Ieee std 1344 1995(r2001)

Recognized as an
American National Standard (ANSI)
The Institute of Electrical and Electronics Engineers, Inc.
345 East 47th Street, New York, NY 10017-2394, USA
Copyright © 1995 by the Institute of Electrical and Electronics Engineers, Inc.
All rights reserved. Published 1995. Printed in the United States of America
ISBN 1-55937-693-7
No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior
written permission of the publisher.
IEEE Std 1344-1995(R2001)
IEEE Standard for Synchrophasors
for Power Systems
Sponsor
Power System Relaying Committee
of the
IEEE Power Engineering Society
Approved 12 December 1995
IEEE Standards Board
Reaffirmed 17 March 2001
IEEE-SA Standards Board
Abstract: The synchronizing input and the data output for phasor measurements made by
substation computer systems is discussed. Processes involved in computing phasors from sampled
data, data-to-phasor conversions, and formats for timing imputs and phasor data output from a
Phasor Measurement Unit (PMU) are also addressed.
Keywords: anti-aliasing, digital computer based measurement, Global Positioning System (GPS),
Nyquist rate, phase lock, phasor, satellites, synchronism, synchronized phasor, time input
Authorized licensed use limited to: Centro De Pesquisas De Energia Eletrica Cepel. Downloaded on February 29,2016 at 17:32:51 UTC from IEEE Xplore. Restrictions apply.

IEEE Standards documents are developed within the Technical Committees of the IEEE Societies and the Standards
Coordinating Committees of the IEEE Standards Board. Members of the committees serve voluntarily and without
compensation. They are not necessarily members of the Institute. The standards developed within IEEE represent a
consensus of the broad expertise on the subject within the Institute as well as those activities outside of IEEE that have
expressed an interest in participating in the development of the standard.
Use of an IEEE Standard is wholly voluntary. The existence of an IEEE Standard does not imply that there are no other
ways 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 Þve years for revision or reafÞrmation. When a document is more
than Þve years old and has not been reafÞrmed, it is reasonable to conclude that its contents, although still of some
value, do not wholly reßect the present state of the art. Users are cautioned to check to determine that they have the
latest edition of any IEEE Standard.
Comments for revision of IEEE Standards are welcome from any interested party, regardless of membership afÞliation
with IEEE. Suggestions for changes in documents should be in the form of a proposed change of text, together with
appropriate supporting comments.
Interpretations: Occasionally questions may arise regarding the meaning of portions of standards as they relate to
speciÞc applications. When the need for interpretations is brought to the attention of IEEE, the Institute will initiate
action to prepare appropriate responses. Since IEEE Standards represent a consensus of all concerned interests, it is
important to ensure that any interpretation has also received the concurrence of a balance of interests. For this reason
IEEE and the members of its technical committees are not able to provide an instant response to interpretation requests
except in those cases where the matter has previously received formal consideration.
Comments on standards and requests for interpretations should be addressed to:
Secretary, IEEE Standards Board
445 Hoes Lane
P.O. Box 1331
Piscataway, NJ 08855-1331
USA
Authorization to photocopy portions of any individual standard for internal or personal use is granted by the Institute
of Electrical and Electronics Engineers, Inc., provided that the appropriate fee is paid to Copyright Clearance Center.
To arrange for payment of licensing fee, please contact Copyright Clearance Center, Customer Service, 222 Rosewood
Drive, Danvers, MA 01923 USA; (508) 750-8400. Permission to photocopy portions of any individual standard for
educational classroom use can also be obtained through the Copyright Clearance Center.
Note: Attention is called to the possibility that implementation of this standard may require use of subject matter
covered by patent rights. By publication of this standard, no position is taken with respect to the existence or
validity of any patent rights in connection therewith. The IEEE shall not be responsible for identifying all patents
for which a license may be required by an IEEE standard or for conducting inquiries into the legal validity or
scope of those patents that are brought to its attention.

iii
Introduction
(This introduction is not a part of IEEE Std 1344-1995, IEEE Standard for Synchrophasors for Power Systems.)
Digital computer-based measurement, protection, and control systems have become common features of electric
power substations. These systems use sampled data to compute various quantities such as voltage and current phasors.
Phasors are of great value in many functions, and their utility is increased further if they can be referenced to a
common time base. This can be accomplished if the signal input processes at various measuring sites are synchronized
with respect to each other. Simultaneous measurement sets derived from synchronized phasors provide a vastly
improved method for tracking power system dynamic phenomena for improved power system monitoring, protection,
operation, and control.
The standard deÞnes the synchronizing input and the data output for phasor measurements made by substation
computer systems. It also discusses the processes involved in computing phasors from sampled data. It is hoped that
this standard will be of considerable value to the developers and users of digital computer-based substation systems.
The members of the Synchrophasor Working Group that developed this standard are as follows:
A. G. Phadke, Chair
K. Martin, Vice Chair
M. G. Adamiak
G. Benmouyal
M. Bergovic
R. O. Burnett, Jr.
K. R. Carr
A. Cobb
S. H. Horowitz
G. R. Jensen
J. A. Kuster
G. L. Michel
R. J. Murphy
M. S. Sachdev
J. S. Thorp
The members of the Relaying Communications Subcommittee who reviewed this standard are as follows:
J. W. Hohn, Chair
M. Adamiak, Vice Chair
P. R. Drum
J. D. Huddleston, III
J. W. Ingleson
G. L. Michel
J. L. McElray
B. D. Nelson
S. L. Nilsson
A. G. Phadke
R. Ryan
E. T. Sage
J. T. Tengdin
E. A. Udren
J. A. Zipp

iv
The members of the PES/IEEE Power System Relaying Committee who reviewed and approved this standard are as
follows:
A. T. Giuliante, Chair
J. C. Appleyard, Vice Chair
R. W. Dempsey, Secretary
M. Adamiak
C. W. Barnett*
E. A. Baumgartner
B. L. Beckwith
G. Benmouyal
J. Berdy*
D.C. Blackburn
J. L. Blackburn*
J. Blose*
J. R. Boyle
B. Bozoki
W. R. Brownlee*
J. Burnworth
H. J. Calhoun
C. H. Castro
T. W. Cease
J. W. Chadwick, Jr.
G. K. Clough
D. H. Colwell*
S. P. Conrad
M. Conroy
C. J. Cook
A. N. Darlington
D. C. Dawson
R. E. Dietrich*
J. B. Dingler*
H. Disante*
C. L. Downs*
P. R. Drum
L. L. Dvorak
W. A. Elmore
J. T. Emery*
E. J. Emmerling*
M. K. Enns
J. Esztergalyos
G. Farley
W. E. Feero*
R. J. Fernandez*
C. W. Fromen
S. E. Grier
C. H. Griffin*
K. R. Gruesen*
E. N. Gulachenski
E. A. Guro
R. W. Haas
E. L. Harder*
R. E. Hart
C. F. Henville
R. W. Hirtler*
J. W. Hohn
S. H. Horowitz*
J. W. Ingleson
J. A. Jodice
E. W. Kalkstein
T. L. Kaschalk
M. Kezunovic
K. J. Khunkhun
J. L. Koepfinger*
W. C. Kotheimer
P. Kotos
L. E. Landoll*
J. R. Latham
J. R. Linders
G. J. Marieni*
W. J. Marsh, Jr.
C. A. Matthews*
G. E. May*
J. McConnell
J. McElray
M. Meisinger
W. Mello
R. J. Moran
J. J. Murphy*
T. J. Murray
K. K. Mustaphi
G. R. Nail
B. Nelson
T. Niessink*
S. L. Nilsson
R. W. Ohnesorge*
G. C. Parr
R. W. Pashley*
R. D. Pettigrew
A. G. Phadke
A. C. Pierce
J. M. Postforoosh
O. F. Pumphrey*
G. D. Rockefeller*
M. S. Sachdev
M. P. Sanders
E. T. Sage
M. Sanders
D. E. Sanford*
D. Smaha
H. S. Smith*
J. E. Stephens
W. M. Strang
F. Y. Tajaddodi
R. P. Taylor
J. T. Tingdin
J. S. Thorp
D. A. Tziouvaras
E. A. Udren
W. H. Van Zee*
V. Varneckas
D. R. Volzka*
F. Von Roeschlaub*
C. L. Wagner
J. W. Walton
W. P. Waudby
T. E. Wiedman
P. B. Winston
J. A. Zipp
S. E. Zocholl*
J. A. Zulaski
*Honorary Member (non-voting)

