎯ EIA-232, 38.4 kbps, commanded or spontaneous
⎯ 10/100BaseT using—TCP, TCP/UDP, or UDP (multicast, unicast), or spontaneous UDP (multicast,
unicast)
PMUs do not necessarily support all communication modes, so the test system must be able to be set up for
the particular DUT.
7.4.6 Automated features
Test time can be substantially reduced if testing is automated. Test equipment could provide options for
performing automated testing (batch mode). The automation of testing should include applying the signals,
waiting for the PMU measurement to settle, and measuring the responses. The results need to be stored in a
convenient way for later review.
7.5 Methods for performing the tests
This subclause provides recommendations for how the tests can be performed. This subclause is ordered
similarly to Clause 5 of IEEE Std C37.118.1-2011, which specifies methods for determining total vector
error (TVE), frequency error (FE), and rate of change of frequency (ROCOF) error (RFE) used to assess
PMU performance in the various tests. This subclause goes on to present recommended test procedures for
steady-state compliance, dynamic compliance, and reporting latency.
7.5.1 Total vector error, frequency error, and rate of change of frequency error
7.5.1.1 Total vector error
The TVE definition that IEEE Std C37.118.1-2011 describes can be rewritten as shown in Equation (2):
true
true Measured
V
V V
TVE r
r r
−
≡
(2)
where
true
V
r
is the true phasor
Measured
V
r
is the PMU measured phasor
Notice that the measured phasor can be a voltage phasor or a current phasor. Figure 7 shows how TVE,
magnitude error and angle error are interrelated. The plot in this figure shows the TVE calculated for angle
errors from –0.5 to 0.5 degrees and magnitude errors of 0%, 0.1%, 0.2%, and 0.3%. As expected, the TVE
is greater than the magnitude error except for the case that the phase error is zero. For even a 0.5 degree
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
41
phase error, the TVE increases considerably. For all practical purposes, the TVE is dominated by phase
errors for phase errors greater than 0.5 degree.
IEEE Std C37.118.1-2011 specifies the minimum amount of time to take measurements: 5 s unless
otherwise noted in the individual test. TVE should be calculated for each measurement during this time.
The error reported for the test should be the maximum of these error values.
Figure 7 —TVE in percentage for angle errors of –0.5 degrees to 0.5 degrees and
magnitude errors of 0%, 0.1%, 0.2% and 0.3%
7.5.1.2 Frequency error and rate of change of frequency error
FE and RFE measurements are tested for accuracy using the following error definitions:
FE = Frequency error = |f
true
– f
measured
| = |Δf
true
– Δf
measured
|
RFE = ROCOF error = |(df/dt)
true
– (df/dt)
measured
|
where
The measured and true values are taken at the same instant of time
Δf
true
is the deviation of the true frequency from the nominal frequency
Δf
measured
is the deviation of the PMU measured frequency from the nominal frequency
(df/dt)
true
is the true rate of change of the frequency
(df/dt)
measured
is the PMU measured rate of change of frequency
7.5.1.3 Reporting rates and reporting times
The PMU should be configured to operate at each of the required reporting rates as specified by
IEEE Std C37.118.1-2011. With near nominal three-phase input voltage and current signals, the TVE of the
PMU should be measured. The reporting time stamps should be examined to see that the fractional second
values are integer multiples of the reciprocal of the reporting rate 1/F
s
(rounded or truncated to the
appropriate significant figure). Thus, if F
s
is 20 per second, the time stamps should occur at intervals of
exactly 1/20 s, or 50 ms.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
42
7.5.2 PMU settling time
During PMU testing, the PMU may be subjected to nonsystem like changes in the input signals. Some
settling time will be required before test data are taken. This settling time will vary from PMU to PMU, for
different PMU settings, and on the type of signal presented by the test system. For some test systems the
nominal test signal will be applied to the PMU before each test. Changes to the new parameter will be a
change in signal magnitude, frequency, distortion, etc. For other test systems the signals may be set to zero
before each test and the new signals applied as a step. For each such test condition transition, the settling
time of the PMU must be estimated.
This settling time can be provided by the manufacturer. Otherwise it can be determined by selective
transition measurements on the PMU. The PMU settling time could be up to 2 s to 4 s or as little as a few
reports.
7.5.3 Steady-state compliance test plans
This subclause describes how to perform tests to assess steady-state compliance of a PMU. Suggestions for
step-by-step methods to perform the tests (test plans) follow.
7.5.3.1 Steady-state compliance
According to IEEE Std C37.118.1-2011, steady-state compliance should be confirmed by comparing the
phasor, frequency, and ROCOF estimates obtained under steady-state conditions to the corresponding input
values and calculating TVE, FE, and RFE as defined above. Steady-state conditions are where X
m
, ω, and
φ of the test signal, and all other influence quantities, other than the quantity being tested, are fixed for the
period of the measurement. Note that for off-nominal frequencies, the measured phase angle will change at
a rate proportional to the difference in frequency between the applied signal and the nominal frequency,
50 Hz or 60 Hz.
The steady-state compliance tests are specified by IEEE Std C37.118.1-2011. Influence quantities,
minimum range of influence quantities, and limits for performance of P and M class PMUs are found in the
standard and will not be repeated in this guide. Testers must refer to the standard for the latest required
limits.
The influence quantities are as follows:
⎯ Signal frequency (range)
⎯ Signal magnitude: voltage
⎯ Signal magnitude: current
⎯ Phase angle
⎯ Harmonic distortion
⎯ Out-of-band interfering signals
For each parameter of an influence quantity (e.g., frequency tested in frequency variation tests) determine
the magnitude error, phase error, FE, and RFE at each time stamp. Calculate the TVE at each time stamp.
Over the time specified for the test, calculate the maximum, minimum, mean, and standard deviation of
these errors. These statistics can be plotted versus the parameter values tested. Use the largest maximum
error for the range of parameters to compare with the test requirements.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
43
7.5.3.1.1 Frequency variation compliance
For certification of compliance, IEEE Std C37.118.1-2011 requires that the signal frequency compliance
tests be performed over the given ranges and meet the given requirements at three temperatures:
T = nominal (~23 ºC), T = 0 ºC, and T = 50 ºC.
This is a series of subtests where magnitudes are held at nominal and frequency is changed by 0.1 Hz
between each test. The first subtest is at nominal frequency minus the bandwidth range limit, and the last
test is at nominal frequency plus the bandwidth range limit. These bandwidths are listed in Table 7 in the
column “Bandwidth” for each reporting rate, F
s
.
See IEEE Std C37.118.1-2011 Table 3 and Table 4 for the required TVE, FE, and RFE limits for the given
F
s
, where F
s
is the reporting rate.
a) Begin at the frequency range: nominal minus the range limit.
b) Wait for the system to settle.
c) Capture the PMU output for 5 s.
d) Calculate the maximum TVE, FE, and RFE.
e) Increase the frequency by 0.1 Hz.
f) Repeat steps b) through e) until the upper frequency range limit is reached.
g) Compare the results to the class limits in IEEE Std C37.118.1-2011.
h) Place the PMU into a temperature chamber set to 0 °C. Allow PMU to reach temperature
equilibrium within 1 °C.
i) Repeat steps a) through g).
j) Set temperature chamber to 50 °C. Allow PMU to reach temperature equilibrium within 1 °C.
k) Repeat steps a) through g).
7.5.3.1.2 Steady-state magnitude compliance: voltage and current
For both current and voltage input ranges, apply steady-state, nominal frequency (50 Hz or 60 Hz),
balanced three-phase inputs. Compare the measurement with the input. The test signals must be known
with uncertainty no greater than 0.1%. Nominal values are determined by the application and are typically
70 V rms or 120 V rms for voltage, and 5 A rms for current. Depending on the level or class, the PMU
must operate over a range of voltages and currents, usually given as magnitudes relative to the nominal
values. See IEEE Std C37.118.1-2011 Table 3 and Table 4 for magnitude values and test limits.
a) Begin at the low magnitude limits.
b) Wait for the system to settle (see 7.5.2).
c) Capture the PMU output for 5 s.
d) Calculate the maximum TVE, FE, and RFE.
e) Increase the input magnitudes by 10% of the nominal value (not a percentage of the actual value).
f) Repeat steps b) through e) until the upper magnitude limits are reached.
g) Compare the results to the class limits in IEEE Std C37.118.1-2011.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
44
7.5.3.1.3 Phase angle compliance
For both current and voltage input ranges, the phase test can be performed with f
in
(input frequency) offset
and | f
in
– f
0
| < 0.25 Hz to provide a slowly varying phase. The phase errors of the phasors can be
determined over several cycles of the absolute phase angle. For example, the frequency can be set to
0.12 Hz above the nominal frequency (50.12 Hz or 60.12 Hz) and measurements taken over 50 s. These
data will cover six cycles of the absolute angle. If the phase errors are plotted versus the true phase-A phase
angles, the plot will show the angle errors versus the true phase angle of ±180º. During this test the
maximum TVE, FE, and RFE must remain within the required limit.
7.5.3.1.4 Response to harmonic distortion signals
Response to harmonic distortion is specified for 2nd harmonic to 50th harmonic individually.
a) Begin at nominal levels for all influence quantities.
b) Inject into the voltage and current inputs harmonics starting with the second harmonic with the
magnitude set to the level specified by IEEE Std C37.118.1-2011 for Class P or M PMUs.
c) Wait for the system to settle (see 7.5.2).
d) Capture the PMU output for 5 s.
e) Calculate the maximum TVE, FE, and RFE.
f) Change to injecting the next harmonic.
g) Repeat steps b) through f) until 50 harmonics have been tested.
h) Compare the results to the class limits in IEEE Std C37.118.1-2011.
Optional: Repeat this test for input frequencies other than nominal such as f
0
± 1 Hz, within the rated input
frequency range of the PMU.
Test input: Both voltage and current.
7.5.3.1.5 Response to out-of-band interfering signals
There are no requirements for Class P response to out-of-band interfering signals. This test is for Class M
compliance only.
Out-of-band compliance (M class only) is to be checked with the frequency of the dominating signal
components within the nominal system frequency ±10% of the Nyquist frequency of the reporting
frequency (F
s
). See Table 7. The in band is defined as the frequency range from the nominal frequency
minus half the reporting rate, f
0
– F
s
/2, to the nominal frequency plus half the reporting rate, f
0
+ F
s
/2.
Out-of-band compliance can be confirmed by adding a single frequency sinusoid to the fundamental power
signal at the required amplitude level and varying the frequency of this signal over the out-of-band range
from below the in band (at least down to 10 Hz) to the second harmonic (2 × f
0
) for reporting rates up to
50 per second or 60 per second. For higher reporting rates, such as 100 per second or 120 per second, start
at the upper end of the in-band and inject frequencies up to f
0
higher. Thus, for reporting rates of 100 per
second and 120 per second the testing will go from 2 × f
0
to 3 × f
0
. If the positive sequence measurement is
being tested, the interfering signal must be positive sequence. This test should be repeated three times with
input signal (f
in
) frequency at nominal (f
0
), at f
in
= f
0
+ 2/3 the signal frequency bandwidth and at
f
in
= f
0
– 2/3 the signal frequency bandwidth. Table 7 shows the bandwidth and the in-band range for each
reporting rate, and nominal frequency. The in-band range is the range in which interfering signals will not
be injected.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
45
Table 7 —Out-of-band interference test with IEEE Std C37.118.1-2011
60 Hz nominal In band
F
s
Bandwidth
(Hz)
10% Nyquist
of F
s
(Hz)
F
s
/2
(Hz)
F
nom
= 60
(Hz)
10 or less ±2 ±0.5 5 55 to 65
12 ±2.4 ±0.6 6 54 to 66
15 ±3 ±0.75 7.5 52.5 to 67.5
20 ±4 ±1 10 50 to 70
30 ±5 ±1.5 15 45 to 75
60 ±5 ±3 30 30 to 90
50 Hz nominal In band
F
s
Bandwidth
(Hz)
10% Nyquist
of F
s
(Hz)
F
s
/2
(Hz)
F
nom
= 50
(Hz)
10 or less ±2 ±0.5 5 45 to 55
25 ±5 ±1.25 12.5 37.5 to 62.5
50 ±5 ±2.5 25 25 to 75
Interfering signals are not injected at the in-band frequencies above.
Apply a steady-state, balanced, three-phase signal to both current and voltage inputs, at nominal steady-
state magnitude and nominal system frequency (50 Hz or 60 Hz). Inject into the voltage and current inputs
a positive-sequence interharmonic at frequency f
i
where, as shown in Equation (3):
|f
i
– f
0
| ≥ F
s
/2 (3)
where
F
s
is the reporting rate
f
0
is the nominal frequency
f
i
= f
interharmonic
is the frequency of the interfering signal frequency
a) Add an interharmonic signal at f
i
= f
0
– F
s
/2 at 10% nominal magnitude.
b) Wait for the system to settle.
c) Capture the PMU output for 5 s.
d) Calculate the maximum TVE, FE, and RFE.
e) Run steps b) through d) repeatedly, decreasing the interharmonic frequency logarithmically until it
reaches 10 Hz:
⎯ For each test run, the amount that the interharmonic frequency decreases should increase
logarithmically, thus providing many tests near f
i
= f
0
– F
s
/2 and fewer tests further away from
f
i
= f
0
– F
s
/2. For example, the first decrease should be 0.1 Hz (f
0
– F
s
/2 – 0.1), the next
decrease 0.2 Hz (f
0
– F
s
/2 – 0.2), the third decrease 0.4 Hz (f
0
– F
s
/2 – 0.4), then 0.8 Hz
(f
0
– F
s
/2 – 0.8) until an interharmonic frequency below 10 Hz is reached. For the last
frequency use 10 Hz rather than the frequency below 10 Hz.
f) Add an interharmonic signal at f
i
= f
0
+ F
s
/2 at 10% nominal magnitude.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
46
g) Run steps b) through e) repeatedly, increasing the interharmonic frequency logarithmically until it
reaches 2 × f
0
:
⎯ For each test run, the amount that the interharmonic frequency increases should increase
logarithmically, thus providing many tests near f
i
= f
0
+ F
s
/2 and fewer tests further away from
f
i
= f
0
+ F
s
/2. For example, the first increase should be 0.1 Hz (f
0
+ F
s
/2 + 0.1), the next increase
0.2 Hz (f
0
+ F
s
/2 + 0.2), the third increase 0.4 Hz (f
0
+ F
s
/2 + 0.4), then 0.8 Hz (f
0
+ F
s
/2 + 0.8),
until an interharmonic frequency above 2 × f
0
is reached. For the last frequency use 2 × f
0
rather
than the frequency above 2 × f
0
.
h) Compare the results to the class limits in IEEE Std C37.118.1-2011.
i) Run all steps a) through h) two more times with the input signal frequency at nominal frequency
plus and minus 10% of F
s
/2 (the Nyquist frequency of the reporting rate, F
s
).
