IEEE Power & Energy Society
1996
TECHNICAL REPORT
PES-TR4 Formerly TP115
Relay Performance Testing PREPARED BY THE Power System Relaying Committee Relaying Practices and Consumer Interface Protection Subcommittee
© IEEE 2013 The Institute of Electrical and Electronic Engineers, Inc. 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.
THIS PAGE LEFT BLANK INTENTIONALLY
RELAY PERFORMANCE TESTING
POWER SYSTEM RELAYING COMMITIEE REPORT OF WORKING GROUP 113 OF THE RELAYING PRACTICES AND CONSUMER INTERFACE PROTECTION SUBCOMMITIEE
MEMBERS OF THE WORKING GROUP
G. E. Alexander H. N. Banerjee J. D. Brandt T. W. Cease L. Champagne G. Chirco A. F. Elneweihi J. Esztergalyos J. Hauber
A. Howard M. Kezunovic P. A. Kotas P. Lerley G. Manchur J. McConnell J. McElray P. McLaren K. Mustaphi
J. Nordstrom M. Sachdev T. Sidhu L. Smith W. Strang E. Udren B. Warwick K. V. Zimmerman
J. A. Jodice, Chairman M. Meisinger, Vice Chairman C. Renville, Secretary
Relay Performance Testing
Abstracting is permitted with creditto the source. For other copying, reprint, or republication permission, write to the IEEECopyright Manager, IEEEService Center445 Hoes Lane, P.O Box 1331,Piscataway, NJ 08855-1331. All rightsreserved. Copyright ©1996by The Institute of Electrical and Electronics Engineers, Inc.
IEEE CatalogNumber: 96TP 115-0 Additional copiesof this publication are available from IEEE ServiceCenter PO Box 1331 445 Hoes Lane Piscataway, NJ 08855-1331 1-800-678-IEEE 1-908-981-1393 1-908-981-9667 (Fax) 833-233 (Telex)
Relay Performance Testing
RELAY PERFORMANCE TESTING CONTENTS
1.0
INTRODUCTION
1.1
General
1.2
Definitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 4
1.2.1
Steady-State Test
4
1.2.2
Power System Simulation Test
5
1.2.2.1
Dynamic-State Test
5
1.2.2.2
Transient Simulation Test
5
1.2.3
Integrity Tests
6
1.2.4
Application Tests
6
1.3
Historical Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • .. 6
2.0
PURPOSE OF RELAY PERFORMANCE TESTING. . . . . . . . . . . . . . . . . . . . . . . . . .. 7
2.1
General
7
2.2
Integrity Testing
7
2.3
Application Testing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 7
3.0
TEST SIGNALS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 8
3.1
General
3.2
Measured Sisnals
3.2.1
AC Signals
3.2.1.1
Fundamental Frequency Steady-State Sine Wave
3.2.1.2
Sine Wave with Controlled Rate of Change in Magnitude. . . . . . . . . . . . . . . . . . . . . . .. 11
3.2.1.3
Sine Wave with Instantaneously Changing Amplitude
11
3.2.1.4
Sine Wave with Control of Phase Angle
11
3.2.1.5
Control Over Signal Duration
3.2.1.6
Line Synchronized and Self Synchronized Operation .........•................. , 12
3.2.1.7
Sine Wave with Controllable Frequency and Rate of Change in Frequency. . . . . . . .. 12
3.2.1.8
Sine Wave with Added Harmonic Content
3.2.1.9
Sine Wave with Transient Components Representing Actual Signals From a Power System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 12
, 4 4
0
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
••
8
8 ,.8 11
'.' . . . . .. 11
- 1-
,
,....... 12
3.2.2
DC Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 12
3.3
Environmental Signals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 12
3.3.1
Auxiliary DC and DC Interference.... . .. .
3.3.2
Conducted Electrical Interference
12
3.3.3
RF Signalsand RF Interference
13
3.3.4
Non Electrical Influencing Factors
13
3.4
Integrity
13
3.5
Application
13
4.0
lEST EQUIPMENT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 13
4.1
General
4.2
.
. .. . . .. ..
. . . . . . . .. . ..