v
The following persons were on the balloting committee:
J. C. Appleyard
E. A. Baumgartner
B. L. Beckwith
J. R. Boyle
B. Bozoki
J. A. Bright
H. J. Calhoun
C. H. Castro
T. W. Cease
J. W. Chadwick, Jr.
G. K. Clough
S. P. Conrad
C. J. Cook
A. N. Darlington
D. C. Dawson
R. W. Dempsey
C. L. Downs
P. R. Drum
L. L. Dvorak
W. A. Elmore
J. Esztergalyos
G. Farley
C. W. Fromen
A. T. Giuliante
S. E. Grier
E. N. Gulachenski
E. A. Guro
R. E. Hart
C. F. Henville
R. W. Hirtler
J. W. Hohn
J. W. Ingleson
J. A. Jodice
E. W. Kalkstein
K. J. Khunkhun
W. C. Kotheimer
J. R. Latham
J. R. Linders
W. J. Marsh, Jr.
R. J. Moran
C. J. Mozinz
T. J. Murray
K. K. Mustaphi
G. R. Nail
S. L. Nilsson
G. C. Parr
R. D. Pettigrew
A. G. Phadke
A. C. Pierce
J. M. Postforoosh
M. S. Sachdev
E. T. Sage
D. Smaha
H. S. Smith
J. E. Stephens
W. M. Strang
F. Y. Tajaddodi
R. P. Taylor
J. S. Thorp
E. A. Udren
V. Varneckas
D. R. Volzka
C. L. Wagner
J. W. Walton
W. P. Waudby
T. E. Wiedman
J. A. Zipp
S. E. Zocholl
J. A. Zulaski
When the IEEE Standards Board approved this standard on 12 December 1995, it had the following membership:
E. G. "Al" Kiener, Chair
Donald C. Loughry, Vice Chair
Andrew G. Salem, Secretary
Gilles A. Baril
Clyde R. Camp
Joseph A. Cannatelli
Stephen L. Diamond
Harold E. Epstein
Donald C. Fleckenstein
Jay Forster*
Donald N. Heirman
Richard J. Holleman
Jim Isaak
Ben C. Johnson
Sonny Kasturi
Lorraine C. Kevra
Ivor N. Knight
Joseph L. Koepfinger*
D. N. "Jim" Logothetis
L. Bruce McClung
Marco W. Migliaro
Mary Lou Padgett
John W. Pope
Arthur K. Reilly
Gary S. Robinson
Ingo RŸsch
Chee Kiow Tan
Leonard L. Tripp
Howard L. Wolfman
*Member Emeritus
Also included are the following nonvoting IEEE Standards Board liaisons:
Satish K. Aggarwal
Steve Sharkey
Robert E. Hebner
Chester C. Taylor
Lisa S. Young IEEE Standards Project Editor

vi
CLAUSE PAGE
1. Overview.............................................................................................................................................................1
1.1 Scope.......................................................................................................................................................... 1
1.2 Purpose....................................................................................................................................................... 1
1.3 Need for this standard .................................................................................................................................1
2. References...........................................................................................................................................................1
3. Definitions...........................................................................................................................................................2
4. Sources of synchronizing signals........................................................................................................................3
4.1 Requirements for a synchronizing source.................................................................................................. 3
4.2 Broadcasts from satellites .......................................................................................................................... 3
4.3 Broadcasts from a central location............................................................................................................. 4
5. Synchronization ..................................................................................................................................................4
5.1 UTC synchronization ................................................................................................................................. 4
5.2 Phase-locked sampling pulses.................................................................................................................... 4
5.3 Time input.................................................................................................................................................. 6
5.4 Time-tag for data samples.......................................................................................................................... 7
5.5 Time-tag for phasors .................................................................................................................................. 7
6. Message format...................................................................................................................................................8
6.1 Transmitted messages ................................................................................................................................ 8
6.2 Data frame.................................................................................................................................................. 8
6.3 Header frame.............................................................................................................................................. 9
6.4 Configuration frame................................................................................................................................. 10
6.5 Received messages................................................................................................................................... 11
Annex A (Informative)..................................................................................................................................................12
Annex B (Informative)..................................................................................................................................................13
Annex C (Informative)..................................................................................................................................................14
Annex D (Informative)..................................................................................................................................................16
Annex E (Informative)..................................................................................................................................................21
Annex F (Informative) ..................................................................................................................................................25
Annex G (Informative) .............................................................................................................30

Copyright © 1998 IEEE All Rights Reserved 1
IEEE Standard for Synchrophasors for
Power Systems
1. Overview
1.1 Scope
This is a standard for synchronized phasor measurement systems in substations. It addresses synchronization of data
sampling, data-to-phasor conversions, and formats for timing input and phasor data output from a Phasor
Measurement Unit (PMU). It does not specify response time, accuracy, hardware, software, or a process for computing
phasors.
1.2 Purpose
This standard deÞnes synchronized phasor measurements in substations so that the measurement equipment can be
readily interfaced with associated systems. It speciÞes data formats and synchronization requirements to allow
correlating phasors from various sources and comparing them with similar data from different measurement systems.
1.3 Need for this standard
Recent developments in the Þeld of power system data acquisition provide users with the ability to record, time-tag,
transmit, and analyze power system phasor quantities in real time. Given the emphasis on future system development,
many different hardware and software approaches are being implemented to make the phasor data available. A
standard is needed for integrating measurement systems into substation environments, to specify data output formats,
and to assure that the measurement processes are producing comparable results. The synchrophasor standard will help
ensure maximum beneÞts from the phasor measurements and will allow interchange of data between a wide variety of
users of both real time and off-line phasor measurements.
2. References
This standard shall be used in conjunction with the following publications:

2 Copyright © 1998 IEEE All Rights Reserved
IEEE Std 1344-1995 IEEE STANDARD FOR SYNCHROPHASORS
IEEE Std C37.111-1991, , IEEE Standard Common Format for Transient Data Exchange (COMTRADE) for Power
Systems (ANSI).1
IRIG STANDARD 200-89, Telecommunications Group, Range Commanders Council, U.S. Army White Sands
Missile Range, NM.2
Network Time Protocol (Version 3) "SpeciÞcation, Implementation, and Analysis," D. L. Mills, DARPA Network
Working Group, RFC-1305, University of Delaware, March, 1992.3
3. Definitions
This clause contains key terms as they are used in this standard.
3.1 anti-aliasing: By the Nyquist Theorem, the maximum reproducible frequency is one-half the sampling rate.
Aliasing is caused when frequencies higher than one half of the sampling rate are present. This results in the higher
frequencies being "aliased" down to look like lower frequency components. Anti-aliasing is providing low pass
Þltering to block out frequencies higher than those that can be accurately reproduced by the given sampling rate.
3.2 Nyquist rate: The minimum rate that an analog signal must be sampled in order to be represented in digital form.
This rate is twice the frequency of that signal.
3.3 phase lock: The state of synchronization between two ac signals in which they remain at the same frequency and
with constant phase difference. This term is typically applied to a circuit that synchronizes a variable oscillator with an
independent signal.
3.4 phasor: A complex equivalent of a simple sine wave quantity such that the complex modulus is the sine wave
amplitude and the complex angle (in polar form) is the sine wave phase angle. See also: vector.
3.5 synchronism: The state where connected alternating-current systems, machines, or a combination operate at the
same frequency and where the phase angle displacement between voltages in them are constant, or vary about a steady
and stable average value.
3.6 synchronized phasor: A phasor calculated from data samples using a standard time signal as the reference for the
sampling process. In this case, the phasors from remote sites have a deÞned common phase relationship. Syn:
synchrophasor.
3.7 abbreviations and acronyms:
¾ BCD: Binary Coded Decimal
¾ CRC16: 16 b Cyclic Redundancy Check, calculated by the generating polynomial X16
+ X12
+ X5
+ 1
¾ DoD: Department of Defense
¾ DoT: Department of Transportation
¾ FRP: Federal Radionavigation Plan
¾ GOES: Geostationary Operational Environmental SatelliteÑconsists of two satellites in a geostationary
orbit.
¾ GPS: Global Positioning SystemÑa DoD navigation system that uses a constellation of 24 satellites
broadcasting a precision signal for location and time synchronization. Basic time synchronization accuracy is
± 0.2 ms.
¾ INMARSAT: International Maritime Satellite System
¾ IRIG-B: InterRange Instrumentation Group Time Code Format B
1IEEE publications are available from the Institute of Electrical and Electronics Engineers, 445 Hoes Lane, P.O. Box 1331, Piscataway NJ 08855-
1331, USA.
2This IRIG standard may be obtained from Secretariat, Range Commanders Council, U.S. Army White Sands Missile Range, NM, 88002, USA.
3This document is available from EE Department, University of Delaware, Newark, DL, 19716, USA. It is also available via the World Wide Web:
http://www.eeeis.udel.edu/~mills.bib.html.