Test input: Both voltage and current.
NOTE—Out-of-band interfering signal testing requires many test runs. Increasing the increment of interharmonic
frequency change logarithmically rather than at small, constant increments reduces the total number of individual test
runs while ensuring that many tests are run near the bandwidth limits.
7.5.3.2 Unbalanced three-phase signals
Unbalanced tests are not required by IEEE Std C37.118.1-2011. Since unbalance only applies to sequence
components of a polyphase system, it is not included in the standard where the requirements apply to both
single-phase and sequence components. It is generally agreed that unbalanced testing is good to do but is
not required for certification. This subclause describes methods for performing unbalanced three-phase
signal test.
7.5.3.2.1 Response to unbalanced signals, phase, and magnitude
For both current and voltage input ranges, apply steady-state, nominal frequency input with two phases at
nominal magnitude and the third being varied as nominal ±10%, and nominal ±20%. Compare the
measurement magnitude and phase angle with that expected, and compute the TVE for each phasor, the
magnitude error for each phasor, and the phase error for each phasor. For example, apply the same test at
nominal frequency minus 1 Hz. Measure the magnitude of any modulation of the TVE, magnitude error,
and phase error.
Test imbalances: Magnitude ±10%, and ±20% on one phase only, and phase ±60°, ±40°,
and ±20° on one phase only
Test frequencies: 49 Hz, 50 Hz (nominal frequency 50 Hz)
Test frequencies: 59 Hz, 60 Hz (nominal frequency 60 Hz)
Test input: Both voltage and current
7.5.4 Dynamic compliance tests
This subclause describes how to perform tests to assess the dynamic performance of a PMU. These are tests
for signals with time varying magnitudes and/or frequencies.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
47
7.5.4.1 Dynamic compliance—measurement bandwidth
The measurement bandwidth is determined by applying sinusoidal amplitude and phase modulated signals
to the PMU. This can be done by applying the modulation applied to balanced three-phase input signals
(voltages and currents) with simultaneous modulation applied to signal amplitudes and phase angles.
Mathematically the input signals may be represented by Equation (4), Equation (5), and Equation (6):
X
a
= X
m
[1 + k
x
cos(ωt)] × cos [ω
0
t + k
a
cos(ωt – π)] (4)
X
b
= X
m
[1 + k
x
cos(ωt)] × cos [ω
0
t – 2π/3 + k
a
cos(ωt – π)] (5)
X
c
= X
m
[1 + k
x
cos(ωt)] × cos [ω
0
t + 2π/3+k
a
cos(ωt – π)] (6)
where
X
m
is the peak amplitude of the input signal
ω
0
is the nominal power system frequency in radians per s (2 × π × f
0
)
ω is the modulation frequency in radians per s
f
m
= ω/2π is the modulation frequency in Hz
k
x
is the amplitude modulation factor
k
a
is the phase angle modulation factor
t is time
k
x
= 0.1 and k
a
= 0.1 for combined phase and amplitude modulation tests.
k
x
= 0 and k
a
= 0.1 for phase modulation tests.
For class P and M, the modulation frequency should range as specified in IEEE Std C37.118.1-2011.
a) Begin with combined phase and amplitude modulated input at ω = 0.1 Hz.
b) Wait for the system to settle (see 7.5.2).
c) Capture the PMU output for at least 2 full cycles of modulation.
d) Calculate the maximum TVE, FE, and RFE.
e) Increase the modulation frequency ω by 0.2 Hz.
f) Repeat steps b) through e) until the upper frequency range limit is reached.
g) Compare the results to the class limits in IEEE Std C37.118.1-2011.
h) Repeat the entire test for phase modulation only.
IEEE Std C37.118.1-2011 only prescribes these combined amplitude-phase and phase only tests. However,
an amplitude only test may be useful to observe if there is an effect on the frequency measurement due to
processing issues in the PMU. Adding this test is an option to consider.
7.5.4.2 Dynamic compliance—performance during ramp of system frequency
PMU performance during system frequency change is tested with a ramp of frequency applied to balanced
three-phase input signals (voltages and currents).
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
48
⎯ The ramp rate is specified by IEEE Std C37.118.1-2011.
⎯ The ramp range for class P and M class is specified by IEEE Std C37.118.1-2011.
For F
s
= 12 FPS, ramp range will be ±2 1/3 (two and one-third) Hz to allow for an integer number of
samples in the result.
Note that the allowed TVE may be exceeded during a “transition time” when a change is made to the
applied ROCOF. A time period of two sample periods after any change in the test ROCOF will not be
included in the TVE calculation for the test. An example of this transition time is when the ROCOF is
changed from a 0 Hz/s value to a ramp at 1 Hz/s. The test should not include frequency discontinuities
(frequency steps or ROCOF steps).
Mathematically, the input signals may be represented by Equation (7), Equation (8), and Equation (9):
X
a
= X
m
cos [ω
0
t + πR
f
t
2
] (7)
X
b
= X
m
cos [ω
0
t – 2π/3 + πR
f
t
2
] (8)
X
c
= X
m
cos [ω
0
t + 2π/3 + πR
f
t
2
] (9)
where
X
m
is the amplitude of the input signal
ω
0
is the nominal power system frequency
R
f
(= df/dt) is the frequency ramp rate (fixed value in this equation)
a) Begin with input at nominal magnitude and lower frequency range.
b) Wait for the system to settle (see 7.5.2).
c) Begin ramping the frequency with a positive ramp rate specified by IEEE Std C37.118.1-2011.
d) Calculate the maximum TVE, FE, and RFE excluding data from the first two sample periods
following the change in ramp rate.
e) Hold the frequency constant for at least two sample periods then begin ramping the frequency at the
negative ramp rate.
f) Calculate the maximum TVE, FE, and RFE excluding data from the first two sample periods
following the change in ramp rate.
g) Compare the results to the class limits in IEEE Std C37.118.1-2011.
7.5.4.3 Dynamic compliance—performance during step changes in amplitude or phase
A recommended practice for measuring a PMU’s response to step changes in input is to run a series of step
tests where the step is displaced in time relative to the PMU report time. The results are recombined to
trace a complete response curve. For each test in the series, the step is applied a fraction of the reporting
period later in time relative to the beginning of a reporting cycle. In each test, the influence quantity
(amplitude or phase) is stepped, the response data are gathered, and the influence quantity is returned to its
previous value in preparation for the next test in the series.
At the end of the series of tests, the data from all tests are indexed and overlaid to create a smooth curve
approximating the step response of the PMU. From the curve, delay time compensation error (called
response delay in IEEE Std C37.118.1-2011), response time, and overshoot can be calculated. This
technique is called equivalent time sampling.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
49
A total number N tests should be run, where N = 10. The first step occurs at the beginning of a reporting
cycle (for example at the same time as the 1 PPS timing signal). PMU data are gathered, then the input
returned to nominal. The next step occurs 1/N times the reporting period after the beginning of a reporting
cycle, the subsequent test 2/N, then 3/N, etc., until all N tests are complete.
When the PMU data are indexed and overlaid, the result will be a curve similar to the line shown in
Figure 8, which shows the curve as well as a set of values that correspond to a step input that is aligned
with a time stamp for t = 0. The delay time compensation error (called delay time in
IEEE Std C37.118.1-2011) is non-zero if the value that corresponds to 50% of the step does not occur at the
zero time stamp. The time difference between the 50% number for the time stamp and the t = 0 time stamp
is the compensation error. In Figure 8, the 50% value is given by a step that occurred slightly before the
t = 0 test.
The response time is the period of time from the time that TVE, FE, or RFE first exceed their specified
limits to the time that they return to and remain below its specified limit.
Overshoot and undershoot have the same meanings as in an analog system and will not be dealt with
further here.
Figure 8 —Sample of step change at t = 0, illustrating response time,
delay time compensation error, and overshoot measurements
[Note that maximum overshoot may be over or under the final value, and the delay time
compensation error may be positive (response after step) or negative (response before step)]
a) For the first test, let n = 0.
b) Begin with three-phase balanced input at nominal amplitude and frequency.
c) At the beginning of a reporting cycle plus n / (N × F
s
) (i.e., n × reporting period / N)) step the
influence quantity (amplitude or phase) by the amount specified in IEEE Std C37.118.1-2011.
d) Gather the PMU data, return the influence quantity to nominal and wait for the PMU to settle.
e) Increment n by one (n = n + 1), then repeat steps c) through d) until n = N – 1.
f) Index and overlay the PMU data to obtain a smooth response curve.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
50
The measurement delay time is the time from the step in influence quantity until the point at which the
response curve exceeds 50% of the difference between the nominal value and the value after the step. Note
that delay time compensation error can be positive (after the step, as shown in Figure 8) or negative (before
the step) depending on how the PMU applies the time stamp.
The response times are the times between TVE, FE, and RFE exceeding the limits and the time where TVE,
FE, and RFE reach and remain within the limits as specified in IEEE Std C37.118.1-2011. Compare the
measurement delay and response times to the limits in IEEE Std C37.118.1-2011.
7.5.4.4 Measurement reporting latency compliance testing
Designs of real-time systems such as control loops rely on measured latency information. Precise
measurement of this information is crucial for the synchrophasor-based system design. In addition,
additional latencies introduced through communications and intermediate processing (PDC) of the data
records must be taken into account for control applications.
The power grid is analog and runs in continuous time. The PMU discretizes continuous time and performs
the power grid measurements at well-defined time instants, which are determined by the reporting rate. An
“event” on the power grid can be any kind of dynamic condition. Examples of events are the time of an
amplitude or phase step change or the time when a dynamically modulated signal has a certain phase or
magnitude. Power system events happen “freely” and are not necessarily synchronized with the PMU
reporting time. The difference between event time and the discrete measurement time is found in all
systems with digital signal processing. This difference is well understood and can be easily accounted for
by using control system theory.
IEEE Std C37.118.1-2011 5.3.4 provides latency definition. The standard may be interpreted as describing
two latencies. The first is defined as the time between the occurrence of an event at the PMU input and the
time the event is reported in the data at the output port of the PMU. It is called latency in measurement
reporting (LMR). A second latency, the PMU reporting latency (LRP) is defined as the maximum time
interval between the data report time as indicated by the data time stamp and the time when the data
becomes available at the PMU output.
Maximum permissible latencies for M and P class PMUs are specified in Table 12 of IEEE Std C37.118.1-
2011. Whether LRP or the more stringent maximum LMR is used for comparison with Table 12 depends
on how the standard is interpreted and should be noted in the specifications for the PMU being tested.
PMUs with a reporting rate synchronous (50 Hz or 60 Hz) or higher will likely have difficulty in meeting
these requirements (particularly maximum LMR). While higher rates are described in the standard,
sufficient details do not exist therein to adequately determine compliance. The user of this guide is advised
to exercise caution when applying the synchrophasor standard at higher rates until proper test and
acceptance criteria are developed. In the event changes are made to clarify these issues, the user is further
advised to use the most recent revision of the standard.
Since a PMU may rely on a microprocessor running a real-time operating system, the measurement
reporting latency may not be consistent from report to report. IEEE Std C37.118.1-2011 requires
measurement of the reporting latency over at least 1000 consecutive reports. The latency is the largest of
these values.
It is recommended that the maximum uncertainty of the reporting latency measurement be 100 μs.
a) No special input is required.
b) Within 100 μs of the report arriving at the output port of the PMU, capture the time.
c) Subtract the time in the time stamp of the report from the time captured in item b).
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
51
d) Repeat steps b) and c) for 1000 consecutive reports.
e) The largest of these 1000 time differences is the LRP.
f) Add one reporting period to the LRP determined in step e) to get the maximum LMR. Decide which
of the two latency definitions will be reported and compare it to the limits in IEEE Std C37.118.1-
2011 Table 12. Report the result along with the latency definition used (LPR or maximum LMR).
7.5.4.4.1 Additional information
The reporting time requirement can be understood by referring to Figure 9, which illustrates the fact that
each individual synchrophasor measurement represents an average value of the PMU input signal observed
over a finite time interval, shown as a measurement window. The length of the measurement window varies
and is determined by the PMU performance class (P or M), and additionally in case of M class, the selected
reporting rate. In the example of Figure 9, the interval is 10 cycles.
Figure 9 —Input signal with step change in magnitude
Since IEEE Std C37.118.1-2011 is a performance-driven standard, the actual position of the window with
respect to the time stamp is not explicitly defined. It is contained within the definition of the TVE concept,
5.3, and Annex C of IEEE Std C37.118.1-2011. The intent of IEEE Std C37.118.1-2011 is that the
performance requirements be met and the PMU manufacturer should determine how the data is processed
to achieve that result. In the examples described in IEEE Std C37.118.1-2011, Annex C, the measurement
window is shown centered on the time stamp. That arrangement is also indicated in Figure 9, which shows
two windows centered on the reporting times 28.1 and 28.2.
Of course, the value being calculated from the input samples is not available until some short time after the
last sample within the time window. There is therefore a short delay before the number is transmitted. This
is shown in Figure 10, which illustrates the LRP.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
52
Figure 10 —Timeline for the event at 28.133
IEEE Std C37.118.1-2011 defines latency measurements in 5.3.4. Two different latencies are defined in the
two paragraphs in this subclause:
Latency in measurement reporting (LMR) defined in the first paragraph is the “time delay from when an
event occurs on the power system to the time that it is reported in data.”
PMU reporting latency (LRP) is the difference between the PMU time stamp and the actual time at which
the record with that time stamp is output (“denoted by the first transition of the first bit of the output
message”).
The two definitions are illustrated in Figure 11. LRP depends only on the time interval from the time tag to
the last sample in the measurement window (generally ½ the length of the window) and the time to process
and emit the measurement. LMR is the measured delay from an event and when it is reported. It is not the
maximum or minimum of such values. It depends upon where within the reporting interval the event occurs
in addition to the time to make the measurement and send it. This will include filtering and windowing
parameters. If we agree to represent an arbitrary event with a step change in value, the time tag is required
to be within ¼ of a reporting period of an event (IEEE Std C37.118.1-2011 5.5.8). Then the minimum
difference between LRP and LMR is ¼ of the reporting period. Detailed analysis reveals that the maximum
value of LMR is one reporting interval greater than LRP, that is, LMR = LRP + 1/Fs. It is called maximum
LMR.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
53
Figure 11 —Illustration of PMU Latencies
7.5.4.5 Time quality tests
IEEE Std C37.118-2005 defines a time quality indication field in the FRACSEC portion of all messages.
This time quality is 0 when the PMU is locked to a traceable source of UTC time and indicates a maximum
time error when the PMU is not locked to its time source.