12
13
. Passive Electromechanicallest Sources. . .. .. . . . .
. .
.. .
.. . . . .. .. ..
. .. . .. 14
Voltage Signals.. .
4.2.2
Current Signals. .. .. .
4.2.3
Phase Shifter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15
4.3
Electronic Test Instruments
4.3.1
Common Characteristics.. . .
4.3.2
Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 16
4.4
Transient Simulators
4.4.1
Analog Simulators. . .. . . .. ..
4.4.2
Real Time Digital Simulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 18
4.4.3
Playback Digital Simulators
19
4.5
Power ConditioningAmplifiers
20
5.0
lEST METIiODS
21
5.1
General
21
5.2
Integrity Test Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 22
5.3
ApplicationTest Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 22
6.0
EXAMPLE
22
6.1
Introduction
22
6.2
Sample Test Plan
23
7.0
CONCLUSION
24
.
..
. . ..
4.2.1
. . . .... . .
..
. . .. . . .. . .
. . .. .. . . . .... .. . . . .
. . . .. .. .. 14 . . .. ... 14
15 .. .
. .. ..
.. . .
. . ..
.. . .. .. 15
17
. .. .. . .. .
- 2-
. .. . .. . . . . . .. . . . . .
.. . . .
17
LIST OF ILLUSTRATIONS Figure 1
Steady-State Test Signals
4
Figure 2
Dynamic-State lest Signals
5
Figure 3
Transient Test Signals
6
Figure 4
Steady-State A-N Thst with Current Action. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 8
Figure 5
Steady-State A-N lest with Voltage Action
Figure 6
Phase-to-Phase lest with Current Action. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 9
Figure 7a
Dynamic-State Test Phasor Diagram
Figure 7b
Dynamic-State Test Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 10
Figure 8
Waveform Sample from Transient Simulation Test
Figure 9
Simple Current Test Circuit
Figure 10
Electronic Test Instrument . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 16
Figure 11
Analog Model Power System
18
Figure 12
Real Time Digital Simulator
19
Figure 13
Playback Digital Simulator
20
9
10
11
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15
- 3 -
RELAY PERFORMANCE TESTING 1.0 INTRODUCTION
which illustrates many test approaches. These examples are not intended to be test guides. The information is included to show how one utility applied specific tests to ensure that relay performance would meet application objectives.
1.1 General The purpose of this report is: •
to provide relay users with an understanding of the strengths and limitations of testing methods used for evaluating the performance of protective relays.
•
to serve as a reference for the development of test plans which determine relay performance and its suitability for application objectives.
1.2 Definitions The primary test methods and test purposes are discussed in this section. 1.2.1 Steady-State Test A steady-state test is used to determine the calibration point or setting for any measured parameter. Phasor test quantities are held stable for a duration much longer than the operating time of the relay, and are then varied in increments much smaller than the resolution of the relay. For the example of an instantaneous overcurrent relay where the nominal measuring time is 2 cycles, the duration of stable test quantities (Delta Time) would typically be more than 20 cycles. If the relay measuring resolution is 1.0 amp, the test current would be varied in increments (Delta Value) of less than 0.1 amp. Figure 1 shows part of a typical steady-state test signal. The voltage is maintained constant at the fault value, the current is increased at a rate of Delta Value/Delta Time.
'[his material is applicable to a wide variety of relays; it is not an instructional guide for testing specific types of relays. Various objectives associated with relay performance testing are identified; means to achieve those objectives are described. A discussion of test signals and the equipment used to produce them is presented to provide a better understanding of test methods. Most of the discussion in the paper is general, with non-specific background information. Section 6.0 concludes with an example of a comprehensive test plan
Figure 1 Steady-State Test Signals
- 4-
synchronously switched means that changes in phasor value (i.e., phase and/or amplitude) from pre-boundary values to post-boundary values occur in all phasors at the same time with no discemable skew. Power system characteristics such as high frequency and de decrement are not represented in this test.
1.2.2 Power System Simulation Test Relays are required to respond to the transient conditions of a disturbed power system. By simulating the signals "seen" by the relay under such conditions, setting and response time may be determined. The disturbance may be simulated in more or less detail by one of the two following test methods.