FOR POWER SYSTEMS IEEE Std 1344-1995
¾ Loran C: LOng RANge navigation system that uses pulsed 100 kHz broadcasts from shore based transmitters
for location, particularly in maritime applications. It also provides precise frequency and limited time
synchronization.
¾ NTP: Network Time Protocol
¾ PMU: Phasor Measurement UnitÑin this standard a generic device that produces synchronized phasors from
voltage and/or current inputs and synchronizing signal.
¾ PPS: Pulse Per Second
¾ ROM: Read Only Memory
¾ SA: Selective Availability
¾ SBS: Straight Binary Seconds
¾ SCADA: Supervisory Control and Data Acquisition
¾ SMPCNT: Sample Count
¾ SOC: Second-Of-Century
¾ UTC: Universal Time Coordinated as distributed by the U.S. Naval Observatory through the GPS system
¾ WWV: Radio broadcast of time and selected weather information. They are provided by the U.S. National
Institute for Standards and Technology using several formats including WWV, WWVB, and WWVH.
4. Sources of synchronizing signals
4.1 Requirements for a synchronizing source
Phasors measured throughout an interconnected grid require a common timing reference provided by a synchronizing
source. The synchronizing source may be local or global, and the signal may be distributed by broadcast or direct
connection.
The signal provided by the synchronizing source shall be referenced to Coordinated Universal Time (UTC). It shall
provide enough time information to determine second-of-century in agreement with UTC. The synchronizing signal
shall have a basic repetition rate of 1 pulse per second (1 PPS) with a stability of at least 1E-07.
The synchronizing signal must be available without interruption at all measurement locations throughout the
interconnected grid. Reliability should exceed 99.87% (1 h of outage per month). The signal shall be accurate enough
to allow the phasor measurement equipment to maintain synchronism within 1 ms of UTC including both
synchronizing source and local receiving equipment error.
4.2 Broadcasts from satellites
Time signals broadcast from satellites have a wide area of coverage, are synchronized to a national standard, and are
little affected by atmospheric conditions. Sites with unobstructed views of satellites have little problem with reliable,
continuous signal reception. Users of satellite based systems do not have to provide primary time and time
dissemination systems, nor do they have to operate and maintain those systems. The only user cost is for the receiver
equipment, although fees may be levied in future satellite systems.
Transit, Geostationary Operational Environmental Satellite (GOES), and Global Positioning System (GPS) are the
three principal satellite systems available today for disseminating time. Once policy issues are resolved, INMARSAT
system satellites will provide precise time in a GPS-like navigation message. Of the three systems available now, only
GPS meets all requirements for this application.
The GPS is a U.S. Department of Defense (DoD) satellite based radio-navigation system. It consists of 24 satellites
arrayed to provide a minimum worldwide visibility of four satellites at all times. Each satellite transmits a timed
navigation signal, from which a receiver can decode time synchronized to within 0.2 ms of UTC. The inherent
availability, redundancy, reliability, and accuracy make it a system well suited for synchronized phasor measurement
systems.

The new INMARSAT system satellites will carry a GPS-like transponder. The signal will be similar to existing GPS
transmissions, so it can be used with slightly modiÞed GPS receivers. Since the signal will not be degraded by the
effects of SA, it should be a better source for time than standard GPS. INMARSAT satellites are geostationary, which
may make reception difÞcult at some locations. The INMARSAT system will provide an alternative to GPS.
4.3 Broadcasts from a central location
Synchronizing signals may also be broadcast from a central location. The viable alternatives include AM radio
broadcasts, microwave, and Þber-optic transmission systems. While AM broadcasts are the least expensive, they are
characterized by the lowest accuracy. Microwave and Þber-optics systems can achieve high accuracy but have high
installation costs.
The most popular and affordable AM systems, WWV, WWVB, and WWVH, rarely achieve accuracy better than 1 ms.
Interference poses problems with the reliability of reception. The Loran C impulse broadcast system is well supported
by a number of countries, including the U.S., through the end of the century and is a popular navigation system. Its
practical accuracy limit is of the order of 1 ms, but careful monitoring and well-designed human interfaces are required
for the best performance.
Fiber-optic and microwave systems can deliver synchronization reliably at 1 ms and better. This, however, requires a
specialized interface that bypasses the usual digital telecommunication interface. The problem with the usual digital
interface is its packet switching, which can cause a 70 ms uncertainty in signal transmission. Until digital systems are
developed with a standard, direct, synchronizing signal, this method of transmission will remain expensive.
5. Synchronization
5.1 UTC synchronization
Phasor measurements shall be synchronized to UTC time derived from the synchronizing source described in 4.1. A
PMU may receive the synchronizing signal directly or from an independent receiver. If an independent receiver is
used, it shall provide time to the PMU using the codes described in 5.3. The basic synchronizing signal is a 1 PPS pulse
train where the rising edge of the pulse marks the second change of UTC time. The PMU shall maintain a timing
accuracy of 1 ms between the synchronizing pulses. The 1 ms synchronization accuracy has been chosen because it is
available and meets all foreseeable data accuracy needs. This corresponds to an angular accuracy of 0.022 degrees for
a 60 Hz system, and 0.018 degree for a 50 Hz system.
If synchronized phasors are calculated from synchronized data samples, the data sampling shall be synchronized to
within 1 ms of UTC time. Details for PMUs that use Þxed frequency synchronized sampling are provided in 5.2, 5.4,
and 5.5.
If synchronized phasors are calculated from non-synchronized data samples, the calculating process shall determine
an equivalent UTC time and sample number as described for synchronized sampling techniques. The phasor thus
determined shall have the same magnitude and phase angle as a phasor bearing the same time and sample number
derived from synchronized samples using the same data window. Annex G includes a discussion of other techniques
for achieving synchronization of phasor measurements.
5.2 Phase-locked sampling pulses
In synchronized sampling systems, pulses that trigger data sampling shall be locked in phase with the 1 PPS signal.
The number of sample pulses per second shall be an integer and evenly spaced throughout the second. The sample
point error shall not exceed 1% of the sample period (0.01/sampling rate). The samples shall be numbered from 0 to

N-1, where N is the sample rate. Sample 0 is the Þrst sample in the new second and shall coincide with the rising edge
of the 1 PPS pulse within ± 100 ns. It is recommended that the sample rate be chosen from the list of Least Common
Multiple sample frequencies as deÞned in IEEE Std C37.111-1991 4
(COMTRADE), repeated here in tables 1 and 2
for convenience.
Table 1ÑSampling frequenciesÑcorresponding to fLCM = 384 ´ fbase
If time or sample pulses are provided by a device external to the PMU, a ßag shall be provided to indicate a loss of time
sync or phase lock. Upon loss of sync or lock, the PMU shall set a ßag in the status word indicating that the resultant
data was computed with "Un-Locked" data. If sampling is being resynchronized to the 1 PPS signal after a period of
being unsynchronized, the sample pulses shall be slewed into alignment at a maximum rate of 70 ms/s. The 70 ms/s will
generate a 0.004 Hz computed frequency error but allows synchronization in 2 h from a worst case 0.5 s offset.
4Information on references can be found in clause 2..
Samples/Cycle
Sample
60 Hz
Frequency
50 Hz
4 240 200
6 360 300
8 480 400
12 720 600
16 960 800
24 1440 1200
32 1920 1600
48 2880 2400
64 3840 3200
96 5760 4800
128 7680 6400
192 11 520 9600
384 23 040 19 200

Table 2ÑSampling frequenciesÑcorresponding to fLCM = 3200 ´ fbase
5.3 Time input
UTC time and synchronization may be provided to the PMU using IRIG-B and 1 PPS or using the high-precision
format time code. The IRIG-B format is commonly available and hence is the most readily implemented. The control
bits are used to enhance standard IRIG-B with year, time zone, and other special ßags. The high-precision format is an
encoding technique compatible with Þber-optic systems that allows high-precision synchronization. The information
format is the same as in IRIG-B. The formats are described in detail in annex F.
¾ IRIG-B format. IRIG-B is fully described in IRIG STANDARD 200-89. Time is provided once per second in
seconds through day of year in a binary coded decimal (BCD) format. The BCD is amplitude modulated with
a 1 kHz carrier signal.
¾ 1 PPS. A one pulse per second positive pulse with the rising edge on time with the second change provides
precise time synchronization. The pulse shall be 5 V with a 50 ns or less rise time and 5 ms to 0.5 s duration.
¾ High precision time code format. The high precision format is the IRIG-B data format transmitted with
modiÞed Manchester coding. The clocking rate is 1 kHz with a rising data edge indicating a binary 1 and a
falling data edge indicating a binary 0. Data edges are in phase with UTC. The high speed digital format
eliminates the need for separate 1 PPS synchronization. This coding is a binary, zero-mean format that is
easily transmitted over and recovered from cable or Þber-optic systems. Modulated and unmodulated IRIG-
B can be regenerated from the code with simple circuits. It is recommended that PMU implementation
include provision for decoding both time code formats.
Samples/cycle
Sample
60 Hz
Frequency
50 Hz
4 240 200
8 480 400
10 600 500
16 960 800
20 1200 1000
32 1920 1600
40 2400 2000
50 3000 2500
64 3840 3200
80 4800 4000
100 6000 5000
128 7680 6400
160 9600 8000
200 12 000 10 000
320 19 200 16 000
400 24 000 20 000
640 38 400 32 000
800 48 000 40 000
1600 96 000 80 000
3200 192 000 160 000

5.4 Time-tag for data samples
Data shall be tagged with the UTC time during the second the sample was taken. The time-tag shall be a 4 B binary
second-of-century (SOC) number derived in accordance with the network time protocol (NTP). This SOC number is
the UTC time in seconds calculated from midnight of January 1, 1900. Leap seconds are added to or deleted from the
scale as necessary to keep NTP synchronized with UTC. (This means NTP can always be determined by multiplying
the number of seconds per day by the days since 1/1/1900. A leap second insertion will result in 2 s having the same
SOC number.)
5.5 Time-tag for phasors
The time-tag for phasor data shall be composed of two words. The Þrst is a 4 B word giving the SOC time in the NTP
time scale. This is the same as the time-tag for data samples. The second word shall be 2 B SMPCNT giving the sample
number of the last sample used in the phasor calculation. This is independent of the length of the data window used
in the phasor calculation, as discussed in annex C. The sample taken on the 1 s mark is sample number 0. Using the
convention that the signal v(t) = V cos (wot + j) corresponds to the phasor V ejj, the phasor computed from a
window beginning at a positive voltage maximum is real while one computed from a window beginning at a positive
going voltage zero is negative imaginary as shown in Þgure 1.
Figure 1ÑConvention for phasor representation
If the 1 PPS signal occurs at time to, the measured phasor corresponding to a sinusoidal signal v(t) = V cos (wot
+ j) with a frequency wo is V ej
(woto + j). For steady-state signals at off-nominal frequency w1, the measured phasor
with time-tag corresponding to the 1 PPS instant to is V ej
(w1to + j). The phasor phase angle shall be corrected for any
phase angle delays occurring inside the PMU. The behavior of the measured phasor between the 1 PPS points and the
response to non-steady-state (transient) conditions will vary with the algorithms used, and is discussed in annex C.
2
2