IEEE Std C37.118.2-2011 renames the above time quality indication the message time quality indication
and adds a second PMU time quality indication that is zero only when it is not implemented (as for pre-
IEEE-C37.118.2 devices) and indicates time quality whether the PMU is locked to a traceable time source
or not. This time quality field has also been added to the IRIG-B time code extensions of
IEEE Std C37.118.2-2011 Annex F and to the synchrophasor profile for IEEE Std 1588-2008.
This guide does not provide a test plan for the PMU time quality indication. At the time of writing of this
guide, time sources and PMUs implementing this indication have not been designed and there has been no
field experience with such devices. Test plans have not been discussed nor devised.
7.5.4.5.1 Message time quality test
Testing the message time quality involves measuring the steady-state phase error while the device is locked
to its time source (this will be the “baseline” phase error), removing the time source from the PMU, and
measuring the phase error and comparing it to the message time quality over a period of no less than
1000 s. The phase error minus the baseline phase error is assumed to be caused by the PMU’s timing
inaccuracy. If the phase error “drifts” beyond the phase error that would be caused by the message time
error, the PMU fails the message time quality test.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
54
7.5.5 Tests of multifuncation PMUs
7.5.5.1 Modified multifuncation PMU testing
Devices that have additional functions above those spelled out in IEEE Std C37.118.1-2011 should adjust
the preceding testing of PMU as follows and should perform the additional testing described as follows.
7.5.5.1.1 Range of input signals for multifuncation PMU
The actual measurement range of a multifuncation PMU is determined by (1) the types of transducers
dictated by its primary functions, and (2) the IEEE C37.118.1 required signal range for phasor
measurement.
7.5.5.1.2 Maximum resources loading
The performance of the phasor measurement function of a multifuncation PMU can be tested under the
maximum computing and communications resources utilization conditions.
7.5.5.1.3 Potential interference scenario verification
The performance of the phasor measurement function of a multifuncation PMU can be tested for scenarios
that may result in interference between primary functions and the phasor measurement function.
7.6 Synchrophasor message format
The synchrophasor message formats are given in IEEE Std C37.118-2005 and are extended in
IEEE Std C37.118.2-2011 with a new CFG-3 message and a continuous time quality (CTQ) field in the
SYNC word. This message system is designed for real-time communications between a PMU measuring
device and a data collection device such as a PDC. It can also be used for data sent from a PDC to another
PDC. The overall data system is described here along with some implementation considerations. Details on
the message format itself can be found in IEEE Std C37.118.2-2011.
7.6.1 Message framework
As defined in IEEE Std C37.118-2005, there are four types of messages: data, configuration, header, and
command. The command message is the only message sent from the receiving device to the PMU. It allows
the receiving device to start and stop the data stream, to request a configuration message or the header
message. The command message has allowance to send extended commands. Since these extensions are
manufacturer specific, methods for testing these are not described here.
Each message includes a CRC-CCITT check word (2 bytes). The value of this check word will be checked
for each message received from the PMU, and any deviations must be reported in the test report indicating
the conditions under which the deviation was detected.
Whenever any changes are made to the PMU configuration (such as when the PMU reporting rate is
changed in tests according to 6.3.2), the CFG-2 message will be requested from the PMU. All further
interpretations of the data messages must be in accordance with this new configuration.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
55
The CFG-1 message and the header message will be requested from the PMU and analyzed for proper
format. Finally, the PMU must be tested to respond properly to the commands to start and to stop sending
the data messages as well as provide configuration and header messages.
7.6.2 Implementation issues
IEEE Std C37.118.2-2011 describes a general framework and messages for data reporting. Implementation
will depend on the type of medium being used and communications protocols. Most implementations will
use UDP/IP or TCP/IP protocol over Ethernet. Some additional specifications of unit operation with the
protocol layers to be used will simplify interoperability between utilities and within utility systems.
7.7 Final comments
Validation of PMU performance does not end with field installation of the unit. Firmware changes and
hardware degradation may produce an instrument that is very different from the one tested in the
laboratory. Examining PMU data for evidence of possible discrepancies is a necessary part of WAMS
operation, and cross-validation of measurement sources is a necessary preface to the analysis of a major test
or disturbance on the system.
The basic strategy for field evaluation of PMUs and other measurement sources is to regularly review the
data for inconsistencies, and then to resolve such inconsistencies as may be found. The former activity is a
normal aspect of WAMS operation. The latter activity may involve such tasks as identifying the cause,
upgrading the measurement system, and developing a way to “clean” or repair the data. Success in such
matters may require supplemental laboratory tests or modeling studies, and it is rarely immediate.
Sometimes it is possible to compare PMU data against data collected on other devices that share the same
inputs, such as a digital fault recorder (DFR) or a PMU of another type. More often, though, one must work
solely from PMU data collected at a number of different locations. It can then be difficult to distinguish
discrepancies in timing or dynamic response within the measurement system from locational differences in
response of the power system itself. Consequently PMU characterization through laboratory tests is
essential and may be the only reference for power system analysis.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
56
Annex A
(informative)
Bibliography
Bibliographical references are resources that provide additional or helpful material but do not need to be
understood or used to implement this standard. Reference to these resources is made for informational use
only.
[B1] ANSI/NCSL Z540.3-2006, American National Standard for Calibration—Requirements for the
Calibration of Measuring and Test Equipment, National Conference of Standards Laboratories.
9
[B2] Cisco Systems, QOS Quality of Service.
10
[B3] Dion, A., Calmettes, V., Boutillon, E., Reconfigurable GPS-Galileo receiver for satellite based
applications, Proc. of ION GNSS 2007, Institute Of Navigation, Global Navigation Satellite Systems
Meeting, September, Fort Worth, Texas, USA.
[B4] EIPP evaluating dynamic performance of phasor measurement units, March 2004.
[B5] EIPP Standards and Performance Task Team preliminary report, August 2004.
[B6] Arifian, A., et al., Optic Technology Monitors HV Bus, Transmission and Distribution, 49, no. 5
(May 1997), pp. 62–68.
[B7] Australian Global Navigation Satellite Systems Coordination Committee, GPS Vulnerability:
Information for CIOs, The Trusted Information Sharing Network (TISN), Publications on Critical
Infrastructure Protection, 2006.
11
[B8] Behrendt, K., and Fodero, K., The Perfect Time: An Examination of Time Synchronization
Techniques, Proc. of the Distributech 06, Feb. 7–9, 2006, Tampa, FLA.
[B9] Carroll, K. M., and Celano, T., Timing via the New LORAN-C System, Proc. of the 2003 IEEE
International Frequency Control Symposium and PDA Exhibition, pp. 245−249.
[B10] Celano, T., et al., Common-View LORAN-C as a Backup to GPS for Precise Time Recovery, 35th
Annual Precise Time and Time Interval (PTTI) Meeting, 2003.
[B11] Deeths, D., and Brunette, G., Using NTP to Control and Synchronize System Clocks—Part I:
Introduction to NTP, July 2001.
12
[B12] Depablos, J., et al., Comparative Testing of Synchronized Phasor Measurement Units, paper
presented at the IEEE/PES General Meeting, Denver, CO, USA, June 2004.
[B13] Dickerson, B., Substation Time Synchronization, Protection, Automation and Control World,
Summer 2007, pp. 39−45.
[B14] Fardanesh, B., et al., Multifunctional Synchronized Measurement Network, IEEE Computer
Applications in Power, 11, no. 1, January 1998, pp. 26−30.
[B15] Fernandez, D., et al., “Comparisons Between Lab Measurements and Digital Simulations for a
Coupling Capacitor Voltage Transformer,” IEEE PES T&D Conference and Exposition, Latin America,
2006.
9
ANSI publications are available from the American National Standards Institute (http://www.ansi.org/).
10
Available at http://www.cisco.com/en/US/tech/tk543/tech_topology_and_network_serv_and_protocol_suite_home.html.
11
Available at http://www.tisn.gov.au.
12
Available at http://anselmo.homeunix.net/Sun-BluePrints/0701/NTP.pdf.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
57
[B16] Hamrita, T. K., Heck, B. S., Meliopoulos, A. P., ”On-Line Correction of Errors Introduced By
Instrument Transformers In Transmission-Level Power Waveform Steady-State Measurements,” IEEE
Transactions on Power Delivery, 15, no. 4, October 2001, pp. 1116–1120.
[B17] Hauer, J. F., “Validation of Phasor Calculation in the Macrodyne PMU for California-Oregon
Transmission Project Tests of March 1993,” IEEE Trans. Power Delivery, July 1996, pp. 1224–1231.
[B18] Hauer, J. F., et al., Dynamic Signatures Recorded on the “WAMS East” Monitor Backbone,
August 14, 2003 vs. Other Days, presented by J. F. Hauer to the IEEE/PES IEEE Work Group on Power
System Dynamics Measurements, IEEE PES General Meeting, Denver, CO, USA, June 7, 2004.
[B19] Hauer, J. F., et al., Performance of “WAMS East” in Providing Dynamic Information for the North
East Blackout of August 14, 2003; IEEE/PES Panel on Major Grid Blackouts of 2003 in North America
and Europe, IEEE PES General Meeting, Denver, CO, USA, June 6–12, 2003.
[B20] Hauer, J. F., Martin, K., Lee, H., Evaluating Dynamic Performance of Phasor Measurement Units:
Experience in the Western Power System, Interim Report of the WECC Disturbance Monitoring Work
Group, partial draft of April 28, 2004.
[B21] Hauer, J. F., Martin, K., Lee, H., Evaluating the Dynamic Performance of Phasor Measurement
Units: Experience in the Western Power System, Interim Report of the WECC Disturbance Monitoring
Work Group, partial draft of August 5, 2005.
[B22] Hauer, J. F., et al., Integrated Monitor Facilities for the Western Power System: The WECC WAMS
in 2003, Interim report of the WECC Disturbance Monitoring Work Group, June 25, 2003.
13
[B23] Hauer, J. F., et al., Use of the WECC WAMS in Staged System Tests for Validation of System
Performance and Modeling: Summary Report for September 2005–August 2006, Interim report of the
WECC Disturbance Monitoring Work Group, April 25, 2007.
[B24] Hauer, J. F., et al. (in association with the Disturbance Monitoring Work Group of the Western
Electricity Coordinating Council), Integrated Dynamic Information for the Western Power System: WAMS
Analysis in 2005, Chapter 14 in the Power System Stability and Control volume of The Electric Power
Engineering Handbook, ed. 2, L. L. Grigsby, ed., Boca Raton, FL: CRC Press, 2007.
[B25] Holbert, K. E., Heydt, G. T., Ni, H., Use of Satellite Technologies for Power System Measurements,
Command, and Control, Proc. of the IEEE 93, 2005, pp. 947−955.
[B26] Huang, H. et al., Laboratory Performance Evaluation Report of SEL 421 Phasor Measurement Unit
(PMU), PNNL technical report PNNL-16852, November 2007.
14
[B27] IEC 60870, Telecontrol Equipment and Systems, Transmission Protocols.
15
[B28] IEEE Std 525™, IEEE Guide for the Design and Installation of Cable Systems in Substations.
16,
17
[B29] IEEE Std 1344™-1995 (Reaff 2001), IEEE Standard for Synchrophasors for Power Systems.
[B30] IEEE Std 1379™-2000, IEEE Trial-Use Recommended Practice for Data Communications Between
Intelligent Electronic Devices and Remote Terminal Units in a Substation.
[B31] IEEE Std 1615™-2007, IEEE Recommended Practice for Network Communication in Electric
Power Substations.
[B32] IEEE Std C37.2™-1991, IEEE Standard Electrical Power System Device Function Numbers and
Contact Designations.
13
Available at www.wecc.biz/committees/JGC/DMWG/documents.
14
Available at http://www.pnl.gov/main/publications/external/technical_reports/PNNL-16852.pdf.
15
IEC publications are available from the International Electrotechnical Commission (http://www.iec.ch/). IEC publications are also
available in the United States from the American National Standards Institute (http://www.ansi.org/).
16
IEEE publications are available from The Institute of Electrical and Electronics Engineers (http://standards.ieee.org/).
17
The IEEE standards or products referred to in this clause are trademarks of The Institute of Electrical and Electronics Engineers,
Inc.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
58
[B33] IEEE Power Systems Relaying Committee, Application Considerations of UCA 2 for Substation
Ethernet Local Area Network Communication for Protection and Control.
18
[B34] InterNational Electrical Testing Association, NETA Acceptance Testing Specifications, International
Electrical Testing Association, 2003.
[B35] InterNational Electrical Testing Association, NETA Acceptance Testing Specifications, International
Electrical Testing Association, January 1, 2007.
[B36] Jacoby, J., et al., Advantages of a Combined GPS/Loran-C Precision Timing Receiver, Proceedings
of the 28th ILA Technical Symposium, 1999.
[B37] Lilley, R., Church, G., Harrison, M., White Paper GPS Backup for Position, Navigation and Timing
Transition Strategy for Navigation and Surveillance, Aug 22, 2006.
19
[B38] Lombardi, M. A., et al., Time and Frequency Measurements Using the Global Positioning System,
The International Journal of Metrology, July-September 2001, pp. 26−33.
[B39] Mack, P., et al., Application of the Galileo System for a Better Synchronization of Electrical Power
Systems, Proc. IEEE Power Tech Conference, Lausanne, Switzerland, 2007.
[B40] Martin, K. E., IEEE Standard for Synchrophasors for Power Systems, IEEE Transactions on Power
Delivery, vol. 13, no. 1, January 1998, pp. 73−77.
[B41] Martin, K. E., Synchronized System Measurement Networks in North America: Operating Process
and System Formats Based Upon BPA’s Phasor Data Concentrator, WAMS Working Note, June 1, 2004.
[B42] Martin, K. E., WECC Disturbance/Performance Monitor Equipment: Proposed Standards for WECC
Certification and Reimbursement, Draft report of the WECC Disturbance Monitoring Work Group,
March 17, 2004.
[B43] Martin, K. E., Hauer, J. F., Faris, T. J., PMU Testing and Installation Considerations at the
Bonneville Power Administration, presented to the Panel of Performance Characteristics and Evaluation of
PMUs, IEEE/PES 2007 General Meeting, Tampa, FL, USA, June 24–28, 2007.
[B44] Martin, K. E., et al., WECC Disturbance/Performance Monitor Equipment: Proposed Standards for
WECC Certification and Reimbursement, March 2004.
[B45] Meliopoulos, A. P., and Cokkinides, G. J., Visualization and Animation of Instrumentation Channel
Effects on DFR Data Accuracy, Proceedings of the 2002 Georgia Tech Fault and Disturbance Analysis
Conference, Atlanta, Georgia, USA, April 29-30, 2002.
[B46] Meliopoulos, A. P., and Cokkinides, G. J., “A Virtual Environment for Protective Relaying
Evaluation and Testing,” IEEE Transactions of Power Systems, no. 1, February 2004, pp. 104–111.