For example, three states may represent the pre-fault, fault, and post-fault power system condition. Additional states may be used to represent evolving faults.
1.2.2.1 Dynamic-State Test A dynamic-state test is one in which phasor test quantities representing multiple power system conditions are synchronously switched between states. The term
Figure 2 shows the test signals that might be used in a dynamic state test; prefault and fault states are shown.
Fault State 2
Prefault State 1
Figure 2 Dynamic-State Test Signals 1.2.2.2 Transient Simulation Test A transient simulation test signal represents in frequency content, magnitude, and duration, actual relay input signals received during power system disturbances. For example, signals may include transient de offset, and the effects of CT saturation and CCVT subsidence. Figure 3 shows Electro-Magnetic Transient Program (EMTP) modeled test signals that were used in a field transient test. Signals indicate that the circuit has tripped
- 5 -
single pole, is open (reclose period), and then recloses. The waveforms represent: a)
prefault load;
b)
the fault condition;
c)
the induced voltage during pole open;
d)
voltage at zero because of HV grounding switch close operation;
e)
voltage recovery and offsetdue to residual charge on the series compensating capacitor when HV groundingswitch opens;
a.
b.
c.
1) voltage transient and chargingcurrent whenlocal
end closes; g) return to load when remote breaker closes.
e.
d.
f.
g.
Figure 3 Transient Test Signals 1.2.3 Integrity Tests These tests are intended to establish whether the relay has been manufactured, delivered, installed, and maintained suchthat it meets itspublished specifications. These are considered routine and are performedon most relays at more than one stage in the relay life cycle. Integritytests are important basicprocedures, andshouldprecedeapplication tests. 1.2.4 Application Tests
Afterintegrity testshaveestablished that aparticularrelay meetsitspublished specifications, morecomprehensive tests may be applied to discover whether performance is satisfactory for its application objectives. Application tests areparticularly recommended whenpublished specifications are not sufficiently detailed to ensure proper application. Digital fault recorder records of specific disturbances maybe playedbackto the relayto assess performance. The disturbancemayalso be recreatedby mathematical simulation, and used to produce test waveforms for performance tests. Simulation offers the opportunityof varying disturbance conditions to more completely assess relaysensitivity and selectivity. 1.3 Historical Background There has always existed a need to test protective relays to confirm performance as designed by the manufacturer, and as required by the user. This process of performance evaluation helps to ensure adequateprotectionof electrical power equipment, and the powersystem as a whole.
- 6 -
Steady-state, and to a lesser degree, dynamic-state simulations have been the methods of choice for many years. These tests have been performed using passive component instruments. These included load box, phase shifter, variac, and switched inductance, capacitance, and resistive elements. In 1974 electronictest instrumentswhich synthesized and regulatedsine waves weredeveloped; these were adopted as an alternativemethod of providing phasor test signals. Electronic test instrument control methods were later improved to perform a complete range of dynamic state simulations. Full transient simulations previously required model power systems. This equipment is large, expensive, and normally onlyavailable in relaymanufacturers'factories, or in research laboratories. Transient simulation tests were usuallyperformedbymanufacturers duringprototypetesting of new relay designs or on special request by users for sophisticated or unusual applications. Usingcommercially available testequipmentitisnowpossible foruserstoperform transient simulation tests in the field or their workshop. Some users have found dynamic state simulations and transient tests cost effective methods for verifying correct application. Similartests havealsobeen used in the field to diagnose unexpected in-service relay operations. Since currentlyavailable test equipment performs a wide variety of tests, understanding the various relay test methods will assist users to develop test plans that minimize unexpected relayoperations and more quickly explain the ones that do occur.
2.2 Integrity Testing
2.0 PURPOSE OF RELAY PERFORMANCE TESTING
There are a number of occasions during the life of a relay when integrity tests are performed. Acceptance testing upon receipt is usually the first instance outside the factory. In-service integrity tests are generally performed on a periodic basis, which depends on several factors; the most common include:
2.1 General This section reviews the purposes for both integrity and application tests [1]. Testing should be complemented by thoroughly reviewing both the manufacturers' specifications and those power system parameters relevant to the relay application. Evaluating the suitability of a relay for a specific application is the first instance when both integrity and application tests may be performed. Integrity tests are general and verify relay conformance with both manufacturer and user acceptance criteria. These are usually steady-state or limited dynamic-state tests designed to examine the performance of all measurements performed by the relay.