6. Message format
6.1 Transmitted messages
The PMU shall transmit three types of information organized as frames. The Þrst 4 B of each frame are a SOC time
mark. The 4 B SOC provides a distinct combination useful for frame synchronization. It is followed by the SMPCNT
word that identiÞes the frame type and gives the sample number or the frame count. The three most signiÞcant bits of
the SMPCNT word determine the frame type as deÞned in table 3. Every frame shall be terminated by a CRC16 to
validate the message. This CRC16 uses the generating polynomial X16
+ X12
+ X5
+ 1.
Only data, conÞguration, and header frames are deÞned in this standard. Other types may be designated in the future
using undesignated combinations in the SMPCNT word. In normal operation the PMU only sends data frames.
Table 3ÑSMPCNT word definition (message type is defined by the three most significant bits)
The header and conÞguration parameters shall be stored in the PMU. These formats generally follow the speciÞcations
of the ConÞguration Þle (*.CFG) and the Header Þle (*.HDR) of IEEE Std C37.111-1991 (COMTRADE).
Real-time data consisting of phasers, frequency estimates, and digital status shall be transmitted in data frames. Each
frame is a data scan corresponding to the time-tag and sample number. The data shall be in binary format with no
delimiters. Information for parsing and converting the information to engineering units shall be stored in the
conÞguration Þle. The central data receiving system shall acquire the conÞguration Þle and store the information for
operation, The header and conÞguration Þles may be transmitted during real time data transmissions, so the
information will be interleaved with data frames. Since these Þles may be too large to send in one block between data
frames, they can be broken into multiple frames for transmission. The minimum frame length for header or
conÞguration frames is 9 B when a single byte is to be transferred. Examples of data, header, and conÞguration Þles
are in annex D.
6.2 Data frame
A data frame shall contain measured data and shall be identiÞed by having the three most signiÞcant bits of the third
word (5th byte) equal to zero. The real-time phasor data frame shall consist of binary data arranged as shown in Þgure
2. SOC shall be the 4 B second-of-century time stamp as deÞned in 5.4. The sample count word, SMPCNT, shall be
as deÞned in table 3 and the status word, STAT, shall be as deÞned in table 4. Data integrity shall be conÞrmed by using
SMPCNT, SOC, and CRC16.
Bits 15Ð14Ð13 Frame type Definitions
000 data frame Bits 0Ð12: sample count, 0Ð8191
001 header frame
Bit 12: first frame flag (0: first frame of header file, 1: continuing
frame)
Bit 11: last frame flag (0: not last frame, 1: last frame of file)
Bits 0Ð10: frame count for file, 0Ð2047 (first frame is count 0)
010 configuration frame
Bit 12: first frame flag (0: first frame of configuration file, 1:
continuing frame)
Bit 11: last frame flag (0: not last frame, 1: last frame of file)
Bits 0Ð10: frame count for file, 0Ð2047 (first frame is count 0)
011Ð100 reserved for expansion Bits 0Ð12: to be assigned
101Ð111 user defined Bits 0Ð12: user defined

Each phasor shall consist of 4 B: 2 B for the real part, and 2 B for the imaginary part, respectively; or, 2 B for
magnitude, and 2 B for angle, respectively. Frequency (FREQ) and rate-of-change of frequency (DFREQ) shall be 2
B each. Digital channel data (DIG) shall be contained in 2 B words, one word for each 16 b channel. The cyclic
redundancy check is the 2 B word CRC16.
Figure 2ÑData frame for phasor data
Table 4ÑStatus word STAT definition
If the phasor is in real and imaginary format, each shall be a two's complement, 16 b binary number, with the scale
deÞned by the conÞguration Þle. If the phasor is in magnitude and angle format, the magnitude shall be a 16 b binary
number in the range 0 to 215
-1; and, the angle shall be in radians ´ 10-4
in the range -p to +p. The frequency (FREQ)
is a 16 b binary number in Hz ´ 10-3
deviation from system nominal. The system nominal frequency shall be given in
the conÞguration Þle. The range shall be -32.767 to +32.767 Hz. The rate-of-change of frequency (DFREQ) shall be
the rate of change in Hz per second ´ 10-2
. The 16 b number thus deÞnes a range of -327.67 to +327.67 Hz/s. The
digital channel data shall reßect the state of the digital inputs (binary 0 or binary 1). The conÞguration Þle deÞnes
whether 0 or 1 is the normal state. Digital inputs shall be assigned starting at the least signiÞcant bit (b 0) of digital
channel word. Zeroes shall be inserted in unused bits.
6.3 Header frame
This shall be an ASCII Þle, and shall contain information about the PMU, the data source, scaling, transducers,
algorithms, analog Þlters used, etc. The frame has the same synchronizing SOC timetag, SMPCNT, STAT, and CRC16,
as the other frames, but the header Þle data has no Þxed format (see Þgure 3).
STAT bits Definitions
bit 15 synchronization status (0: valid, 1: invalid)
bit 14 Data integrity (0: valid data, 1: invalid data)
bits 13Ð11
Trigger status:
111: Frequency trigger
110: df/dt trigger
101: Angle trigger
100: Overcurrent trigger
011: Undervoltage trigger
010: Rate trigger
000Ð001: Unused, user defined
bits 10Ð0
Number of bytes in frame including SOC,
SMPCNT, STAT, and CRC16, 10Ð2047

Figure 3ÑHeader frame organization
6.4 Configuration frame
This shall be a machine-readable binary Þle containing information about processing parameters associated with the
real-time data frame shown in Þgure 2. The conÞguration Þle shall have the following information:
a) Station name and identiÞcation
b) Number of phasors
c) Number of digital channels
d) Channel and phasor names, units, and conversion factors
e) Nominal line frequency
f) Transmission period in number of fundamental frequency cycles
There shall be two types of conÞguration Þles: a system conÞguration Þle CFG-1, and a data conÞguration Þle CFG-
2. The CFG-1 Þle shall indicate all possible inputs. The CFG-2 shall indicate the phasors actually being sent since only
a subset of available phasors may be transmitted by the PMU in the data frame. All Þelds shall be Þxed length and no
delimiters shall be used. SOC, SMPCNT, STATUS, and CRC16 are as previously described. Other data types are as
follows:
¾ STN: Station Name; 16 B in ASCII format.
¾ IDCODE: 8 B, PMU hardware identiÞcation, normally coded into the ROM.
¾ PHNMR: Number of phasors; 2 B integer word (0Ñ32767).
¾ DGNMR: Number of digital channels; 2 B integer word (0Ñ32 767).
¾ CHNAM: Phasor and channel names; 16 B for each phasor and each digital channel in ASCII format in the
same order as they are transmitted.
¾ PHUNIT: Conversion factor for analog channels; 4 B for each phasor. The Þrst byte designates voltage or
current (00 = voltage; 01 = current). The last 3 B form a 24 b word in 10-5
V or amperes per bit.
¾ DIGUNIT: Conversion factor for digital channels; 2 B word for each digital channel. Bit 4 indicates the
normal state in the digital word (0 or 1); bit 0 (least signiÞcant bit) designates the input normal state (0 =
open, 1 = closed).
¾ FNOM: Nominal line frequency code and ßag in two byte word. Bit 8 set (=1) indicates FREQ is omitted
from the data frame and bit 9 set indicates DFREQ is omitted. Bit 0 (least signiÞcant bit) set indicates the
fundamental line frequency is 50 Hz, and cleared indicates 60 Hz. The remaining bits can be used to
designate other codes and frequencies.
¾ PERIOD: Period of phasor data transmissions; 2 B integer word (0Ñ32 767) of the number of fundamental
frequency cycles ´ 100 between consecutive data transmissions (200 means a data frame every 2 cycles).

Figure 4ÑConfiguration frame organization
The conÞguration Þle may be sent in several frames. The overall transmission has the organization shown in Þgure 4
but the conÞguration Þle data from STN through PERIOD may be broken into multiple frames. Each frame shall start
with SOC, SMPCNT, and STAT, and end with CRC16.
Figure 5ÑReceived command frame definition
6.5 Received messages
The PMU shall receive commands from a control system and take appropriate actions. The received message frame is
shown in Þgure 5. IDCODE shall be an 8 B identiÞcation code that shall be matched with a pre-stored code in the PMU
for the command to be accepted by the PMU. The IDCODE shall be a unique code as determined by the user of the
PMU. CMD shall be a 2 B command code as deÞned in table 5. SOC and CRC16 are the timestamp and error check as
previously described.
Table 5ÑCommands received by the PMU
Command word bits Definition
Bits 15Ð4 Reserved for future use
Bits 3Ð2Ð1Ð0:
0001 Turn off real-time data
0010 Turn on real-time data
0011 Send HDR file
0100 Send CFGÐ1 file
0101 Send CFGÐ2 file
1000 Receive reference phasor in data frame format

Annex A
(Informative)
Bibliography
[B1] Brigham, E. O. The Fast Fourier Transform, Prentice Hall, Inc., New York, 1974, p. 105.
[B2] "1992 Federal Radionavigation Plan," DoT-VNTSC-RSPA-92-2/DoD-4650.5, Stock No. 008-047-00402-8,
U.S. Government Printing Office. Washington, D.C. 20402.