[B47] Meliopoulos, A. P., Zhang, F., Zelingher, S., Hardware and Software Requirements for a
Transmission System Harmonic Measurement System, Proceedings of the Fifth International Conference
on Harmonics in Power Systems (ICHPS V), Atlanta, GA, USA, September 1992, pp. 330–338.
[B48] Meliopoulos, A. P., Zhang, F., Zelingher, S., “Power System Harmonic State Estimation,” IEEE
Transactions on Power Systems, no. 3, July 1994, pp. 1701–1709.
[B49] Meliopoulos, A. P., et al., GPS-Synchronized Data Acquisition: Technology Assessment and
Research Issues, Proceedings of the 39th Hawaii International Conference on System Sciences, 2006.
[B50] Meliopoulos, A. P., et al., “Transmission Level Instrument Transformers and Transient Event
Recorders Characterization for Harmonic Measurements,” IEEE Transactions on Power Delivery, vol. 8,
no. 3, July 1993, pp. 1507–1517.
[B51] Mills, D. L., Network Time Protocol Version 4 Reference and Implementation Guide, Electrical and
Computer Engineering Technical Report 06-06-1, University of Delaware, June 2006, p. 83.
18
Available at www.pes-psrc.org/h/DRAFT%235(7-16-04)H6%20PAPER.zip.
19
Available at www.loran.org/news/GPS-Backup-Released.pdf.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
59
[B52] NASPI, PMU System Test Guide Full 12_07a document.
20
[B53] NASPI, Performance and Standards Task Team—A Guide for PMU Installation, Commissioning,
and Maintenance, Part I: PMU Acceptance Test Checklist.
[B54] NASPI, Performance and Standards Task Team—A Guide for PMU Installation, Commissioning,
and Maintenance, Part II: PMU Installation Procedures, Performance, and Standards Task Team.
[B55] NASPI, Performance and Standards Task Team—SynchroPhasor Accuracy Characterization,
Performance, and Standards.
[B56] NASPI, PRTT Testing and Interoperability Guide, 2007.
[B57] Orpen, O., and Zwaan, H., “Dual Frequency DGPS Service for Combating Ionospheric
Interference,” The Journal of Navigation, vol. 54, no. 1, 2001, pp. 29, 36.
[B58] Power System Dynamic Performance Committee, Blackout Experiences and Lessons, Best Practices
for System Dynamic Performance, and the Role of New Technologies, IEEE Task Force Report, Special
Publication 07TP190, July 2007.
[B59] Stenbakken, G. N., et al., Reference Values for Dynamic Calibration of PMUs, Proceedings of the
41st Hawaii International Conference on System Sciences (HICSS-41), January 2008, IEEE Computer
Society, compact disk.
[B60] Volpe, J., Vulnerability assessment of the transportation infrastructure relying on the Global
Positioning System (The Volpe report), John A. Volpe National Transportation System Center,
Washington, DC, 2001.
[B61] Wesson, K., Shepard, D., and Humphreys, T., “Straight Talk on Anti-Spoofing: Securing the Future
of PNT,” GPS World, January 2012.
[B62] Zelingher, S., Stillmann, G. I., Meliopoulos, A. P., Transmission System Harmonic Measurement
System: A Feasibility Study, Proceedings of the Fourth International Conference on Harmonics in Power
Systems (ICHPS IV), Budapest, Hungary. October 1990, pp. 436-444.
20
NASPI documents are available at www.naspi.org.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
60
Annex B
(informative)
Responses of the reference signal processing model to test signals
B.1 General
This annex provides sample typical responses of the reference signal processing model presented in
Appendix C of IEEE Std C37.118.1-2011. Typical responses are informative only and do not construe
specifications or limits. PMUs being tested may respond in a similar or quite different manner but would
still be in compliance if they meet or exceed the limits of IEEE Std C37.118.1-2011.
The PMU model response examples from this annex do not include time-stamp round-off error as the
model is simulated with a floating-point time resolution with much finer resolution than 1 μs.
B.2 Introduction to the graphs
The following graphs were made using a mathematical model of the PMU reference signal processing
model of IEEE Std C37.118.1-2011 Annex C. The graphs are made at the full system sampling rates of
960 samples/second for M class and 900 samples/second for P class. This is done before the decimation to
the output reporting rate in order to show the actual internal response to the PMU filters that were used to
verify the performance limits of the standard.
B.3 Response of the model to steady-state test signals
B.3.1 Steady-state–signal frequency range
B.3.1.1 M class signal frequency range
The TVE of the M class reference model at nominal frequency is dominated by the attenuation of the filter
following the complex multiplication of the input signal with the nominal frequency carrier. The output of
this multiplication is both the difference of the signals and the sum of the signals. The M class filter passes
the difference signal and attenuates 20 dB of the sum signal.
Following are plots of TVE, magnitude error, and phase error for nominal frequency input at a PMU
reporting rate (F
s
) of 60 FPS. These plots are taken at the internal sampling rate of the PMU model at
960 samples per second and the sum signal can clearly be seen in Figure B.1:
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
61
0.4 0.5 0.6 0.7 0.8 0.9
0.0313
0.0313
0.0313
0.0313
0.0313
Time (seconds)
T
V
E
(
%
)
TVE vs Time
0 0.2 0.4 0.6 0.8
-4
-2
0
2
4
x 10
-4
Time (seconds)
M
a
g
n
i
t
u
d
e
e
r
r
o
r
Magnitude Error vs Time
0 0.2 0.4 0.6 0.8 1
-0.02
-0.01
0
0.01
0.02
Time (seconds)
P
h
a
s
e
e
r
r
o
r
(
d
e
g
r
e
e
s
)
Phase Error vs Time
Figure B.1—TVE, magnitude, and phase error for nominal input at 60 FPS reporting rate
B.3.1.1.1 TVE at bandwidth limits
TVE as the input frequency changes is dominated by the magnitude roll off of the PMU filter. This can be
observed by looking at plots of TVE, magnitude, and magnitude error as the frequency ramps from the low
bandwidth limit to the high bandwidth limit shown in Figure B.2 and Table B.1:
55 60 65
0
0.5
1
Delta Frequency from Nominal
T
V
E
(
%
)
TVE vs Frequency
55 60 65
0.7
0.71
0.72
0.73
0.74
Delta Frequency from Nominal
M
a
g
n
i
u
d
e
Magnitude vs Frequency
55 60 65
-5
0
5
10
x 10
-3
Delta Frequency from Nominal
M
a
g
n
i
t
u
d
e
E
r
r
o
r
Magnitude Error vs Frequency
Figure B.2—TVE, magnitude, and magnitude error for F
s
= 60 FPS as frequency ramps
between the bandwidth limits (M class)
More on frequency ramping later.
Table B.1—TVE, FE, and RFE at the bandwidth limits (M class)
Reporting rate (F
s
)
upper frequency (Fu)
lower frequency (Fl)
Upper bandwidth frequency
(fu)
Lower bandwidth frequency (Ll)
F
s
= 120
Fu=65 Hz
Fl=55 Hz
Max TVE(%) = 1.0
Max FE = 3.7e-013
Max RFE = 1.2e-010
Max TVE(%) = 0.83
Max FE = 4.5e-013
Max RFE = 1.4e-010
F
s
= 60
Fu=65 Hz
Fl=55 Hz
Max TVE(%) = 0.92
Max FE = 1.7e-013
Max RFE = 9.1e-011
Max TVE(%) = 0.92
Max FE = 2.1e-013
Max RFE = 1.4e-010
F
s
= 30
Fu=65 Hz
Fl=55 Hz
Max TVE(%) = 0.995
Max FE = 1.4e-013
Max RFE = 1.3e-010
Max TVE(%) = 1.0207
Max FE = 1.3e-013
Max RFE = 1.16e-010
F
s
= 20
Fu=64 Hz
Fl=56 Hz
Max TVE(%) = 0.89
Max FE = 1.1e-013
Max RFE = 1.2e-010
Max TVE(%) = 0.88
Max FE = 1.1e-013
Max RFE = 1.4e-010
F
s
= 15
Fu=63 Hz
Fl=57 Hz
Max TVE(%) = 0.88
Max FE = 9.6e-014
Max RFE = 1.4e-010
Max TVE(%) = 0.89
Max FE = 1.2e-013
Max RFE = 1.4e-010
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
62
Table B.1—TVE, FE, and RFE at the bandwidth limits (M class) (continued)
Reporting rate (F
s
)
upper frequency (Fu)
lower frequency (Fl)
Upper bandwidth frequency
(fu)
Lower bandwidth frequency (Ll)
F
s
= 12
Fu = 62.4 Hz
Fl = 57.6 Hz
Max TVE(%) = 0.89
Max FE = 1.2e-013
Max RFE = 1.7e-010
Max TVE(%) = 0.88
Max FE = 1.0e-013
Max RFE = 1.6e-010
F
s
= 10
Fu = 62 Hz
Fl = 58 Hz
Max TVE(%) = 0.85
Max FE = 1.1e-013
Max RFE = 1.4e-010
Max TVE(%) = 0.86
Max FE = 1.0e-013
Max RFE = 1.7e-010
B.3.1.2 P class signal frequency range
For steady-state input at the nominal frequency, the P class reference model filter has extremely good
attenuation of all harmonics. This means that the effect of the sum output of the complex multiplication is
insignificant and TVE for nominal signals is effectively 0.
The P class PMU model also uses magnitude compensation for off-nominal signals, which improves
performance; however the attenuation for off-nominal sum signal can be as little as 25 dB.
Figure B.3 shows plots of TVE, magnitude, and magnitude error for a frequency ramp from 5 Hz below to
5 Hz above the nominal frequency:
55 60 65
0
0.1
0.2
0.3
0.4
Delta Frequency from Nominal
T
V
E
(
%
)
TVE vs Frequency
55 60 65
0.7
0.705
0.71
0.715
0.72
Delta Frequency from Nominal
M
a
g
n
i
u
d
e
Magnitude vs Frequency
55 60 65
-2
-1
0
1
2
x 10
-3
Delta Frequency from Nominal
M
a
g
n
i
t
u
d
e
E
r
r
o
r
Magnitude Error vs Frequency
Figure B.3—TVE, magnitude and magnitude error for P class
In the preceding figures, the effect of the sum signal at off-nominal frequencies is apparent. The effect of
the frequency compensation for magnitude is also apparent when compared to the M class magnitude plot.
See Table B.2.
Table B.2—TVE, FE, and RFE at the bandwidth limits (P class)
Reporting rate (F
s
)
upper frequency (Fu)
lower frequency (Fl)
Upper bandwidth
frequency (Fu)
Lower bandwidth frequency (Ll)
F
s
= 60
Fu = 62 Hz
Fl = 58 Hz
%TVE =0.03
FE =2.0e-013
RFE = 1.3e-010
%TVE = 0.032
FE = 3.5e-013
RFE = 7.5e-011
Note that the P class reference model filter is the same at all reporting rates so the performance will be the
same for all reporting rates.
B.3.2 Steady-state signal magnitude range
The PMU reference model performs the same under any steady-state signal magnitude. The results of this
test for the limits of the magnitude range are trivial and are not shown here. The results will not be trivial
for real-world PMUs.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
63
B.3.3 Steady-state phase angle range
The PMU reference model performs the same under any steady-state phase angle. The results of this test
for the limits of the magnitude range are trivial and so are not shown here. The results will not be trivial for
real-world PMUs.
B.3.4 Steady-state worst-case harmonic distortion
The PMU reference model is most susceptible to interference from lower harmonics. This is due to filter
design tradeoffs between reporting latency, relation of frequency and ROCOF measurements and the time
stamp of the phasor, and applications need for out-of-band noise immunity.
B.3.4.1 M class response to harmonic distortion
Table B.3—M class response to 10% 2nd through 8th harmonics
2
nd
3
rd a
4
th
5
th
6
th
a
7th 8
th
F
s
=
120
%TVE =
1.12
FE = 0.018
RFE = 19.5
%TVE = 0.13
FE = 4e-014
RFE = 2e-011
%TVE =
0.13
FE = 0.018
RFE = 19.5
%TVE = 0.13
FE = 0.0015
RFE = 6.56
%TVE = 0.13
FE = 5e-014
RFE =
2.5e-011
%TVE = 0.13
FE = 0.0015
RFE = 6.56
%TVE = 0.13
FE = 0.003
RFE = 5.37
F
s
= 60 %TVE =
0.038
FE = 0.0023
RFE = 2.46
%TVE =
0.036
FE = 2.5e-014
RFE =
2.4e-011
%TVE =
0.035
FE = 0.0023
RFE = 2.46
%TVE =
0.034
FE = 0.0002
RFE = 0.81
%TVE =
0.034
FE = 3.5e-014
RFE =
3.5e-011
%TVE =
0.034
FE = 0.0002
RFE = 0.81
%TVE =
0.034
FE = 0.0002
RFE = 0.81
F
s
= 30 %TVE =
0.012
FE = 0.0016
RFE = 1.66
%TVE =
0.020
FE = 1.3e-014
RFE =
2.6e-011
%TVE =
0.014
FE = 0.0016
RFE = 1.656
%TVE =
0.019
FE = 0.0002
RFE = 0.72
%TVE =
0.014
FE = 3.1e-014
RFE =
4.5e-011
%TVE =
0.019
FE = 0.0002
RFE = 0.72
%TVE =
0.014
FE = 0.0015
RFE = 2.83
F
s
= 20 %TVE =
0.013
FE = 0.0013
RFE = 1.40
%TVE =
0.018
FE = 2.9e-014
RFE = 5e-011
%TVE =
0.014
FE = 0.0013
RFE = 1.40
%TVE =
0.016
FE = 3.4e-005
RFE = 0.15
%TVE =
0.016
FE = 1.7e-014
RFE = 3e-011
%TVE =
0.015
FE =
3.4e-005
RFE = 0.15
%TVE =
0.017
FE = 0.0007
RFE = 1.26
F
s
= 15
%TVE =
0.004
FE = 0.0006
RFE = 0.62
%TVE =
0.007
FE = 7e-014
RFE =
9.6e-011
%TVE =
0.005
FE = 0.0006
RFE = 0.62
%TVE =
0.007
FE = 6e-005
RFE = 0.28
%TVE =
0.005
FE = 6e-014
RFE =
6.9e-011
%TVE =
0.007
FE = 6e-005
RFE = 0.28
%TVE =
0.005
FE = 0.0006
RFE = 1.11
F
s
= 12
%TVE =
0.007
FE = 0.0001
RFE = 0.13
%TVE =
0.010
FE = 5e-014
RFE = 9e-011
%TVE =
0.010
FE = 0.0001
RFE = 0.13
%TVE =
0.009
FE = 2.0e-005
RFE = 0.09
%TVE =
0.010
FE = 4e-014
RFE = 8e-011
%TVE =
0.010
FE = 4e-014
RFE = 8e-011
%TVE =
0.010
FE = 2.0e-005
RFE = 0.09
F
s
= 10
%TVE =
0.010
FE = 0.0002
RFE = 0.16
%TVE =
0.009
FE = 5e-014
Rfe = 6e-011
%TVE =
0.008
FE = 0.0002
RFE = 0.16
%TVE =
0.008
FE = 1.8e-005
RFE = 0.08
%TVE =
0.008
FE = 5e-014
RFE = 1e-010
%TVE =
0.009
FE = 1.8e-
005
RFE = 0.08
%TVE =
0.009
FE = 0.0002
RFE = 0.43
a
Harmonic numbers with multiples of 3 are zero sequence and have no effect on the Frequency and ROCOF measurements.