• •
manufacturers recommendations operating history of the particular relay type
•
complexity of the protection function
• •
criticality of the protected circuit maintenance schedule of the protected apparatus
2.3 Application Testing Similarly application tests are performed periodically during the installed life of a relay. Instances include: •
Design Verification/Type Testing
Manufacturers employ a multitude of tests to ensure the relay is suitable for the application. Design specifications and critical performance characteristics are verified at the pre-production phase. Special tests may be performed to ensure a particular users application is satisfied. [3]
The diagnostic quality of integrity and application tests has improved as advanced test instruments have become available and affordable. Computer automated integrity tests examine relay performance more rapidly than manual methods. More extensive integrity and application tests may be performed, at lower costs. Application tests using dynamic state and transient simulation methods are being applied more frequently during the relay life cycle, improving protection system reliability.
•
Application tests are more comprehensive and are designed to examine relay performance specific to the application. Dynamic state and transient simulation tests simulate the power system conditions which relays will experience during service. Sufficient internal and external faults should be applied to examine the suitability of the relay for the given application. Multiple faults within the zone of protection should be applied to examine relay sensitivity. Faults outside the protected zone should be applied to examine selectivity. A test plan may incorporate a combination of both integrity and application tests. Complexity of both relay function and application form the criteria for selecting tests. For example: integrity tests verifying performance of simple instantaneous overcurrent relays generally include calibration at a defined current and measurement of operate time. These tests use sinusoidal steady-state and dynamic-state current, respectively. For applications where power system currents may not be sinusoidal due to CT performance or system harmonics, experience has shown instantaneous relay performance varies substantially from that which isexpected [2]. Applying a test current waveform which approximates the magnitude, frequency components, and d.c, offset of the protected circuit secondary is recommended to determine whether performance is suitable for the application. [2]
Relay Selection Prior to selecting a relay for a specific protection application, its response to expected power system conditions should be evaluated. A combination of dynamic state and transient simulation tests may be used, dependingupon the complexity of the application. Some utilities have modelled the transient behavior of critical circuits, and performed transient simulation tests prior to purchase; these establish the suitability of a particular relay design for the application. •
Commissioning/System Testing Application tests on individual relays minimize the possibility of setting errors. Since protection systems often comprise several discrete relays, the protection system should be tested as a whole to ensure appropriate interaction and coordination. •
Operation Analysis After an unexpected operation, application tests offer insight to the probable causes of malfunction. Recent experiences reveal inappropriate interaction between relays to have been the cause of scheme misoperation. A scheme test verifies coordination between relays. Operation of associated communications equipment, reclosers, breakers and transfer schemes is also verified. Fault and event recorder operation may also be confirmed. During the last few years end-to-end satellite synchronized dynamic-state and transient simulation tests have been used to examine overall protection and communication system operation. These have successfully identified malfunctions of individual relays in the scheme; of
- 7 -
solid-state relays, and of associated communication equipment. Failure of circuit breakers to operate properly has also been discovered. Application tests are notably appropriate in the following instances: •
To evaluate effects of less than ideal instrument transformer performance, such as (ct) saturation and (cvt) subsidence.
•
To evaluate effects of unusual system conditions such as ferroresonance, harmonic resonances, off nominal frequency, low voltage, loss of stability, high load, high source to line impedance ratio.
Tests should be performed with the relay mounted in the manner prescribed by the manufacturer. For electromagnetic relays, normal physical orientation of the case (vertical, level, etc) should be observed. For relays with electronic circuitry, the case should be grounded. The burden, and especially the capacitance and inductance, of test operation indicators should be low with relationship to the rating of contact or other output sensing devices. AVOID ERRORS!
•
To evaluate the effects of power electronic devices and their controls, such as static var compensators, variable series compensation, and local HVDC circuits.