Annex B
(Informative)
Measurement accuracy and precision of synchronization
B.1 Measurement accuracy
Phasor accuracy is limited by the data sampling as follows. For a minimum error requirement Emin and a full-scale
rating Fs, the A/D converter needs the following:
( B-1)
The factor 2 scales the formula from rms to bipolar peak values, which is how A/D converters must be speciÞed.
For example, if the full-scale input Fs is 1.5 times the nominal input and a 1% resolution is required (Emin=0.01), the
minimum A/D resolution required is 9 b. Practical choices for number of bits are dictated by component availability
and are typically 8, 12, 16, or 18 b.
This computation will be reasonably true for measuring bus voltage phasors since the magnitude typically varies little.
Line currents are much more dynamic, with 10% of full load considered light load, short term overload ratings 2XÐ3X
full load, and fault ratings 40X full load. Since phasor measurements are not very meaningful during faults, a
reasonable scaling may set the full-scale 3X nominal and require a 1% accuracy at light load. This would add a factor
of 10 to the full-scale accuracy and result in a 13 b requirement. The user needs to balance dynamic range requirements
with system costs.
Other factors may have as great an effect on accuracy as data sampling. Anti-aliasing Þlters, harmonics, and phase
imbalance can signiÞcantly effect the measurement. Without anti-aliasing Þlters, harmonics are aliased directly into
the measurements. Power ßow computed from positive-sequence phasors can be inaccurate if there are large phase
imbalances. Anti-aliasing Þlters can introduce signiÞcant errors. For example, a 2 pole R-C Þlter with a cutoff
frequency of 300 Hz will have a phase shift of about 22 degrees and a 2% magnitude rolloff at 60 Hz.
B.2 Required precision of synchronization
Measurement accuracy requirements are determined by the application. A tenth of a degree at the power system
frequency is sufÞcient accuracy for state estimation, stability monitoring, control, and relaying. This requires a total
synchronization error less than 5 ms for a 60 Hz system. System hardware synchronization, sampling, and the data
window all must be included in the synchronization uncertainty. Clauses 4 and 5 and annex G discuss synchronization
further.
Minimum resolution ln 2 2 Fs( ) Emin¤[ ] ln(2)¤=
2

Annex C
(Informative)
Time tagging phasors
C.1 Transient response
As a consequence of the requirements of 5.1, two PMUs of different brands should be expected to yield the same
phasor measurement in steady-state. Steady-state is a condition deÞned here as one where no change is taking place in
magnitude, phase angle, or frequency. Also, identical PMUs (deÞned as having identical hardware and algorithms)
should yield the same phasor measurement under all conditions. However, two PMUs of different brands (with
different algorithms and/or different analog circuitry) can be expected to yield different results for the same phasor
measurement in transient state (deÞned here as the time a change in magnitude, phase angle, or frequency takes place).
Since this document does not point to a particular algorithm as a reference, an acceptable tolerance is not deÞned here
and the amount of discrepancy to be expected from two PMUs can be evaluated only by inspection and analysis of the
individual units. In particular the following should be analyzed:
a) The type of sampling algorithm used and the value of the nominal sampling frequency. With conventional
FIR type Þltering (Discrete Fourier Transform, Least Squares Fitting, etc.), a Þxed sampling rate will yield a
phasor Þxed in the complex plane only at rated frequency. This phasor normally undergoes an angular
rotation, with the speed proportional to the frequency deviation when the frequency is different from rated.
On the contrary, with an adaptive sampling rate where the number of samples per cycle is Þxed whatever the
incoming frequency, the measured phasor will be normally Þxed in the complex plane, even if the frequency
is different from rated frequency.
b) The data window length.
c) The type of Þlter used to get the phasor measurement. Particularly, the frequency response of the algorithm
and its rejection of harmonics and off-nominal frequencies and its time response to a voltage or current step-
function, together with the frequency response, should be investigated.
C.2 Alternate time-tags
Time-tagging the phasor measurement is an issue that is largely a matter of convention. Since a single measurement is
the result of processing data in a time window, the time-tag may technically be associated with any of the samples that
belong to the time window. The three most logical alternatives are as follows:
a) Time-tag at the Þrst sample of the window
b) Time-tag in the middle of the window (at sample number [N/2], or [N/2]+1, where N is the number of
samples in the window)
c) Time-tag at the last sample of the window
In practice, there is no reason why any of the alternatives would be given preference over any other. A few remarks
may be helpful. Figure C-1 shows the phase response of a phasor measurement system to a waveform x(t) = Xmax cos
{wot+0.1 h(t-0.05)}, where h(t) is a step-function that occurs at time t=0.
The waveform therefore represents a step change of phase angle of 0.l rad at t=0.05. The response was calculated using
a time window of 12 samples. The curves labeled "start of window," "center of window," and "end of window"
represent the response when time-tag is associated with the Þrst, sixth, and last sample of the window respectively.
Obviously, the only difference between the methods occurs in the interval equal to two time windows centered around
the moment of disturbance; phase angles before and after are calculated correctly.

Figure C-1ÑIllustration of transient response of phasors
When monitoring is the only objective of phasor measurement, the choice of time-tag is irrelevant, as long as
consistency is maintained. If measurement may be used for real time control, the following should be taken into
account in interpreting the results:
¾ Tracking: From the point of view of "the best Þt" between the original and measured waveforms, the time-tag
in the center of the time window represents the best choice.
¾ Causality: When time-tag is associated with anything but the last sample of the window, the calculated
response appears to begin before the actual disturbance occurs. Only when the tag is on the last sample does
the response appear to begin after the occurrence of the disturbance.
¾ Data Availability: The phasors are available for further processing (real time control, etc.) only after the time
window of measurement. Therefore, choosing time-tags before the end of the time window amounts to
representing the phasor measurement system as a system with a delay for real time control purposes. This, of
course, should be of no major concern given that the measurement time windows are normally very short
compared to the dynamics of the system that may qualify for control.
¾ Initiation: Finally, the time-tag corresponding to the initiation of measurement corresponds to the Þrst sample
of the time window. This may be considered as a logical choice, with the proviso that when the objectives of
the measurement encompass the above properties, appropriate compensatory action is taken.

Annex D
(Informative)
Supporting information for synchronized phasor data
Real time phasor data is transmitted according to the format shown in Þgure 2. Interpretation of the real time data
involves various parameters for the data to be processed. These parameters are saved into two Þles in the phasor
measurement systems. These Þles are "ConÞguration Þle" (*.CFG) and a "Header Þle" (*.HDR) that generally follow
the speciÞcations of IEEE Std C37.111-1991 (COMTRADE). The information may be accessed remotely, typically
only at unit startup. Each Þle may be broken into several frames whose length is short enough to interleave with real
time data.
In general, the selected data communication structure is designed to support the following requirements:
a) Only measured and computed data should be transmitted in real time. Informational data should be
transmitted only on request.
b) All data must be traceable to an absolute time reference. The data transmitted should be in the most compact
form possible to Þt the available channel bandwidth. However, consideration must be given to optimization of
host computer hardware and software.
c) A wide range of basic sampling rates as well as data transmission rates must be supported.
d) The format should support bi-directional real time control functions in a full-duplex communication mode.
e) A mechanism to transmit bi-directional status information should be provided.
f) Data integrity checks and an option for re-transmission should be provided.
g) The amount and type of data transmitted over a sample interval must be user deÞnable to adjust to the wide
range of data requirements.
D.1 Header file
This is an ASCII Þle, and should contain information about the PMU, the data source, scaling, transducers, algorithms,
analog Þlters used, etc. This Þle has no Þxed format.
The following is an example of a header Þle:
DATA ORIGINATING FROM SUBSTATION XXX, COMPANY
YYY. CONTACT PERSON XYXY. PHONE NUMBER:
(XXX) XXXÐXXXX. GPS SYNCHRONIZED PMU.
ANALOG FILTERS HAVE A PHASE LAG OF
10° ± 0.03° AT 60 Hz. NOMINAL FREQUENCY 60 Hz.
VOLTAGE CLASS 500 kV. XXC400 AND XXC800 TYPE CT.
CVTS MANUFACTURED BY YYY.
5 ANALOG CHANNELS AND 4 DIGITAL CHANNELS.