ROCOF is most sensitive to harmonic distortion. The higher the reporting rate, the more susceptible the
ROCOF measurement is.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
64
B.3.4.2 P class response to harmonic distortion
Since the P class reference model filter has excellent harmonic rejection, the TVE, FE, and RFE for
harmonics are very small. See Table B.4.
Table B.4—P class response to 1% 2nd through 8th harmonics
2
nd
3
rd a
4
th
5
th
6
th
a
7th
F
s
=
Any
%TVE =
1.8e-012
FE = 1.1e-013
RFE| =
3.1e-011
%TVE =
1.8e-012
FE = 1.0e-013
RFE =
2.6e-011
%TVE =
1.5e-012
FE = 8e-014
RFE =
2.1e-011
%TVE =
2.2e-012
FE = 1.5e-013
RFE =
3.9e-011
%TVE =
2.7e-012
FE = 2.4e-013
RFE =
4.2e-011
%TVE =
2.1e-012
FE = 1.2e-013
RFE = 4.0e-011
a
Harmonic numbers with multiples of 3 are zero sequence and have no effect on the Frequency and ROCOF measurements.
B.3.5 Steady-state worst-case out-of-band interfering signal
For each PMU reporting Rate F
s
, the passband is defined as |f – f
0
|≤ F
s
/2
where f is the frequency of an interfering signal and f
0
is the nominal frequency (50 Hz or 60 Hz).
IEEE Std C37.118.1-2011 5.5.5 Table 3 and Table 4 specify limits for TVE, FE, and RFE in the presence
of out-of-band interfering signals where |f – f
0
|≥ F
s
/2. There are no specified limits for P class, only for M
class.
For the test, the input test frequency is varied between f
0
and 10% of the Nyquist frequency of the reporting
rate. The worst-case performance happens when the input frequency is furthest away from nominal
(f
0
± 10% of Nyquist)
IEEE Std C37.118.1-2011 Annex C PMU reference model has a passband of ± F
s
/5 on either side of
nominal frequency. The worst-case interference is when the interfering signal is near the band limit
frequency. Performance improves rapidly as the frequency of interfering signal gets further from the
nominal frequency.
For the model, the off-nominal input test frequency contributes to TVE due to the magnitude error of the
filter response. The worst case is when the input frequency is at f
0
± 10% of Nyquist.
Example
For IEEE Std C37.118.1-2011 Annex C model with nominal frequency f
0
= 60 Hz and F
s
reporting rate
30 FPS, Table B.5 shows TVE, FE, and RFE for input and interfering frequencies on either side of nominal
under worst-case conditions.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
65
Table B.5—Worst-case effect of out-of-band interfering signals for F
s
= 60 and 10 FPS
F
s
F
in
29 Hz 29.5 Hz 30 Hz 90 Hz 90.5 Hz 91 Hz
TVE (%) 0.99727 1.1041 1.1497 1.1515 1.1024 0.98901
FE (Hz) 1.99E-13 2.27E-13 2.34E-13 2.20E-13 2.34E-13 2.20E-13 63.0 Hz
RFE (Hz/s) 1.24E-10 8.48E-11 7.16E-11 8.44E-11 9.12E-11 1.04E-10
TVE (%) 0.95245 1.0698 1.1907 1.1925 1.0688 0.95627
FE (Hz) 2.20E-13 2.27E-13 2.27E-13 2.42E-13 2.27E-13 2.27E-13
60 FPS
57.0 Hz
RFE (Hz/s) 5.88E-11 9.76E-11 9.08E-11 8.48E-11 1.04E-10 9.81E-11
54 Hz 54.5 Hz 55 Hz 65 Hz 65.5 Hz 66 Hz
TVE (%) 0.13599 0.29562 0.99495 0.99515 0.3094 0.13072
FE (Hz) 4.97E-14 7.82E-14 4.974E-14 6.39E-14 8.53E-14 6.39E-14 60.5
RFE (Hz/s) 9.12E-11 7.17E-11 8.473E-11 1.24E-10 7.17E-11 9.44E-11
TVE (%) 0.13516 0.30457 1.0008 1.001 0.29098 0.13338
FE (Hz) 4.97E-14 7.11E-14 5.68E-01 3.55E-14 5.68E-14 5.68E-14
10 FPS
59.5
RFE (Hz/s) 7.81E-11 1.11E-10 1.43E-10 5.21E-11 1.24E-10 7.33E-11
Note that the effect of out-of-band interfering signals on FE and RFE are negligible. The errors shown here
are primarily due to floating point errors in the simulation.
B.4 Response to dynamic input signal tests
B.4.1 Dynamic measurement bandwidth tests response
To understand how the PMU reference model responds to modulated input, it is helpful to look at the
response to phase modulation (PM) and amplitude modulation (AM) separately.
Plots of PMU response with respect to input signal frequency can be quite informative, especially when
troubleshooting. Figure B.4 shows plots for measured phase at reporting rate of 60 FPS versus both time
and frequency for a ω
a
= ω
x
= 1 Hz phase modulation at K
a
= K
x
= 0.1 Hz modulation index:
0 0.5 1 1.5 2
-10
-5
0
5
10
Time (seconds)
M
e
a
s
u
r
e
d
P
h
a
s
e
(
d
e
g
r
e
e
s
)
Phase vs Time
-0.1 -0.05 0 0.05 0.1
-10
-5
0
5
10
Delta Frequency from Nominal
M
e
a
s
u
r
e
d
P
h
a
s
e
(
d
e
g
r
e
e
s
)
Phase vs Frequency
Figure B.4—Phase versus time and frequency for 1 Hz modulation at 0.1 index
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
66
The phase modulation is specified to be Ka × cos(ω
m
t – π) so the measured phase begins at –5 degrees at
0 Hz delta frequency and increases to 0 degrees at 0.1 Hz delta frequency. A. 0.1 Hz delta frequency is
where the modulation is crossing 0 (going positive) so the phase continues to increase towards 5 degrees
while the frequency decreases back to 0. Frequency crosses 0 at the positive peak of the modulation signal
so phase begins to decrease as the frequency decreases and phase reaches 0 at –0.1Hz, which is the
negative going zero crossing of the modulation signal. Frequency begins to increase and phase continues to
decrease as the modulation continues towards the negative peak where it began.
Note that for a modulated phase, the input frequency is at its maximum and minimum at the zero crossings
of the modulation; this is where the ROCOF is at its maximum.
Dynamic measurement bandwidth uses combined amplitude and phase modulated input signals to test
PMU response. For all tests, amplitude and phase modulations indices are fixed at 0.10 (10%). Modulation
frequency (Fmod) is varied from 0.1 Hz up to the reporting rate divided by 5 (Fs/5).
The worst-case response for the reference PMU at any reporting rate is at the maximum modulation
frequency of F
s
/5. Table B.6 shows maximum TVE, FE, and ROCOF error for the worst-case modulation
frequencies:
Table B.6—Response to combined amplitude and phase modulated input, 10% index
M-class P-class
F
s
= 120 Hz
Fmod = 24 Hz
%TVE = 3.01
FE = 1.16
RFE = 215.0
Fmod = 12 Hz
%TVE = 1.98
FE = 0.33
RFE = 30.5
F
s
= 60 Hz
Fmod = 12 Hz
%TVE = 1.75
FE = 0.34
RFE = 33.5
Fmod = 6 Hz
%TVE = 0.52
FE = 0.078
RFE = 3.6
F
s
= 30 Hz
Fmod = 6 Hz
%TVE = 0.56
FE = 0.095
RFE = 5.59
Fmod = 3 Hz
%TVE = 0.13
FE = 0.019
RFE = 0.43
F
s
= 20 Hz
Fmod = 4 Hz
%TVE = 0.35
FE = 0.047
RFE = 2.04
Fmod = 2 Hz
%TVE = 0.06
FE = 0.008
RFE = 0.13
F
s
= 15 Hz
Fmod = 3 Hz
%TVE = 0.74
FE = 0.038
RFE = 1.20
Fmod = 1.5 Hz
%TVE = 0.03
FE = 0.005
RFE = 0.05
F
s
= 12 Hz
Fmod = 2.4 Hz
%TVE = 0.35
FE = 0.022
RFE = 0.65
Fmod = 1.2 Hz
%TVE = 0.02
FE = 0.003
RFE = 0.03
F
s
= 10 Hz
Fmod = 2
%TVE = 0.76
FE = 0.023
RFE = 0.48
Fmod = 1 Hz
%TVE = 0.015
FE = 0.002
RFE = 0.02
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
67
B.4.1.1 Effect of phase modulation on TVE
The effect of phase modulation on TVE is also quite informative, especially what happens as the
modulation frequency increases. The next few plots show TVE for low modulation frequencies increasing
to higher modulation frequencies.
Figure B.5 shows TVE versus time and frequency for 60 FPS reporting rate, 3 Hz modulation frequency,
and 0.1 modulation index. The maximum TVE is well within the limits but is maximum at 0 Hz delta
frequency. This is where the modulation is crossing 0 and the ROCOF is at its highest. At either end of the
delta frequency (0.1 index × 3 Hz modulation frequency = 0.3 Hz delta frequency) the TVE is at its lowest.
As the modulation frequency increases, so does the maximum TVE:
0 0.2 0.4 0.6 0.8 1
0
0.01
0.02
0.03
Time (seconds)
T
V
E
(
%
)
TVE vs Time
-0.4 -0.2 0 0.2 0.4
0
0.01
0.02
0.03
Delta Frequency from Nominal
T
V
E
(
%
)
TVE vs Frequency
Figure B.5—TVE versus time and frequency for F
s
= 60, 3 Hz phase modulation
at 0.1 index
The plots in Figure B.6 were made at 6Hz modulation frequency, 0.1; the maximum TVE is about 0.15%
and the minimum TVE is about 0.05. Notice that the TVE samples are symmetrical on either side of 0 delta
frequency. This is an effect of the relation between the modulation frequency and the sample rate.
0 0.2 0.4 0.6 0.8 1
0.05
0.1
0.15
0.2
0.25
Time (seconds)
T
V
E
(
%
)
TVE vs Time
-1 -0.5 0 0.5 1
0.05
0.1
0.15
0.2
0.25
Delta Frequency from Nominal
T
V
E
(
%
)
TVE vs Frequency
Figure B.6—TVE versus time and frequency for F
s
=60, 6Hz phase modulation at 0.1 index
An interesting and important effect occurs then the TVE samples are not symmetrical about 0 delta
frequency, as shown in the plots of 8 Hz modulation frequency in Figure B.7:
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
68
0 0.2 0.4 0.6 0.8 1
0
0.1
0.2
0.3
0.4
Time (seconds)
T
V
E
(
%
)
TVE vs Time
-1 -0.5 0 0.5 1
0
0.1
0.2
0.3
0.4
Delta Frequency from Nominal
T
V
E
(
%
)
TVE vs Frequency
Figure B.7—TVE versus time and frequency for F
s
= 60, 8 Hz phase modulation
at 0.1 index
Notice on the plot of TVE versus time the lines between the TVE sample points. This effect could be a
problem an insufficient number of samples is taken. The maximum TVE may not be found, as illustrated by
the following set of plots taken at 9.8 Hz modulation frequency.
In the plots in Figure B.8, it is clear that an insufficient number of samples were taken to determine the
maximum TVE. The pair of plots took a sufficient number of samples.
0 0.1 0.2 0.3 0.4 0.5
0
0.2
0.4
0.6
0.8
Time (seconds)
T
V
E
(
%
)
TVE vs Time
-1 -0.5 0 0.5 1
0
0.2
0.4
0.6
0.8
Delta Frequency from Nominal
T
V
E
(
%
)
TVE vs Frequency
0 0.5 1 1.5 2
0
0.2
0.4
0.6
0.8
Time (seconds)
T
V
E
(
%
)
TVE vs Time
-1 -0.5 0 0.5 1
0
0.2
0.4
0.6
0.8
Delta Frequency from Nominal
T
V
E
(
%
)
TVE vs Frequency
Figure B.8—TVE versus time and frequency for F
s
= 60, 9.8 Hz phase modulation at
0.1 index
Finally, the plots in Figure B.9 show what happens at the upper limit of the measurement bandwidth test.
The upper set of plots was taken at modulation frequency of 12 Hz (still at 60 FPS reporting rate) and the
lower set taken at modulation frequency of 11.9 Hz. Both show the maximum TVE at about 1.15%.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
69
0 0.5 1 1.5 2
0
0.5
1
1.5
Time (seconds)
T
V
E
(
%
)
TVE vs Time
-2 -1 0 1 2
0
0.5
1
1.5
Delta Frequency from Nominal
T
V
E
(
%
)
TVE vs Frequency
0 0.5 1 1.5 2
0
0.5
1
1.5
Time (seconds)
T
V
E
(
%
)
TVE vs Time
-2 -1 0 1 2
0
0.5
1
1.5
Delta Frequency from Nominal
T
V
E
(
%
)
TVE vs Frequency
Figure B.9—TVE versus time and frequency for 12 Hz and 11.8 Hz modulation at 0.1 index
B.4.1.1.1 Combined amplitude and phase modulation
An interesting thing happens to TVE when phase and amplitude modulation is combined (see Figure B.10).
0 0.2 0.4 0.6 0.8 1
0
0.1
0.2
0.3
0.4
Time (seconds)
T
V
E
(
%
)
TVE vs Time
59 59.5 60 60.5 61
0
0.1
0.2
0.3
0.4
Delta Frequency from Nominal
T
V
E
(
%
)
TVE vs Frequency
Figure B.10—TVE versus time and frequency for AM and PM at 6 Hz, 0.1 index
TVE as frequency is increasing is lower than TVE as frequency is decreasing. The maximum TVE is where
the frequency crosses zero, which is also the maximum ROCOF because this is where the phase error is at
its maximum. The main contributor to TVE at the frequency endpoints (where the ROCOF is 0) is
magnitude error due to the filter roll-off described in B.3.1.