•
To evaluate response to unusual conditions such as evolving faults, slow clearing external faults, and faults with high arc or ground resistance. 3.0 TEST SIGNALS
3.1 General AVOID DAMAGE!
Relays are measuring instruments. As with all measuring instruments the accuracy, resolution, and stability of calibration and test signals must exceed (relay) measurement specifications. Instability or errors in test signals may cause inaccuracies in settings. Subsequent indeterminate response to power system disturbances may result. 3.2 Measured Signals 3.2.1 AC Signals This section reviews the requirements for voltage and current test signals used for both integrity and application tests. Control of parametric values of frequency, amplitude, and phase angle relationships is presented in terms of relay measurements. Test source specifications and their effect on test results is also discussed.
Test signals may exceed the continuous rating of the relay. Test equipment may be capable of damaging the relay if carelessly applied; the damage may cause immediate failure or operation may be slowly degraded. Marginal operation followed by failure when called on to operate under power system transient conditions is a more serious event.
The voltage and current parameters for most relay tests approximate a power system fault condition. Voltage and current amplitudes and phase angle relationships at the power system frequency must be independently controlled to simulate the fault phasor relationships.
Persons responsible for designing the test must be cognizant of relay I 2t specifications, and voltage limitations. Test designers must ensure that those responsible for carrying out the tests are aware of the precautions necessary to avoid damage or reduction of life. Unnecessary tampering or adjustment of the relay must be discouraged to reduce "adjustment induced" failure.
For steady-state tests, phasor parameters are slowly changed to determine relay calibration setting. Single phase faults require a change in one parameter, while the remaining phasors describing the fault remain constant. The phasor diagram in Figure 4 shows the constant, (preset) voltage phasor relationships which describe an A-N Fault Condition.
VA = VFAULT
Figure 4 Steady-State A-N Test With Current Action
- 8-
(II) represents an incrementing current used to determine A-N reach.
In the Phasor Diagram of Figure 4, VA is set at the fault voltage; VB and Vc represent balance. The dotted phasor
VA =ACTION
+ !'+ I I
11=1 FAULT
Vc Figure 5 Steady-State A-N Test With Voltage Action In the Phasor Diagram of Figure 5, 11 is set at the fault impedance ratio and phase-to-phase fault voltage current value. VB and Vc represent balance. The dotted magnitude. Vc is set at the unfaulted value. A single test phasor (V A) represents a decrementing voltage used to current (IAB) is injected through the series connected Phase determine A-N reach. A and Phase B inputs of the relay. Depending on test protocol, the fault current lAB is incremented or In both cases a single parameter is varied to determine decremented. VA and VB are maintained at their fault the steady-state reach calibration setting. values. Phase-to-phase faults require changes in two, Alternatively, the fault current may be held constant phase-to-neutral phasors. These geometrically sum to while VA and VB are synchronously decremented to form the faulted phase-to-phase voltage. In Figure 6, VA determine reach. and VB may be set to represent any source to fault
~\
I \ I \ I \ I I A I I I I
11
V \
lAB
=ACTION
/ / /
/
//
VAB=V FAULT
Vc Figure 6 Phase-to-Phase Test with Current Action
These simple examples demonstrate amplitude control requirements for basic steady-state tests. Tests for system disturbances involving additional phases require similar control of the involved phasor qualities.
- 9 -
Dynamic state tests are used to determine the time response of a relay to a predetermined fault condition. For example, run the previously described steady-state test to determine A-N reach; then perform a dynamic state test to reveal operate time for the fault condition.