D.2 Configuration file
This is a machine-readable binary Þle, and contains the parameters associated with the data frame shown in Þgure 2.
The Þle is structured as described in 6.4. Figure 4 illustrates the frame structure, and shows the additional frame header
(SOC, SMPCNT, STAT) and trailer (CRC16) words. When the Þle is transmitted as several frames, structure within
each frame and the order of frames will preserve the original Þle structure.
The following is an example of a conÞguration Þle:
58 58 58 58 58 58 20 35 30 30 20 4B 56 20 20 20 XXXXXXÐ500ÐkVÑ
F2 45 AA 9D 2C 03 A9 4C binary id
00 05 five phasors
00 04 four digital channels
20 42 55 53 20 56 4F 4C 54 41 47 45 20 20 20 20 BUS VOLTAGE
20 4C 49 4E 45 31 20 43 55 52 52 45 4E 54 20 20 LINE1 CURRENT
20 4C 49 4E 45 32 30 43 55 52 52 45 4E 54 20 20 LINE2 CURRENT
20 4C 49 4E 45 33 20 43 55 52 52 45 4E 54 20 20 LINE3 CURRENT
20 4C 49 4E 45 34 20 43 55 52 52 45 4E 54 20 20 LINE4 CURRENT
42 52 45 41 4B 45 52 31 20 20 20 20 20 20 20 20 BREAKER1
42 52 45 41 4B 45 52 32 20 20 20 20 20 20 20 20 BREAKER2
42 52 45 41 4B 45 52 33 20 20 20 20 20 20 20 20 BREAKER3
42 52 45 41 4B 45 52 34 20 20 20 20 20 20 20 20 BREAKER4
00 17 48 A5 voltage; 32 767 = 500 kV
01 00 0B EB current; 32 767 = 1000A
01 00 77 36 current; 32 767 = 10000 A
01 00 05 F5 current; 32 767 = 500A
01 00 05 F5 current; 32 767 = 500A
00 11 1 is normal; normally closed
00 10 1 is normal; normally open
00 11 1 is normal; normally closed
00 01 0 is normal, normally closed
00 00 FREQ & DFREQ present, 60 Hz nominal frequency
00C8 transmission each two cycles
XXXX CRC16

D.3 Data file
The following example shows four data frames that might be found in a data Þle. They illustrate a system sending two
cycle data, or, one sample every two cycles of the power system frequency with Þve phasors and four digital words.
The following is an example of a data Þle:
Hexadecimal Decimal Quantity
B0 15 58 00 2 954 188 800 SOC
02 28 552 Sample #
30 16 12 310 Status
04 9C 1180 Phasor #1 real
DA 26 -9690 Phasor #1 imaginary
E7 E6 -6170 Phasor #2 real
E7 F8 -6152 Phasor #2 imaginary
F2 75 -3467 Phasor #3 real
27 E4 10 212 Phasor #3 imaginary
EB 37 -5321 Phasor #4 real
12 38 4664 Phasor #4 imaginary
F8 CD -1843 Phasor #5 real
28 67 10 343 Phasor #5 imaginary
00 00 0 Frequency
00 00 0 df/dt
55 55 21 845 dig channel #1
AA AA 43 690 dig channel #2
55 55 21 845 dig channel #3
XX XX Ñ CRC16
B0 15 58 00 2 954 188 800 SOC
02 40 576 Sample #
30 16 12 310 Status
E7 EB -6165 Phasor #2 real
27 E1 10 209 Phasor #3 imaginary

F8 D2 -1838 Phasor #5 real
00 00 0 Frequency
00 00 0 df/dt
55 55 21 845 dig channel #1
55 55 21 845 dig channel #3
XX XX Ñ CRC16
B0 15 58 00 2 954 188 800 SOC
02 58 600 Sample #
30 16 12 310 Status
E7 EF -6161 Phasor #2 real
27 DE 10 206 Phasor #3 imaginary
EB 2D -5331 Phasor #4 real
F8 C3 -1853 Phasor #5 real
00 00 0 Frequency
00 00 0 df/dt
55 55 21 845 dig channel #1
55 55 21 845 dig channel #3
XX XX Ñ CRC16
B0 15 58 00 2 954 188 800 SOC
02 70 624 Sample #
30 16 12 310 Status

DA 1D -9699 Phasor #1 imaginary
E7 F4 -6156 Phasor #2 real
E7 ED -6163 Phasor #2 imaginary
F2 7C -3460 Phasor #3 real
27 DC 10 204 Phasor #3 imaginary
F8 C8 -1848 Phasor #5 real
28 5F 10 335 Phasor #5 imaginary
00 00 0 Frequency
00 00 0 df/dt
55 55 21 845 dig channel #1
55 55 21 845 dig channel #3
XX XX Ñ CRC16

Annex E
(Informative)
Sources of synchronization
E.1 Synchronizing source requirements
A synchronizing source must have sufÞcient availability, reliability, and accuracy to meet power system requirements.
Protection and control systems in the power industry operate continuously without interruption. A synchronization
source should have continuous uninterrupted availability. The synchronizing signal source must be available to all sites
where the data is to be compared. This could range from two sites at either end of a single power line, to an entire
utility service area. To be as inclusive as possible, the same signal should be available to the entire region where data
might be reasonably compared. The source should be an order of magnitude more reliable than the system requirement
since it is the single most crucial element in the synchronizing system.
The synchronizing signal must have sufÞcient accuracy to assure that the samples are taken at the correct time. Its
basic accuracy must be greater than the sampling accuracy requirement. The signal must be repeated often enough to
keep the sampling clock within the sampling accuracy requirement. The maximum sampling timing error Et in second
is as follows:
Et = Es + RiAc
where
Es is the synchronizing signal maximum error in seconds
Ri is the repetition interval in seconds and
Ac is the sampling clock accuracy Þgure (a ratio of seconds/seconds)
A GPS based 1 PPS synchronizing signal with a maximum error of 1 ms coupled with a 10E-6 sampling clock yields
a combined error of 2 ms.
E.2 Broadcasts from satellites
Satellite broadcast timing has signiÞcant advantages over other timing systems. Satellites have a wide area of
coverage. The signal is little affected by atmospheric conditions or seasonal variations. The time signal is continuously
referenced to a national standard. Cost is low because the primary reference and time dissemination system are
provided by the satellite system sponsor.
The principal problem with satellite broadcasts has been control. All satellite broadcast systems have been put up for
purposes other than time dissemination. During crises the primary purposes take priority, and timing function users
have occasionally lost access. Satellite systems are expensive to put up and maintain, so the time function user is also
at the mercy of funding provided for the primary satellite system function.
For most users, the advantages still make satellite systems worth the risk. As time becomes more critical in
applications, users will need to invest more in backup or alternative systems. They will also demand more reliability
from government provided services. The need for industry support is being recognized, and timing users are receiving
a higher priority. It is expected that satellite based time will eventually be provided with the reliability sought by the
power industry, though at that point funding may be by subscription fees.
Transit, GOES, and GPS are the three principal satellites systems available in 1996 for disseminating time. The
INMARSAT system should also provide precise time in GPS-like format once policy issues are resolved.

Transit is the oldest of these services and is due to be phased out in 1997. It is a Navy satellite navigation system whose
signal can be decoded for time synchronization. It has a reliable 10 ms accuracy but poor availability due to few
satellites. With these limitations, it is not a viable alternative for phasor time synchronization.
The GOES system's primary mission is weather monitoring, particularly for hurricane activity in the Western
Hemisphere. The system consists of two geostationary satellites situated to provide coverage across the entire United
States. Due to satellite failures, the system has occasionally been reduced to one operating unit, leaving some areas
without usable time reception. Small dish antennas are frequently used and must be readjusted if the satellites are
moved. The radio link at 468 MHz suffers some interference problems with land based mobile communications and
outages due to solar eclipses in the spring and fall. It provides a time synchronization referenced to UTC with a base
accuracy of 25 ms, although a more realistic operating accuracy is 100 ms. Overall, it provides a synchronizing signal
that is marginally acceptable for phasor measurements.
The Global Positioning System (GPS) is a United States Department of Defense (DoD) satellite based radionavigation
system. Completed in 1994, it consists of 24 satellites that circle the earth twice a day in six orbital planes. Range and
time is broadcast on three L-band frequencies. Time can be derived from the C/A code transmitted at 1575 MHz.
Each satellite modulates the 1575 MHz signal with a unique pseudo-random (PN) sequence that spreads the signal
power over a wide bandwidth. The spread spectrum makes the signal resistant to interference. Code division
multiplexing allows many satellites to share the same frequency. The signal can be received by a simple
omnidirectional antenna. The receiver applies the same PN code to the received signal to de-spread it and acquire
precise synchronization with the satellite signal. The distance from each satellite can be calculated by comparing
receiver time with the satellite time; that is, distance is the time the signal takes to travel from the satellite to the
receiver traveling at the speed of light. Receiver position is determined by triangulating the distance from three
satellites using a fourth satellite signal as a time reference. If time or any receiver coordinates are known, fewer
satellites are required. A receiver in Þxed position needs only one satellite to determine time. Additional information
is modulated on the satellite signal including time of day, position, and other system parameters.
GPS is steered by a ground-based Cesium clock ensemble, which itself is referenced to Coordinated Universal Time
(UTC), the world standard. Each satellite provides a correction to UTC time that the receiver automatically applies to
the outputs. With this continuous adjustment, timing accuracy is limited only by short term signal reception whose
basic accuracy is 0.2 ms. That baseline accuracy can be improved by advanced decoding and processing techniques.
For general applications, a 0.5 ms accuracy is a safer Þgure to depend on. The short term frequency accuracy of the
received signal is on the order of 1 ´ 10E-11.
A number of considerations make GPS a system that is very available, reliable, and accurate. DoD designed to provide
navigation throughout the world without fail in the worst conceivable scenarios, including all weather, system
degradation, and interference conditions. With a minimum of four satellite coverage at all times, even sites with
restricted sky view are unlikely to lose signal reception. The spread spectrum decoding technique has a high inherent
rejection of other signals and noise. Interference from other systems at 1.5 GHz is minimal. Atmospheric conditions
and power system noise have been found to have little inßuence on signal reception. The antennas do not need to be
pointed. The receivers are mostly digital and require little tuning or adjustment. Large receiver circuit portions can be
built on VLSI chips, reducing the overall chip count for higher reliability.
The only requirement for a phasor system synchronizing signal that GPS does not clearly meet is the access issue.
There is still some uncertainty that civilian use will be supported at a level acceptable to the power industry. Congress
directed the DoD and the Department of Transportation (DoT) in 1978 to develop a plan for joint use of GPS, rather
than spending large sums on both military and civilian systems. GPS was to be further examined as a replacement for
older civilian and military systems for additional cost savings. However, GPS has proven successful beyond
expectations and the DoD has concerns that a system open to general civilian use may be used by a military adversary.
To address these concerns, they have degraded the accuracy of the C/A signal with a process called Selective
Availability (SA), and have encrypted the precise signal. Neither of these have a signiÞcant effect on timing required
by power systems, but the possibility of further restrictions is an issue of concern.