The reason that the TVE versus frequency is different for increasing and decreasing frequency is because
the amplitude modulation is 180 degrees out of phase with the frequency modulation. When the frequency
is increasing, the amplitude is decreasing, and its contribution to TVE is reduced compared to when the
frequency is decreasing and the amplitude is increasing.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
70
B.4.1.2 Effect of modulation on frequency error and rate of change frequency error
The plots in Figure B.11 illustrate FE and RFE under 1 Hz phase modulation with 0.1 modulation index:
0 0.5 1 1.5 2
-2
-1
0
1
2
x 10
-3
Time (seconds)
F
r
e
q
u
e
n
c
y
e
r
r
o
r
Frequency Error vs Time
59.8 59.9 60 60.1 60.2
-2
-1
0
1
2
x 10
-3
Frequency
F
r
e
q
u
e
n
c
y
E
r
r
o
r
Frequency Error vs Frequency
Figure B.11—Frequency error versus time and frequency for 1 Hz phase modulation at
0.1 index
The frequency error is greatest when the delta frequency is crossing zero. This is because the rate of change
of frequency is at its greatest at this point, and the effect of the very small difference between the phasor
time stamp and the actual time of the frequency measurement has its greatest effect. The sign of the
frequency error changes and is positive while the frequency is increasing and negative when it is
decreasing. See Figure B.12.
0 0.5 1 1.5 2
-2
-1
0
1
2
x 10
-3
Time (seconds)
R
O
C
O
F
e
r
r
o
r
ROCOF Error vs Time
59.8 59.9 60 60.1 60.2
-2
-1
0
1
2
x 10
-3
Frequency
R
O
C
O
F
E
r
r
o
r
ROCOF Error vs Frequency
Figure B.12—FE versus time and frequency for 1 Hz phase modulation at 0.1 index
RFE is greatest where the delta frequencies are greatest and least where the delta frequency crosses zero.
The reason for this is that ROCOF itself is zero when the modulation signal is at its positive and negative
peaks and maximum where the modulation signal is crossing zero.
As the phase modulation frequency increases, so does the error due to the effect of the small time offset
from the phasor estimate (see Figure B.13):
An improvement can be made to the frequency and ROCOF estimation by centering the estimate over the
time stamp. This is made at the cost of one or two internal PMU sample periods of additional latency.
Figure B.14 shows plots at the same modulation as above using a three sample estimate of frequency
centered on the phasor being reported at the time of the time stamp. Maximum FE is improved from 0.069
Hz to 0.024 Hz and RFE is improved from 3.08 Hz/s to 1.37 Hz/s.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
71
0 0.5 1 1.5 2
-0.1
-0.05
0
0.05
0.1
Time (seconds)
F
r
e
q
u
e
n
c
y
e
r
r
o
r
Frequency Error vs Time
59 59.5 60 60.5 61
-0.1
-0.05
0
0.05
0.1
Frequency
F
r
e
q
u
e
n
c
y
E
r
r
o
r
Frequency Error vs Frequency
0 0.5 1 1.5 2
-4
-2
0
2
4
Time (seconds)
R
O
C
O
F
e
r
r
o
r
ROCOF Error vs Time
59 59.5 60 60.5 61
-4
-2
0
2
4
Frequency
R
O
C
O
F
E
r
r
o
r
ROCOF Error vs Frequency
Figure B.13—FE and RFE versus time and frequency for 6 Hz phase modulation
at 0.1 index
0 0.5 1 1.5 2
-0.01
-0.005
0
0.005
0.01
Time (seconds)
F
r
e
q
u
e
n
c
y
e
r
r
o
r
Frequency Error vs Time
59 59.5 60 60.5 61
-0.01
-0.005
0
0.005
0.01
Frequency
F
r
e
q
u
e
n
c
y
E
r
r
o
r
Frequency Error vs Frequency
0 0.5 1 1.5 2
-1
-0.5
0
0.5
1
Time (seconds)
R
O
C
O
F
e
r
r
o
r
ROCOF Error vs Time
59 59.5 60 60.5 61
-1
-0.5
0
0.5
1
Frequency
R
O
C
O
F
E
r
r
o
r
ROCOF Error vs Frequency
Figure B.14—Similar plots using a three sample estimate of frequency centered on the
phasor being reported at the time of the time stamp
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
72
Table B.7—Response to combined amplitude and phase modulated input, 10% index
M class P class
F
s
= 120 Hz
Fmod = 24 Hz
%TVE = 0.83
FE = 0.28
RFE = 26.5
Fmod = 12 Hz
%TVE = 1.79
FE = 0.33
RFE = 29.4
F
s
= 60 Hz
Fmod = 12 Hz
%TVE = 1.60
FE = 0.29
RFE = 26.3
Fmod = 6 Hz
%TVE = 0.47
FE = 0.078
RFE = 3.6
F
s
= 30 Hz
Fmod = 6 Hz
%TVE = 0.56
FE = 0.077
RFE = 3.9
Fmod = 3 Hz
%TVE = 0.12
FE = 0.019
RFE = 0.43
F
s
= 20 Hz
Fmod = 4 Hz
%TVE = 0.35
FE = 0.047
RFE = 2.04
Fmod = 2 Hz
%TVE = 0.06
FE = 0.008
RFE = 0.13
F
s
= 15 Hz
Fmod = 3 Hz
%TVE = 0.34
FE = 0.030
RFE = 0.75
Fmod = 1.5 Hz
%TVE = 0.03
FE = 0.005
RFE = 0.05
F
s
= 12 Hz
Fmod = 2.4 Hz
%TVE = 0.34
FE = 0.021
RFE = 0.44
Fmod = 1.2 Hz
%TVE = 0.02
FE = 0.003
RFE = 0.03
F
s
= 10 Hz
Fmod = 2
%TVE = 0.34
FE = 0.017
RFE = 0.29
Fmod = 1 Hz
%TVE = 0.01
FE = 0.002
RFE = 0.02
B.4.1.3 Response to amplitude modulation only
Though IEEE Std C37.118.1-2011 does not specify AM only testing, the graphs of the PMU response
should be useful to the practitioner.
Note that there are no plots versus frequency. That is because there is no frequency shift to AM only.
It is interesting to note that the maximum modulation frequency may expose the worst-case TVE.
Figure B.15 shows a plot of TVE at F
s
= 60 FPS with a 12 Hz modulation frequency:
0 0.2 0.4 0.6 0.8 1
0
0.5
1
1.5
Time (seconds)
T
V
E
(
%
)
TVE vs Time
Figure B.15—TVE for AM at 12 Hz, index 0.1 at F
s
= 60 FPS
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
73
However at 11.9 Hz, it is apparent that the TVE goes above where it was at 12 Hz (see Figure B.16):
0 0.2 0.4 0.6 0.8 1
0
0.5
1
1.5
Time (seconds)
T
V
E
(
%
)
TVE vs Time
0 1 2 3 4
0
0.5
1
1.5
Time (seconds)
T
V
E
(
%
)
TVE vs Time
Figure B.16—TVE for AM at 11.9 Hz, 0.1 index at F
s
= 60 FPS
The second plot shows once again why it is important to make sure enough samples are gathered to help
ensure that maximum TVE is reached.
B.4.2 Dynamic response to linear frequency ramp test signals
B.4.2.1 M class TVE response to frequency ramps
In B.3, a plot from a frequency ramp was used to illustrate the impact of magnitude roll off of the M class
reference model’s filter on TVE. For reporting rates of F
s
below 30 FPS, the TVE is above 1% until the
input frequency reaches reporting rate divided by 5 (F
s
/5). This is the basis for the M class limits of
IEEE Std C37.118.1-2011. See Figure B.17.
55 60 65
0
0.5
1
Frequency
T
V
E
(
%
)
TVE vs Frequency
Figure B.17—TVE versus time and frequency for 1Hz frequency ramp from
–5hz to +5hz at F
s
= 30 FPS
The primary contributor to the M class TVE is the magnitude roll off of the M class reference model’s
filter, which is not required to be frequency compensated. Figure B.18 shows graphs of magnitude and
Figure B.19 shows phase error over the ramp:
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
74
0 2 4 6 8 10
-5
0
5
10
x 10
-3
Time (seconds)
M
a
g
n
i
t
u
d
e
e
r
r
o
r
Magnitude Error vs Time
55 60 65
-5
0
5
10
x 10
-3
Frequency
M
a
g
n
i
t
u
d
e
E
r
r
o
r
Magnitude Error vs Frequency
Figure B.18—Magnitude error versus time and frequency for 1 Hz frequency ramp from
–5Hz to +5Hz at F
s
= 30 FPS
0 2 4 6 8 10
-0.1
-0.05
0
0.05
0.1
Time (seconds)
P
h
a
s
e
e
r
r
o
r
(
d
e
g
r
e
e
s
)
Phase Error vs Time
55 60 65
-0.1
-0.05
0
0.05
0.1
Frequency
P
h
a
s
e
E
r
r
o
r
(
d
e
g
r
e
e
s
)
Phase Error vs Frequency
Figure B.19—Phase error versus time and frequency for 1 Hz frequency ramp from
–5Hz to +5Hz at F
s
= 30 FPS
B.4.2.2 M class frequency and RFE response to frequency ramps
The primary contributor to PMU reference model frequency and ROCOF error under ramp test is the small
offset of the frequency measurement from the time stamp. See Figure B.20 and Figure B.21.
Figure B.20—Frequency error versus time and frequency
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
75
0 2 4 6 8 10
-4
-2
0
2
x 10
-5
Time (seconds)
R
O
C
O
F
e
r
r
o
r
ROCOF Error vs Time
55 60 65
-4
-2
0
2
x 10
-5
Frequency
R
O
C
O
F
E
r
r
o
r
ROCOF Error vs Frequency
Figure B.21—ROCOF error versus time and frequency
The ROCOF error for frequency ramp tests is very low.
B.4.2.3 P class TVE response to frequency ramps
The P class reference PMU model requires magnitude compensation for the frequency roll off of the filter.
Since magnitude roll off is the main contributor to TVE in the M class model, the P class model responds to
frequency ramps with lower TVE (see Figure B.22):
0 2 4 6 8 10
0
0.05
0.1
0.15
0.2
Time (seconds)
T
V
E
(
%
)
TVE vs Time
55 60 65
0
0.05
0.1
0.15
0.2
Frequency
T
V
E
(
%
)
TVE vs Frequency
Figure B.22—TVE versus time and frequency for P class frequency ramp
Despite the improved TVE performance of the P class PMU, the bandwidth (BW) is limited to ±2Hz for the
ramp test. This gives even better TVE. See Table B.8.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
76
Table B.8—Response to frequency ramp rate = 1 Hz/s, ramp range = ±BW
M class
P class same for all F
s
BW = ±2 Hz
F
s
= 120 Hz
BW = 5 Hz
%TVE = 1.02
FE = 0.003
RFE = 3e-008
%TVE = 0.04
FE = 0.003
RFE = 6e-008
F
s
= 60 Hz
BW = 5 Hz
%TVE = 0.92
FE = 0.003
RFE = 1.6e-006
—
F
s
= 30 Hz
BW = 5 Hz
%TVE = 0.97
FE = 0.003
RFE = 3.4e-005
—
F
s
= 20 Hz
BW = 4 Hz
%TVE = 0.93
FE = 0.004
RFE = 0.0003
—
F
s
= 15 Hz
BW = 3 Hz
%TVE = 1.05
FE = 0.004
RFE = 0.0008
—
F
s
= 12 Hz
BW = 2.4 Hz
%TVE = 0.96
FE = 0.005
RFE = 0.0021
—
F
s
= 10 Hz
BW = 2 Hz
%TVE = 1.13
FE = 0.007
RFE = 0.0042
—
B.4.3 Dynamic response to step changes in amplitude
When a step change in input occurs, the Annex C PMU model output amplitude is a finite impulse step
response centered around the 50% point of the step because the Annex C model uses a symmetrical FIR
filter centered around the time of each report.
IEEE Std C37.118.1-2011 limits the response time, which is the difference in time that the TVE leaves the
specified accuracy limit until it returns to and remains within the specified limit. In Figure B.23, the
response time is shown. Recall that the response time is the time that the TVE exceeds 1%. Figure B.24
shows the TVE during the step change in amplitude.
A minor difference may exist between the time reported for the instant a step change is applied and the
actual time, shown in Figure B.23 and Figure B.24 as a delay-time compensation error. The measurement
delay time compensation error can be very small.
IEEE Std C37.118.1-2011 also limits the maximum overshoot, which is a maximum percentage of the step
magnitude that the signal can reach before it settles to the final value. Figure B.23 shows this overshoot and
undershoot.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
77
Figure B.23—Measurement response to a 10% step change in magnitude
Figure B.24—TVE for a 10% step change in amplitude
B.4.4 Dynamic response to step changes in phase
When a step change in the input occurs, the Annex C PMU model output phase is a finite impulse step
response centered around the 50% point of the step. It makes no difference if the step change was
amplitude or phase because the Annex C model uses a symmetrical FIR filter centered around the time of
each report. Figure B.25 shows the response to a 90º step change in phase at time t = 0.
IEEE Std C37.118.1-2011 limits the response time, which is the difference in time that the TVE leaves the
specified accuracy limit until it returns to and remains within the specified limit. In Figure B.26, the
response time is the time elapsed between the vertical lines shown about one eighth of a second before and
one eighth of a second after the step change in phase. These are the times that the TVE exceeds the
specified limit of error in response to the step change in phase.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
78
Figure B.25—Measurement response to a 90° step change in phase—example
Figure B.26—TVE for a 90° step change in phase
It is worth pointing out that the error is not calculated in quite the conventional way, by taking the
difference between the measurement and the “true value” and finding a ratio. Because of the way TVE is
defined, the errors are Vee-shaped functions, such as shown in Figure 7 in 7.5.1.1 of this guide.
It is observed in 7.5.1.1 that the TVE is dominated by phase errors for phase errors greater than 0.5 degrees.
In fact, if the amplitude error is zero, the TVE due to phase error can be expressed as shown in
Equation (B.1):
TVE = 1.75 ε (B.1)
where ε is the phase error in degrees and the TVE is in percent.
In the case of a 90º phase shift, the error will be 45º at the center of the response, so the TVE will be nearly
79%, as shown in Figure B.26.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
79
B.5 Synchrophasor TIMEBASE and TVE
According to IEEE Std C37.118.1-2011 5.4.2 on reporting times:
For a reporting rate N frames per second (FPS) where N is a positive integer, the reporting
times shall be evenly spaced through each second with frame number 0 (numbered 0 thru
N–1) coincident with the UTC second rollover (e.g., coincident with a 1 PPS provided by
GPS). These reporting times (time tags) are to be used for determining the instantaneous
values of the synchrophasor as defined in 4.2.
Synchrophasors carry with them a TIMESTAMP that is specified to represent the absolute time of the
phasor measurement. The fractional part of the second is represented by a 24-bit unsigned integer called
FRACSEC. To recover the fraction of a second, the 24-bit unsigned integer is divided by the TIMEBASE,
which is a number transmitted in the CFG-2 message. A popular TIMEBASE is 1 million (10
6
). The
resolution of the fraction of a second will be the inverse of the TIMEBASE; typically 1/10
6
second or 1 μs.