VA = V FAULT
30V~
70v 1-240 0
VB
Vc
VB
Vc
70v 1-120 0
70v 1-240 0
70v 1-120 0
Fault State 2
Balanced Prefault State 1 Figure 7a Dynamic-State Test Phasor Diagram Prefault State 1
Fault State 2
Figure 7b Dynamic-State Test Waveforms
The dynamic-state test requires synchronous switching between phasor states. In this example the balanced, pre-fault phasor quantities are STATE 1: the previously determined steady-state reach conditions are STATE 2. Both phasor states are shown in Figiure 7a. Operate time is measured beginning at the transition from STATE 1 to STATE 2 asshownin FigureTh. Responseto evolving faults may be determined by synchronous switching between successive fault conditions each represented by one set of phasor states. Responseto faultsat varying distances from the relay mayalso be determined. A profileof the operate timeversus faultlocation maybe obtainedby tabulatingthe results of a sequence of tests in which the phasors are switched from the balanced condition to conditions
representing the desired fault locations. Complex microcomputer relays which measure additional phasor quantitiesaresimilarly evaluated. Dynamicstate transitions may include magnitude and phase angle for one, or all phasor values, as required to properlyrepresent the faulted system condition. Transient simulation application tests require informationdescribing system waveforms fromtransient mathematicalmodelsor from digital fault recorders. Microcomputer relays nowofferwaveform data, but at samplingrates lower than the other sourcesmentioned. Transientwaveforms are described by tens to hundreds of digital coefficients per cycle. Each coefficient represents the instantaneous value
- 10 -
and frequency to properly reproduce the system disturbance. Discussions on sampling frequencies and data format may be found in the COMTRADE Standard [4] recommended for anyone considering use of transient simulation application tests.
of the waveform, at a point in time. For polyphasesystem representation, synchronism between all sources of digital information must be precise to ensure correct phase relationship. Any amplifying means must similarly provide constant or predictable phase shift vs. both load impedance
VA
VB VC
11
IAAQ(l \TV v V
V
==ooaA!\f\f\f\
v
v
v
v
C7
V \TV
\TV
ADAAAAAAAAAAAAAAAAAI JVVVVVVVVVVVl}VVV\[VVV \AAAAAAAAAAAAAAAAAAA VV \[\[V VVvVV\[V \)\[VVVVV ~ Df\ f\ f\
12 13 Figure 8 Waveform Sample from Transient Simulation Test 3.2.1.1 Fundamental Frequency Steady-State Sine Wave
This signal is used for steady-state tests. Total harmonic distortion should be substantially less than specified in IEEE Standard No. 519[5] "IEEE Recommended Practices and Requirements for Harmonic Control in Electric Power Systems". Standard 519 defines different limits of total harmonic distortion for different primary voltage levels. To ensure repeatable test results, test equipment should limit total harmonic distortion of currents and voltages to 1.5%. The maximum for an individualharmonic should be wellless than 1%. The magnitude of test phasors should be wellregulated ( < 0.5%). Phase angle variations should also be regulated to less than 0.50 • Variations may seriously affect the accuracy of a test. A periodic check of the test equipment for possible variation can be used to establish test tolerances. Tolerances published by the relay manufacturers should be considered when selecting test equipment. 3.2.1.2 Sine Wave with Controlled Rate of Change in Magnitude.
The change may be by means of continuous ramp or step function, with delta value/delta time approximatinga linear ramp as defined in 1.2.1. The rate of change should be adjustablefrom near instantaneous (reachingthe newvalue in lessthan 5% of the nominal relayoperating time), to near steady state, with rate of change less than 1% of maximum value per second. This signal is used for pickup/dropout tests.
3.2.1.3 Sine Wave with Instantaneously Changing Amplitude.
This signal may be used for dynamic state simulation tests. Synchronously switchingbetween numerous states provides a dynamic state simulation of evolving fault conditions. Control over the point on wave at which the change in all phasors occurs may be required. 3.2.1.4 Sine Wave with Control of Phase Angle.
Instantaneous and controllable rates of change in phase anglerelationshipsmaybe required. Stabilityof phase angle relationships with changes in power line and relay load, should preferably be 0.50 maximum. Resolution in phase angle control should be less than 1 0 ; 0.1 0 is appropriate for microcomputer relay designs. 3.2.1.5 Control Over Signal Duration.
Some relays also measure time as part of their function. To establish the time response of a relay, the duration at which test signals remain constant at a test value must be controllable. The required precision of time control will depend on the precision with which the time is being measured by the relay. When a dynamic state test is used to measure time response at operating quantities, the number of cycles at the fault valueshould represent the measuring time of the relay, as a minimum.