These issues are being resolved as the DoD and DoT jointly develop the Federal Radionavigation Plan (FRP) (see
[B2]). It basically details what radio navigation aids will be provided by the U.S. government and gives a timetable for
their availability. The plan commits GPS to be made available to civilian users at all times except in a national
emergency. It limits the inaccuracy introduced by SA so that 92% of the time it will not exceed 0.5 ms and 99.9% of
the time it will not exceed 1.1 ms. It also commits the U.S. to provide the signal worldwide without fee for a minimum
of 10 years. The FRP is updated every two years, usually in the direction of more civilian access. The large and
growing base of users in many diverse applications exerts enough inßuence that withdrawal of civilian access is
unlikely.
The International Maritime Satellite System (INMARSAT) is a series of satellites in geostationary orbits put up for
maritime communications and navigation. The new satellites being built for launch in 1995 include a transponder
capable of transmitting a GPS-like signal. No SA will be applied, allowing time reception at full precision. Since it is
a commercial enterprise, there is little risk that it will be shut down for military purposes. The only real drawback is
that geostationary signals are very low on the horizon in the northern latitudes, making reception difÞcult at some sites.
The GPS-like signal should allow reception with existing GPS equipment with only minor modiÞcations.
Overall, the use of GPS for phasor systems synchronization looks very promising. All requirements are well met, the
system is well funded for the foreseeable future, and areas of doubt are becoming resolved. GPS is recommended as
the synchronizing source of choice.
E.3 Broadcasts from a central location
A variety of alternatives exist for synchronization of remote measurement units using a signal from a central location
that has been sent to all the measurement units at the same time. Among the available systems the following groups
exist:
¾ AM radio broadcasts
¾ Microwave transmission systems
¾ Fiber-optic transmission systems
AM radio broadcasts are the least expensive, but due to a variety of problems with their implementation, their accuracy
is very limited (to a few milliseconds). Both microwave and Þber-optic circuits require substantial hardware and
maintenance investment while reaching the accuracies of 1 ms or better. It should also be noted that leased wire line
circuits could also be used in place of microwave or Þber systems. However, the user generally has no control over
long distance circuits. Routing changes, costs, and reliability combine to make this method undesirable.
The AM radio broadcasts are among the most popular and affordable ways to disseminate time information. The
transmitting medium is the atmosphere via electromagnetic waves, hence no hardware investment is needed and the
receivers may be placed anywhere. The transmitters are land-based. The most commonly used radio stations in the
U.S. are WWV, WWVB, and WWVH (transmitting from Colorado and Hawaii). When distance between the
transmitter and receiver is known, the propagation delay can be accounted for and the theoretical accuracy limit is
about 100 ms.
Unfortunately, propagation delays in AM transmission are difÞcult to account for, since the signal can travel both by
sky wave and ground wave. The difference in propagation delay between these modes can be several milliseconds, and
occasionally the modes can cancel each other. Due to this lack of reliability and accuracy, these broadcasts are not
adequate for use as a synchronization source.
Loran C solves these problems by transmitting discontinuous (impulse) signals. Receivers easily differentiate and lock
onto the ground wave only. Once the exact location of the receiver is known, the delay is Þxed and can be calculated.
Only one station needs to be tracked to provide accurate synchronization. Timing accuracy can reach 200 ns, although
1 ms is a more typical and reliable number.

Loran C signals are very stable and continuously available around the clock. The system is funded in the U.S. through
the year 2000 and actively supported by a number of other countries. Timing is controlled by atomic clocks and
referenced to UTC. However, the signal does not contain time stamp data, and therefore an additional coarser time
reference (such as WWV or WWVB) may be needed to establish initial synchronization within 10 ms. A big problem
with application of Loran C in substations is its inherent sensitivity to electrical interference, noise, and corona. Power
line carrier in the 100 kHz frequency range causes direct interference. Corona causes noise in the same frequency
range that degrades performance. For reliable performance careful monitoring is required, a great disadvantage for
unattended operation in substations.
A utility may prefer to keep its primary frequency and time source at a central site with transmission to many remote
sites. A time code signal output from the central site may then be transmitted via microwave or Þber-optics to
equipment requiring synchronizing. The best resolution using microwave facilities approaches 1 ms by utilizing direct
phase shift modulation of a continuous wave signal on the baseband. The baseband requirements are about 60 kHz of
spectrum. Fiber-optic transmission would be similar, but could achieve higher accuracy due to the greater bandwidth
available and lower noise.
This method works for analog microwave but is difÞcult to implement on digital microwave and Þber-optic systems.
Digital systems gather slower rate data into packets and send them over the main high speed link. The time delays from
one packet to the next may be as great as 70 ms resolution over several hops, but greater accuracy requires custom
facilities. Custom Þber-optic links not used for data transmission routinely achieve synchronizing accuracy of 10 ns
and better.

Annex F
(Informative)
Time and synchronization formats
F.1 One pulse per second (1 PPS)
The 1 PPS signal shall be a positive pulse at least 5 ms wide and have a rise time less than 50 ns. It shall be capable of
driving a 50 W load to at least 5 V over 50 m of RG-58 cable.
F.2 IRIG-B format
IRIG-B is fully described in IRIG STANDARD 200-89, published by the Range Commanders Council of the U.S.
Army White Sands Missile Range. The minimum format shall be B122, which has seconds through day-of-year coded
in BCD and amplitude modulated on a 1 kHz carrier. Information needed for continuous precision timekeeping shall
be added to the code in the control functions as described below. Straight binary seconds may be added as allowed in
the standard, but shall not be required by either transmitting or receiving equipment. The amplitude shall be 1Ð6 V
peak-to-peak for the mark (peak), with a mark-to-space amplitude ratio 10:3, as provided in the standard. The time
code format is as follows:
<sync> SS:MM:HH:DDD <control> <binary seconds>
where
<sync> is the on-time sync marker
SS is the second of the minute [00 to 59 (60 during leap seconds)]
MM is the minute of the hour (00 to 59)
HH is the hour of day in 24 format (00 to 23)
DDD is the day of year (001 to 366)
<control> is a block of 27 binary control characters
<binary seconds> is a 17 b second of day in binary
F.3 Control bit assignment
By using IRIG-B with additional extensions, old and new time sources and time users can be easily integrated. PMUs
should be programmed to check the control bit Þeld and use this additional information where it is provided, but rely
on user-entered data where it is not. Where possible, these new assignments are made with zero indicating a normal
state since unused control Þeld bits are normally set to zero. This will minimize the possibility of creating a false
alarm. For example, if a control Þeld was all zeroes, the time quality code would indicate the clock was locked with
full accuracy that would not accidentally be interpreted as an error condition.
Virtually every timekeeping system is run by some kind of processor. Since IRIG time code numbers arriveAFTER the
on-time mark, the timekeeping system must generate the time-tag based on the anticipated number, rather than on what
it just got. Consequently, time counts that are not in exact sequence require advance notice. Non-sequence clock
counts include leap year, leap second, and daylight savings time changes. The leap second and daylight savings change
bits warn of impending special clock counts, and the last two digits of the year alert the timing system of leap year
changes.
As an interpretation of the IRIG standard, BCD time and Straight Binary Seconds (SBS) should always be consistent.
If BCD time changes by an hour for a daylight time change, SBS should change at the same time to reßect a consistent
count. The year will rollover with BCD time regardless of whether it corresponds with UTC time.