Note that only 20 bits would be required to carry FRACSEC with a 10
6
TIMEBASE, so four FRACSEC
bits are typically unused. If all 24 bits were used, the resolution would be 16 times more precise.
A typical PMU with steady-state nominal frequency will output a very accurate phasor. However with a
time base of 10
6
, the time stamp for many of the measurements will be rounded off to 1 μs precision.
Consider a nominal frequency of 60 Hz and synchrophasor reporting rate (F
s
) of 30 reports per second:
Beginning with a report made on the UTC second, the tenth report will be made at 1/3 of a second and
twentieth report at 2/3 of a second. However the FRACSEC representations will be rounded to 0.333333
and 0.666667, for an error of minus and plus 1/3 μs, respectively. In fact, two out of every three
synchrophasor measurements at F
s
= 30 will have a round-off error of 1/3 μs.
Applications and PMU test systems may use different interpretations of the time tag. In this guide, in line
with IEEE Std C37.118.1-2011 5.4.2., it is recommended that the time tag in a PMU report is interpreted as
evenly spaced through each second, regardless of the resolution of the time tag. It is worthwhile to be
aware that some applications and test equipment might have used the digitally reported value of the time
tag, possibly affected by the truncation of the value.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
80
Annex C
(informative)
Effects of signal channels
C.1 Introduction
Even though a PMU is required by IEEE Std C37.118.1-2011 to have errors less than 1% (TVE), the
overall error in a synchrophasor measurement system may be larger due to other sources. In particular, the
errors in instrument transformers and the associated cabling may be significant. This annex examines such
sources of error.
From the measurement point of view, two observations must be made. First, such sources of error are what
is termed Type B uncertainty by metrologists. That means that no amount of repeated measurement and
data processing will reduce their effect. A value that is wrong because (for example) of a scaling error is
always wrong, no matter how many times the measurement is made.
Second, the source of error is external to the PMU hardware. A PMU that meets all the requirements of the
PMU standard (IEEE Std C37.118.1-2011) in terms of its accuracy can still furnish synchrophasor readings
that (for example) do not agree with readings from another PMU in a nearby part of the power system, even
if both are compliant with the same standard. The differences are not the result of PMU problems. They are
system differences, caused by the differences of the instrument transformers, the cabling, and the
transformer burdens.
Established standards set classes for instrument transformers based on application-dependent limits in the
phase and magnitude errors. These standards were developed when there were, for all practical purposes,
two applications for the instrumentation transformer: metering and protection. For metering, with the
implication of energy billing, it was crucial to establish performance specifications that gave very accurate
results provided the quantities being measured were close to “nominal.” The frequency could be expected
to be very close to the nominal value of power frequency, for example, and no significant attention had to
be paid to the frequency response of the device. Typically, accurate performance within 1% or 2% of the
rated power frequency was required. For a protection application, on the other hand, conditions were
expected to be at times far from nominal. A CT was expected to furnish good current values even if there
was a very large overcurrent, for example, 20 times greater than the rated current. In this situation, the
stringent accuracy requirements of metering were relaxed.
It may be worthwhile to note that the 2011 revision to IEEE Std C37.118.1-2011 (as compared to its
predecessor IEEE Std C37.118-2005) effectively acknowledges different requirements for protection and
metering applications and sets different dynamic performance requirements for the “P class” and “M class”
synchrophasor data.
PMUs are expected to furnish good results in a wide variety of circumstances, and yet at the time this is
written (2012) no particular PMU-application standards have been written to specify the performance of the
instrument transformers that they are connected to. It is common when a PMU is installed to use existing
voltage and CTs. Often little consideration is given to the suitability of the transformer characteristics or to
the cabling system that connects it to the PMU because there are few cost-effective options.
Testing performed on instrument transformers prior to installation determines accurately the phase and
magnitude errors of an individual device or of a type under certain conditions. In principle, such errors, if
known, and if constant as a function of time and environmental conditions, could be stored in the PMU and
used to correct the readings if necessary. It is important to keep in mind that some of these corrections may
be suitable in only a limited or narrow range of conditions.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
81
C.2 Instrument transformer standards
The magnitude and phase errors permitted by some current standards are presented for the convenience of
the reader in Table C.1, Table C.2, and Table C.3. They are representative as of November 2012:
Table C.1—Maximum magnitude and phase error for ANSI class type CTs
(IEEE Std C57.13)
ANSI CT type Load current
(%)
Max. magnitude
error (%)
Max. phase error
(degrees)
Relaying 10 to 2000 10 Not specified
10 2.4 2.08 Metering 1.2
100 1.2 1.04
10 1.2 1.04 Metering 0.6
100 0.6 0.52
10 0.6 0.52 Metering 0.3
100 0.3 0.26
Table C.2—Maximum magnitude and phase error for ANSI class VTs (IEEE Std C57.13)
ANSI VT type Max. magnitude error
(%)
Max. phase error
(degrees)
Relaying 10 Not specified
Metering 1.2 1.2 2.08
Metering 0.6 0.6 1.04
Metering 0.3 0.3 0.52
Table C.3—Maximum magnitude and phase errors of IEC class CTs (IEC 60044-x series)
IEC CT type Load
Max. magnitude error
(%)
Max. phase error
(degrees)
100% 3 Not specified
Relay type 10P
Max. limit 10 (composite error) Not specified
100% 1 1
Relay type 5P
Max. limit 5 (composite error) Not specified
5% 3.0 3.0
20% 1.5 1.5
100% 1.0 1.0
Metering type
1.0 accuracy
120% 1.0 1.0
5% 1.5 1.5
20% 0.75 0.75
100% 0.5 0.5
Metering type
0.5 accuracy
120% 0.5 0.5
5% 0.75 0.5
20% 0.35 0.25
100% 0.2 0.167
Metering type
0.2 accuracy
120% 0.2 0.167
5% 0.4 0.250
20% 0.2 0.133
100% 0.1 0.083
Metering type
0.1 accuracy
120% 0.1 0.083
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
82
The maximum error values given in this summary are, of course, all applicable at the nominal power
frequency. No guidance is given if the frequency is not close to nominal; for example, IEC 60044-7
specifies “the standard reference range of frequencies shall be from 99% to 101% of the rated frequency for
the accuracy classes for measurement, and from 96% to 102% for the accuracy classes for protection.” That
is an unfortunate circumstance, because it has been established experimentally
21
that many transformers
exhibit resonance conditions with their cabling and burden that can result in very significant performance
challenges. Certainly, the errors can greatly exceed those allowed in the PMU itself.
C.3 Measured results on instrument transformers
The solution for the challenge of not knowing the errors caused by the instrument transformers and/or the
cabling involves field calibration methods that are beyond the scope of this guide. Nevertheless, it is worth
examining the problem in enough detail to understand, at least approximately, how serious a problem it
might be.
Figure C.1 shows a simplified equivalent circuit of the arrangement shown as a physical picture in Figure 1
of this guide.
Figure C.1—Simplified equivalent circuit of VT and CT feeding PMU
The instrument transformers shown in Figure C.1 will typically have significant inter-winding capacitances
and capacitances to ground; these are not shown in the figure. Such features complicate the response
further.
Figure C.2 shows the frequency response (amplitude and phase) of a commercial wound-type VT as
measured and reported in Meliopoulos et al. [B50].
21
See for example Meliopoulos et al. [B50].
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
83
Figure C.2—Magnitude and phase error of a VT
(adapted from Meliopoulos et al. [B50])
The resonance exhibited by the particular device whose response is shown here is thought to be broadly
representative. At and near power frequency, the transformer is within the limits established by standards.
The magnitude error and phase error are both negligible (i.e., within the allowed errors specified on the VT
nameplate) at power frequency (as expected for VTs that are built and installed properly). Nor, according to
these authors, are the errors significantly affected by the transformer burden, provided that the burden is
within a reasonable range (as specified on the VT nameplate).
The pronounced resonance peak in the response above 1 kHz should be noted. Nevertheless, for most PMU
applications, the filtering in the PMU should reduce the effect of this resonance to insignificance.
The other common kind of VTs, coupling capacitor voltage transformers (CCVTs), are dismissed in
Meliopoulos et al. [B50] as not being useful over the frequency range of interest (for harmonics
measurement). Some interesting field measurements have been made on CCVTs, nevertheless, see for
example Fernandez et al. [B15]. An example is shown in Figure C.3.
Readers who are familiar with the behavior of the sort of tuned coupled inductors used in the IF
transformers of radios may recognize a familiar shape. It certainly seems that the circuit parameters have
been adjusted so that the amplitude and phase response are flat across the region of power frequency
(60 Hz in this example). However, the rest of the response is likely to give significant difference, even at
just the second harmonic.
It may be pointed out that CTs do not, in general, have such pronounced peaks in their response. A device
tested in Meliopoulos et al. [B50], for example, had only 0.17 degree error across the frequency range
tested.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
84
Figure C.3—Magnitude and phase error of CCVT versus frequency
(adapted from Fernandez et al. [B15])
C.4 Conclusions
While PMUs are required to meet the requirements on TVE as per IEEE Std C37.118.1-2011, there is no
such requirement on a complete synchrophasor measurement system as installed. Significant errors can
arise, particularly in systems with capacitive voltage transformers (CVTs), that can be very difficult to
determine without field calibration or verification. It is important to keep in mind that if differences are
observed between the results of measurements from PMUs that are expected to give similar results, the
cause might be in the instrumentation channels, including instrument transformers, the cabling or the
burden.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
85
Annex D
(informative)
Examples of instrumentation channel impact on accuracy using
approximate models
D.1 General
This annex provides a few examples of instrumentation channels. In selecting the instrumentation channels,
the following voltage measurement technologies have been considered:
⎯ Wound-type potential transformers (PTs)
⎯ Capacitive coupled voltage transformers (CCVTs)
⎯ Capacitive voltage dividers (CVDs)
⎯ Resistive voltage dividers (RVDs)
Generally voltage dividers provide accurate wide bandwidth measurements with flat frequency response;
however, these technologies are usually limited to laboratory grade instrumentation. In high-voltage power
system applications (outdoor relaying and metering units), PTs and CCVTs are commonly used,
accordingly examples below include PTs and CCVTs only.
Current transformers (CTs) are typically wire-wound ferromagnetic core type transformers with current
output. The ratings depend on the maximum permissible current that can be developed on the secondary
before saturation occurs. For PMU type applications, saturation is generally not a concern as the current
through the CT is typically much lower than the capability of the CT.
In the following examples, some basic models for instrumentation channels have been defined. The
parameters of these instrumentation channel models have been selected to represent near practical systems.
They have been constructed by considering combinations of the following important instrumentation
channel parameters: CT types, VT types, CCVT types, instrumentation cables (type, shielded/unshielded,
length). Instrumentation channels based on optical CTs and VTs have not been considered in this annex.
The instrumentation channel examples listed in Table D.1 are based on data provided from measurements
in the field and laboratory settings. Using these sample instrumentation channels, the variability and
amount of errors potentially introduced are evaluated and tabulated here. Please note that the examples
modeled here are a very small subset of what may actually be installed in the field. Also, the models may
not be detailed or accurate enough to represent actual instrumentation channels in some cases.
Nevertheless, the results provided in the tables in this annex are simply meant to provide an appreciation
for the approximate impact of various parameters on measurement variability.
It is also important to note that the resolution of the numbers provided in the tables in this annex may be
significantly finer than the accuracy of the models and the resultant numbers. Nevertheless, this higher
resolution shows the level of variability and dependency (albeit sometimes very small) as certain system
parameters change.
D.2 Description of typical instrumentation channels
In this annex, samples of instrumentation channel models are provided, taking into account the possible
effects of the following parameters:
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
86
⎯ Instrument transformer (ratio, impedance, parasitic capacitance)
⎯ Instrumentation cable type, size, and length
⎯ Attenuators, if present
⎯ Burden (e.g., a PMU’s input impedance)
The error characterization for each one of these channels is as follows.
⎯ VT instrumentation channels: The voltage magnitude error is defined as the percentage difference
between the primary voltage of the PT and voltage seen by the relay or PMU scaled by a nominal
factor k
nominal
, as show in Equation (D.1):
% 100 %
,
×
⋅ −
=
primary
relay nominal primary
magnitude V
V
V k V
Error
where:
nominal relay
nominal primary
nominal
V
V
k
,
,
=
(D.1)
In addition, the phase error is defined as the phase angle difference between the reading at the
primary side of the PT and voltage reading seen by the PMU.
⎯ CT instrumentation channels: The current magnitude error is defined as the percentage difference
between the primary current of the CT and current seen by the relay or PMU scaled by a nominal
factor k, as shown in Equation (D.2):
% 100 %
,
×
⋅ −
=
primary
relay primary
magnitude I
I
I k I
Error
(D.2)
In addition, the phase error is defined as the phase angle difference between the current reading at
the primary side of the CT and current reading seen by the PMU.
D.3 Potential transformer instrumentation channel error
A PT instrumentation channel example has been analyzed using a model, and the estimated errors have
been tabulated. These results provide an approximate quantitative measure of the impact of various
elements of the PT instrumentation channel on the quality and accuracy of the synchrophasor data reported
by the PMU.
The PT instrumentation channel model is shown in Figure D.1. The model includes the effect of parasitic
capacitances and instrumentation cable. The PMU burden is also accounted for in the model by a large
resistance (typically on the order of few thousand ohms).
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
87
Figure D.1—Sample PT instrumentation channel model parameters
The simulation is performed for different values of the model parameters. The errors are quantified in terms
of magnitude and phase errors.
D.3.1 Example, a 199kV/69V PT
The base case model parameters for this example are shown in Table D.1. The calculated error results are
presented in Table D.2, Table D.3, Table D.4, and Table D.5. The parameters not shown in the results
tables can be assumed to be the same as the base case values.
Table D.1—Base-case parameters for 199 kV/69 V PT instrumentation
channel model simulation
Parameter description Value
Nominal ratio 199 kV/69V
PT VA rating 30 kVA
PT resistance 0.01 pu
PT leakage reactance X
ℓ
0.025 pu
PT nominal core loss 0.005 pu
PT nominal magnetization current 0.005 pu
PT parasitic capacitance C
p
50 nF
Instrumentation cable type #10 copper pair
Instrumentation cable length ℓ 60 m (200 ft)
PMU burden resistance R
B
10 kΩ
Table D.2—199 kV/69 V PT instrumentation channel voltage magnitude error
in % for different burden resistances and cable lengths
Instrumentation cable length ℓ
Burden resistance R
B
(Ω)
30 m (100 ft) 60 m (200 ft) 150 m (500 ft) 240 m (800 ft)
5 k 0.009 0.016 0.22 0.52
10 k 0.009 0.015 0.19 0.48
20 k 0.009 0.014 0.18 0.41
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
88
Table D.3—199 kV/69 V PT instrumentation channel voltage phase error
in degrees for different burden resistances and cable lengths
Instrumentation cable length ℓ
Burden resistance R
B
(Ω)
30 m (100 ft) 60 m (200 ft) 150 m (500 ft) 240 m (800 ft)
5 k 0.121 0.212 0.425 0.638
10 k 0.115 0.198 0.411 0.614
20 k 0.115 0.196 0.399 0.589
For this high-voltage PT, Table D.2 and Table D.3 show that the impact of burden resistance combined
with cable length can be significant, especially for longer cable lengths. We can see estimated voltage
magnitude errors close to 0.5% TVE for longer cable runs (240 m or longer is a typical cable run in many
230 kV and higher voltage substations). Phase angle errors are also significant and can exceed 0.57° (1%
TVE) for longer cable runs.
Table D.4—199 kV/69 V PT instrumentation channel voltage magnitude error
in % for different parasitic capacitances and cable lengths
Instrumentation cable length ℓ (ft)
Parasitic capacitance C
p
(nF)
30 m (100 ft) 60 m (200 ft) 150 m (500 ft) 240 m (800 ft)
25 0.008 0.012 0.17 0.41
50 0.009 0.015 0.19 0.48
75 0.015 0.026 0.22 0.61
Table D.5—199 kV/69 V PT instrumentation channel voltage phase error in degrees for
different parasitic capacitances and cable lengths
Instrumentation cable length ℓ (ft)
Parasitic capacitance C
p
(nF)
30 m (100 ft) 60 m (200 ft) 150 m (500 ft) 240 m (800 ft)
25 0.090 0.185 0.382 0.538
50 0.115 0.198 0.411 0.614
75 0.196 0.235 0.479 0.779
Similarly, Table D.4 and Table D.5 show that the impact of parasitic capacitances combined with cable
length can be significant, especially for longer cable lengths. We can see estimated voltage magnitude
errors close to 0.6% TVE for longer cable runs. The estimated phase angle errors at longer cable length are
on the order of 1% TVE (0.6°), noticeably higher than voltage magnitude errors.
D.4 Current transformer instrumentation channel error
In this subclause, a CT instrumentation channel example has been analyzed using a model, and the
estimated errors have been tabulated below. These results provide an approximate quantitative measure of
the impact of various elements of the CT instrumentation channel on the quality and accuracy of the
synchrophasor data reported by the PMU.
The CT instrumentation channel model is shown in Figure D.2. The model includes the parasitic
capacitances and instrumentation cable. The PMU burden is also accounted for in the model (usually
0.1 Ω or 0.14 Ω).
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
89
R
PMU
Instrumentation Cable
CT
Phase Conductor
C
p
Figure D.2—Sample CT instrumentation channel model parameters
The simulation is performed for different values of the model parameters. The error is quantified in terms
of magnitude and phase error.
D.4.1 Example, a 3000/5 A/A CT
The base case model parameters shown in Table D.6 represent the model used here for a 3000/5A/A (600:1
ratio) CT. The value of the current magnitude error is calculated using this model to be less than 0.07%,
effectively independent of the cable length, the burden resistance, and the parasitic capacitance values. The
calculated phase error results are shown in Table D.7 and Table D.8; burden resistance and parasitic
capacitance (in a reasonable range) have negligible impact on phase errors. Longer cable lengths appear to
be the only significant factor affecting the estimated synchrophasor phase errors, amounting to 0.2° to 0.3°
(close to 0.5% TVE) for a 240 m cable length.
Table D.6—Base-case parameters for 3000/5 A/A CT instrumentation channel
model simulation
Parameter description Value
Nominal ratio 3000/5 A
CT VA rating 500 VA
CT resistance 0.001 pu
CT leakage reactance X
ℓ
0.003 pu
CT nominal core loss 0.004 pu
CT nominal magnetization current 0.005 pu
CT parasitic capacitance C
p
5 nF
Instrumentation cable type #10 pair
Instrumentation cable length ℓ
60 m (200 ft)
PMU burden resistance R
B
0.1 Ω
Table D.7—3000/5 A/A CT instrumentation channel current phase error in degrees for
different burden resistances and cable lengths
Instrumentation cable length ℓ (ft)
Burden resistance R
B
(Ω) 30 m
(100 ft)
60 m
(200 ft)
150 m
(500 ft)
240 m
(800 ft)
0.1 0.03
o
0.04
o
0.09
o
0.23
o
0.14 0.03
o
0.04
o
0.09
o
0.23
o
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
90
Table D.8—3000/5 A/A CT instrumentation channel current phase error in degrees for
different parasitic capacitances and cable lengths
Instrumentation cable length ℓ (ft)
Parasitic capacitance C
p
(nF) 30 m
(100 ft)
60 m
(200 ft)
150 m
(500 ft)
240 m
(800 ft)
1 0.02
o
0.03
o
0.08
o
0.20
o
5 0.03
o
0.04
o
0.09
o
0.23
o
10 0.05
o
0.08
o
0.14
o
0.32
o
D.5 CCVT instrumentation channel error
A sample CCVT instrumentation channel has been analyzed using a model and the estimated errors have
been tabulated. The results provide an approximate quantitative measure of the impact of various elements
of the CCVT instrumentation channel on the quality and accuracy of the synchrophasor data reported by
the PMU.
The basic CCVT configuration is illustrated in Figure D.3. The capacitive divider is formed by capacitors
C1 and C2. In typical high-voltage CCVTs (115 kV to 500 kV), the capacitor values are selected so that the
voltage at the tap point A is in the order of 4 kV to 10 kV. The transformer scales this voltage to standard
instrumentation voltage level (e.g., 69 V or 120 V). The output of the transformer is connected to a burden
via instrumentation cable (usually a multi-conductor cable, sometimes with an overall shield). The burden
represents the PMU or relay input impedance, which is typically resistive. Since the interaction of the
capacitive divider and the resistive burden introduces considerable phase shift, a series inductor L is added
to compensate for the divider output capacitance. A CCVT equivalent model is shown in Figure D.4. The
CCVT model parameters are summarized in Table D.9.
Figure D.3—CCVT physical circuit
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
91
Figure D.4—CCVT equivalent model
Table D.9—CCVT parameters summary
Parameter description Schematic reference
Upper capacitor size C
1
Lower capacitor size C
2
Drain inductor L
D
Compensating reactor inductance L
C
Compensating reactor resistance R
C
Burden resistance R
B
Ferroresonance suppression damping resistor R
F
Ferroresonance suppression circuit inductor L
F
Ferroresonance suppression circuit capacitor C
F
Parasitic capacitance C
P
In designing a CCVT, the inductor L is selected by setting the sum of the equivalent capacitive reactance
and the inductive reactance to zero at the power frequency. Note that the Thevenin equivalent capacitive
reactance at point A is the sum of the upper and lower leg capacitances, C
1
+ C
2
, thus, as shown in
Equation (D.3):
( )
0
1
2 1
=
+
+
C C j
L j
ω
ω
(D.3)
Or equivalently:
( )
2 1
2
1
C C
L
+
=
ω
The reactor L
D
, known as the drain reactor, is for the purpose of power line carrier filtering and may be
optionally shorted by a manual switch. For the purpose of low frequency analysis (0 to 1 kHz) this reactor
has negligible effect and is ignored.
In standard CCVT designs, typical values of capacitor dividers are selected so that the sum of the
capacitances C
1
+ C
2
is in the order of 100 nF. The corresponding compensating reactor inductance for
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
92
60 Hz power frequency is in the order of 70 Henries. A reactor of such large inductance must have a
magnetic core, and will also have a substantial resistance. As a result, the compensating reactor is also
subject to saturation. Furthermore, its resistance makes the CCVT transformation ratio dependent on the
burden resistance. Recently, CCVTs with larger capacitance values and smaller compensating reactors have
become available. These devices exhibit improved transient response, as well as less sensitivity to burden
resistance.
The interaction of the transformer saturation characteristics with the divider capacitance makes this circuit
subject to ferroresonance. Specifically, during transients, a resonance may occur at the frequency
determined by the transformer magnetizing reactance and the circuit equivalent capacitance. This results in
overvoltages developing across the transformer that drive the core into saturation. This nonlinear high-
amplitude oscillation causes severe measurement errors and can damage the circuit components. For this
reason, CCVTs include a ferroresonance suppression circuit, usually located across the transformer
secondary winding (Z
F
in Figure D.3). Several ferroresonance suppression circuit topologies are presently
in use. These circuits are considered proprietary by some manufacturers, and thus the circuit details are not
readily available. However, two generic circuit models capture the basic behavior of these filters: The
“active” suppression circuit illustrated in Figure D.5, and the “passive” suppression circuit illustrated in
Figure D.6.
The active suppression circuit inductor L
F
and capacitor C
F
are tuned to the power frequency. Thus during
normal steady-state operation, the impedance of the L
F
//C
F
branch is very large and the suppression circuit
draws negligible current. During transients, the impedance of L
F
//C
F
branch is lower and thus the resonating
energy is dissipated through the filer resistor R
F
.
The passive suppression circuit (see Figure D.6) consists of a saturable core reactor L
F
and a damping
resistor R
F1
connected at the center tap of the transformer secondary, plus a spark gap in series with a
second damping resistor R
F2
.
Figure D.5—Equivalent circuit of a CCVT with active ferroresonance suppression circuit
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
93
Figure D.6—Equivalent circuit of a CCVT with passive ferroresonance suppression circuit
During steady-state 60 Hz operation, both of these filter circuits have negligible effect on the CCVT
response. However during transients they generally prolong the CCVT transient response. It has been
shown that CCVTs with passive ferroresonance suppression circuits have better transient response
characteristics (error decays to negligible levels faster).
The CCVT simulation model used here is that shown in Figure D.4. The model explicitly represents the
upper and lower capacitors, the drain inductor, the compensating reactor inductance and resistance,
ferroresonace suppression damping, the parasitic capacitance, and the burden resistance.
D.5.1 Example, a 288 kV:120 V CCVT
The base case parameter values for this example are given in Table D.10. The simulation is repeated for
various instrumentation cable lengths. In addition, different values of burden resistance, and different
inductance and capacitance values are used to calculate CCVT phase errors under different conditions. The
calculated phase errors are presented in Table D.11 and Table D.12. Table D.11 shows the results of a
parametric analysis with respect to burden resistance and instrumentation cable length. Table D.12 shows
the results of a parametric analysis with respect to CCVT component parameter inaccuracies. Specifically
the varied parameters are the compensating reactor inductance and the capacitive divider capacitance.
Table D.10—Base-case CCVT instrumentation channel simulation parameters
Parameter description Schematic reference Value
CCVT capacitance class — Normal
Input voltage — 288 kV
Output voltage — 120 V
Upper capacitor size C
1
1.407 nF
Lower capacitor size C
2
99.9 nF
Drain inductor L
D
2.65 mH
Compensating reactor inductance L
C
68.74 H
Compensating reactor resistance R
C
3000 Ω
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
94
Table D.10—Base-case CCVT instrumentation channel simulation parameters (continued)
Parameter description Schematic reference Value
Burden resistance R
B
200 Ω
Ferroresonance suppression damping resistor R
F
70 Ω
Ferroresonance suppression circuit inductor L
F
0.398 H
Ferroresonance suppression circuit capacitor C
F
17.7 uF
Cable type — #10 pair
Cable length — 100 ft
Transformer power rating — 300 VA
Transformer voltage rating — 4 kV/120 V
Leakage reactance — 3%
Parasitic capacitance C
P
500 pF
Table D.11—CCVT phase error (in degrees) versus burden resistance and cable length
Cable length
Burden resistance (Ω)
3 m (10 ft) 300 m (1000 ft) 600 m (2000 ft)
50 1.33 1.44 1.55
100 0.7 0.77 0.85
200 0.37 0.41 0.45
400 0.2 0.22 0.25
1000 0.1 0.11 0.12
Table D.12—CCVT phase error (in degrees) versus capacitive divider capacitance
and compensating reactor inductance
Inductance error (%)
Capacitance error (%)
0% 1% 5%
0 0.10 0.12 0.18
–1 0.11 0.13 0.19
–5 0.17 0.19 0.24
The results in Table D.11 show that low burden resistances (higher burdens) can have a very significant
impact on synchrophasor measurement accuracy. Nevertheless, since most modern PMUs have a fairly
high input impedance (more than several kilohms), the impact of burden on voltage phasor accuracy is
expected to be minimal.
The results also show that changes in the capacitive divider’s capacitance (e.g., due to aging or temperature
changes) or inaccuracies in selecting the compensating reactor’s inductance can have noticeable impacts on
phase accuracy. The impact of capacitive divider’s capacitance change on voltage magnitude accuracy is
not shown here since they have a commonly well-understood near-linear relationship; e.g., a 2% change in
C
1
’s capacitance will result in nearly 2% error in CCVT voltage magnitude error (ratio error).
Additional examples and references can be found in Meliopoulos et al. [B49].
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
IEEE Std C37.242-2013
IEEE Guide for Synchronization, Calibration, Testing, and Installation of Phasor Measurement Units (PMUs)
for Power System Protection and Control
Copyright © 2013 IEEE. All rights reserved.
95
Annex E
(informative)
Example of commissioning tests and measurements
Various tests can be performed or repeated at the time of commissioning a PMU. The extent and
complexity of the tests depend on a number of factors, most importantly, the functions that the PMU is
serving and the importance of its accuracy and reliability. In most cases during commissioning, PMUs are
tested for simple accuracy checks (approximate performance, e.g., within 1% to 2% of what is expected),
polarity checks, data format (according to a standard or manufacturer’s specification) and communication
quality.
Figure E.1 shows setup for performance testing by a user. This setup can be used in a laboratory or in the
field for measuring the performance (including approximate accuracy) of a PMU. The test setup uses 1 PPS
signal from GPS to synchronize the 3-phase signal generator software and produce –90° positive sequence
angle when the positive-going zero crossing of A-phase is aligned with this 1 PPS. At nominal frequency of
60 Hz (or 50 Hz) this zero crossing does not change (nor does it change for exact integer off-nominal
frequencies of 61 Hz or 59 Hz, etc.).
Figure E.1—PMU performance test setup example
Various kinds of tests can be performed, including steady-state accuracy tests, dynamic performance tests
(if the signal generator supports the step and ramp functions), and various protocol tests and delay
measurements.
A connection testing software (an off-the-shelf tool) can act like a local PDC and be used to confirm
correct expected magnitudes and angles. Final results may be taken from a PDC located at a control center
at a distance in order to include various types of delays such as the delays within the communications
network.
Authorized licensed use limited to: UNIVERSIDADE FEDERAL DE SANTA CATARINA. Downloaded on April 24,2013 at 17:44:54 UTC from IEEE Xplore. Restrictions apply.