Table F-1ÑControl bit assignments
F.3.1 Year
The last two digits of the year is in straight BCD in the same format as the rest of the IRIG-B code and follows Þrst
after day of year. It will rollover with the day-of-year in the BCD time count.
IRIG-B
Pos ID
CTRLB
IT#
Designation Explanation
P 50 1 Year, BCD 1 Last 2 digits of year in BCD
P 51 2 Year, BCD 2 IBID
P 54 5 Not used Unassigned
P 55 6 Year, BCD 10 Last 2 digits of year in BCD
P 59 Ñ P6 Position identifier # 6
P 60 10 Leap second pending (LSP) Becomes 1 up to 59 s BEFORE leap second insert
P 61 11 Leap second (LS) 0 = Add leap second, 1 = Delete leap second
P 62 12 Daylight saving pending (DSP) Becomes 1 up to 59 s BEFORE DST change
P 63 13 Daylight savings time (DST) Becomes 1 during DST
P 64 14 Time offset sign Time offset signÑ0 = +, 1 = -
P 65 15 Time offsetÑbinary 1
Offset from coded IRIG-B time to UTC time IRIG
coded time plus time offset (including sign) equals
UTC time at all times (offset will change during
daylight savings)
P 70 19 Time offsetsÑ0.5 hour 0=none, 1=additional 0.5 h time offset
P 71 20 Time quality
4 b code representing approx. clock time error
0000 = clock locked, maximum accuracy
1111 = clock failed, data unreliable.
P 75 24 PARITY Parity on all preceding data bits

F.3.2 Leap second
The leap second pending (LSP) and polarity (LS) bits show that one is about to happen and whether it will be inserted
or deleted. Leap seconds have only been positive for the last 20 years, so LS = 0 is almost certain. The LSP bit should
be asserted at least 1 s and less than 60 s before the hour it is to be inserted. The bit should go to 0 when the second
count goes to 00. Leap seconds are always inserted at UTC midnight by altering the second time count only. Thus, in
UTC time the time count goes from 23:59:59 to 23:59:60 to 00:00:00 to add the extra second. In another time zone, say
PaciÞc Standard Time, 8 h behind UTC, the same count will be 15:59:59 to 15:59:60 to 16:00:00. SBS should give the
count 57 600 (=16:00:00) twice.
F.3.3 Daylight savings
The Daylight Savings Pending (DSP) and Daylight Savings Time (DST) bits indicate that a change is about to happen
and whether daylight savings is in effect. If DST = 0, then the impending change will be to ON, which will delete 1 h
from the time scale (leap forward 1 h in the spring) and the Daylight Savings bit will go to one. If DST = 1, the opposite
will occur. Daylight time changes will be 1 h and are asserted at the minute rollover. The DSP bit should be asserted
at least 1 s and less than 60 s before time is to be changed. The DSP and DST bits should change at the same time
between the 59 s and 00 s counts. In the U.S., where the time change is put into effect at 2 a.m., the time count in the
spring is 01:59:59 to 03:00:00. In the fall, the count is 01:59:59 to 01:00:00.
F.3.4 Local time offset
The local time offset is a 4 b binary count with a sign bit. An extra bit is included for an additional 1/2 h offset used by
a few countries. The offset gives the hours difference (up to ± 16.5 h) between UTC time and the IRIG-B time (both
BCD and SBS codes). Adding the offset to the IRIG-B time using the included sign gives UTC time (e.g., if the IRIG-
B time is 109:14:43:27 and the offset is -06 given by the code 0110 (.0), then UTC time is 109:08:43:27). The local
time offset should always give the true difference between IRIG code and UTC time, so the offset changes whenever
a daylight savings time change is made. Keeping this offset consistent with UTC simpliÞes operation of remote
equipment that uses UTC time.
F.3.5 Time quality
A 4 b time quality indicator is used by several manufacturers and is in several existing standards. It is an indicator of
time accuracy or synchronization relative to UTC and is based on the clock's internal parameters. The code
recommended here is by order of magnitude relative to 1 ns. It is basically the same as used in the HaveQuick and
STANAG 4430 (NATO) time codes, but with a more practical scale. The 1 ns basic reference is Þne enough to
accommodate all present industry uses now and into the foreseeable future. With present GPS technology at the 100 ns
accuracy level, a 0000 code indicating locked will go to a 0011 or a 0100 code at unlock.
F.3.6 Parity
A parity bit is easy to implement by simply adding all the bits in the message from BCD seconds through the time
quality control bits. (Straight Binary Seconds would not be included.) It provides some assurance the data is correct
and a secondary veriÞcation that the control bit Þeld has been implemented. An unused bit could not be mistaken for
parity, since parity will change each second most of the time as the second count increments. The last three unused bits
are left after the parity for user speciÞc assignment without affecting this code. They could be used for higher order bits
in a 4 b parity or LRC.

Table F-2Ñ4 b quality indicator code
F.4 High precision time code format
IRIG-B format transmitted using modiÞed Manchester modulation is recommended as an alternative to the AM
modulated IRIG-B with separate 1 PPS sync. This modulation is better adapted for both Þber and metallic digital
systems. With the previous control bit assignments, this time code format can serve all power industry requirements
now and in the foreseeable future.
Manchester coding provides a 0 mean code that is easy to decode, even at low signal levels. The 1 kHz clock provides
a precise on-time mark that is always present. The coding method mimics 1 kHz modulated IRIG-B with binary 1s and
0s in place of high and low amplitude cycles. A Manchester binary 1 is equivalent to a high amplitude cycle in the AM
modulation and a binary 0 indicates a low amplitude cycle. Using this modulation, an IRIG-B code "0" will be two 1s
followed by eight 0s. An IRIG-B code "1" will be Þve 1s followed by Þve 0s (see Þgure F-1). This conversion keeps
the codes compatible and makes translation or regeneration of the AM IRIG-B very simple.
ModiÞed Manchester modulation is not included in the IRIG standard, so this is not an IRIG code. However, the Range
Commander's Council is considering adding a Manchester modulation into the IRIG standard. This or a similar
Manchester modulation format may be speciÞed by 1997; users of this standard should consult IRIG and IEEE-PSRC
for the most recent recommendation.
F.5 Modified Manchester coding
Manchester modulation or encoding is a return-to-zero type where the pulse transition indicates binary 0 or 1. In this
case, a 1 kHz square wave is the basic clock modulated by the data to produce a rising edge to indicate a binary one (1)
and a falling edge to indicate a binary zero (0). The transition at every data bit provides good receiver synchronization.
Each bit period is half high and half low, so the mean is always half, making it easy to decode even at low levels. In
standard Manchester coding, the data edge occurs in the middle of the clock window to indicate a binary one or zero.
Since there is a data bit (i.e., transition) every clock period, it is easier to synchronize on the data than the clock cycle.
The "modiÞcation" moves the data window so the data is at the edge of the clock window that is on time with UTC (see
Þgure F-2). In another view, the modiÞcation simply deÞnes the middle of the window as "on time." What is important
is that the data edge is the "on time" mark in the code. This simpliÞes the construction of readers and regeneration of
the other IRIG code forms.
Binary Hex Value (worst case accuracy)
1111 F FaultÑclock failure, time not reliable
1011 B 10 s
1010 A 1 s
1001 9 100 ms (time within 0.1 s)
1000 8 10 ms (time within 0.01 s)
0111 7 1 ms (time within 0.001 s)
0110 6 100 ms (time within 10-4 s)
0101 5 10 ms (time within 10-5 s)
0100 4 1 ms (time within 10-6 s)
0011 3 100 ns (time within 10-7 s)
0010 2 10 ns (time within 10-8 s)
0001 1 1 ns (time within 10-9 s)
0000 0 Normal operation, clock locked

Figure F-1ÑIRIG-B coding comparisons: level shift, 1 kHz AM, and modified Manchester
Figure F-2ÑModified Manchester coding

Annex G
(Informative)
Alternative synchronization methods
Clause 5. deÞnes an algorithm for synchronization of the phasors by phase-locking the sampling clock to a time
reference. Alternative methods exist that meet the other requirements of this standard for producing synchronized
phasors. The following is one of the alternative methods.
For the purpose of direct comparison of phasors at different locations within a power system, each phasor must have
a common time and angular relationship, a common data format, and a common time reference. UTC has been chosen
as the time base for the reason previously shown. The common phasor format has also been described earlier. This
annex describes an alternative method for giving phasors a common time and angular relationship.
Synchronization is accomplished by forcing all phasors to be directly referenced to the synchronized 1 PPS. This
synchronization can be achieved by establishing an angular reference to allow the measured phasors to be referenced
to a common angle with an equivalent angular accuracy. The required level of precision for the synchronized sampling
method is 1 ms, and the corresponding angular reference precision is 0.022 degrees for a 60 Hz system and 0.018
degrees for a 50 Hz system.
It is not necessary to force the individual samples to be phase-locked to a time reference in order to achieve direct
phasor comparison. The same result can be achieved by forcing the samples to be phase-locked to local frequency.
Using adaptive sampling to adjust the time between samples, from which the phasors are calculated, automatically
avoids "leakage" effect (see [B1]) caused by off-nominal power system frequencies. Under the conditions shown in
5.5, the two methods will lead to identical phasors.
Phase-locking the sampling can also permit calculation of mixed quantity engineering units (watts, VARs, apparent
impedance, etc.) during the same cycle that the phasors are calculated. The data format will then allow real time
transmission of mixed quantities expressed as complex numbers as well as the voltage and current phasors.
The 1 ms level of precision has been chosen because of the availability of synchronization sources capable of this
accuracy. Depending upon the particular application, less demanding synchronization precision may be satisfactory. If
analog or digital processing of the sampled data produces a phase shift that effectively gives rise to errors in either the
sampling times or phase angle reference, these errors must be corrected and may be included as part of the header
information so that comparison between different systems can be made.

Ieee std 1344 1995(r2001)

More Related Content

Similar to Ieee std 1344 1995(r2001) (20)

Recently uploaded (20)

Ieee std 1344 1995(r2001)