- 11 -
3.2.1.6 Line Synchronized and Self Synchronized Operation Test signal sources require frequency accuracy and stability in proportion with relay measurement capability. Typically, quartz crystal based references are used to attain accuracy of approximately0.01 Hz and stabilityof 50 to 100 ppm.
Some electromechanical relays are susceptible to electromagnetic fields from generator and distribution busses. These relays may appear to perform incorrectly when tested with signals having independent crystal based references since these are asynchronousto the local power system fields. Variations of the relay operating point may be observed as the test signal beats with the interfering system frequency. Such relays are in wide serviceso care must be observed and interference effects noted. To determine the setting error due to interfering magnetic fields, run one test in line synch mode to determine one set of test results. Run a second test with phasors reversed 1800 • The actual setting is the averagebetween the two results. If the variationfrom the average is significant,check the applicationof the relay in its environment. 3.2.1.7 Sine Wavewith Controllable Frequency and Rate of Change in Frequency The accuracy and stability of the frequency should be some small fraction of the resolutionof the relay.This varies from 0.02 Hz for axial turbine synchronizing relays to 0.1 Hertz for general purpose frequency relays. The rate of change should be controllable similarly to that described above for amplitude control.
Rate of change in frequency vs. time ranges from 0.01 Hz/second to tenths of Hz/second depending on power system design. 3.2.1.8 Sine Wavewith Added Harmonic Content Control of harmonic frequency, its magnitude and time (phase angle at each harmonic) relationship may be required. These signals are used for testing relays such as harmonicallyrestrained differential relays, and tuned relays such as third harmonic sensing stator ground fault relays. Reference [2] discusses the effect of harmonics on relays, and is particularly useful when designing application tests. Commonly required harmonics are 2nd, 3rd ,5th, 7th, 9th, 11th, 13th, and 15th. Harmonic frequencies higher than 1 kHz are rarely required in a measured signal. 3.2.1.9 Sine Wavewith Transient Components Representing Actual Si~nals From a Power System These transients may include;
•
Fault or switchinginduced de offset,
•
Instrument transformer (Cf) saturation, or (CVT) subsidence, or ferroresonance,
•
Power system swings or current reversals,
•
Power transformer ferroresonance,
•
Inrush from transformer, reactor, transmission line, cable, and shunt capacitor energization,
•
Effects of nonlinear devices such as surge arresters, thyristors,spark gaps,and powerdiodes,
•
Circuit breaker current chopping and transient recoveryvoltages. The precedingitems are not meant to form an exhaustive list; they give an idea of the variety of transients that must exist in the test signal which simulates system conditions, This signal is used for transient simulation tests where the applied test signal represents the faulted power system as closely as necessary to evaluate proper relay performance. 3.2.2 DC Signals Some relays which measure de quantities are special purpose devices. DC test signals which examine performance should be a subject of agreement between the manufacturer and user. Sincegeneralizationisnot appropriate for such specialpurpose relays,they willnot be discussed further. 3.3 Environmental Signals Manufacturers normally specify applicableenvironmental standards for whichtheir relays are designed. Users will normally only apply such signals when the manufacturers specificationis silent on a particular aspect of performance, or for troubleshooting. This document does not address environmental signals. Other standards and references discuss environmental signals in great depth; they will be mentioned briefly here.
Some of the test instruments used to provide a.c, test signalsmay be susceptibleto the environmental test signals when simultaneously applied to the relay. AC test instruments should conform to the same environmental test standards as the relay. Alternately, the test instruments should be evaluated to ensure its performance in the presence of environmental test signals is known. 3.3.1 Auxiliary DC and DC Interference If a relay requires a de power source, various test signals may be applied to the relay's internal dc power supply. Maximum and minimum de voltage levels may be applied (with or without superimposed ripple) for verifying operation under actual measurement conditions, and for measuring dc burden [3]. Voltagedips and interruptions of various magnitudes and duration may be applied to the power supply [6]. 3.3.2 Conducted Electrical Interference The effect of conducted AC surges and fast transients which appear on any relay terminals are important, and
- 12 -
should be investigated. These signals are applied to all terminals except those specifically exempted by the manufacturer's specifications in accordance with the standard on Surge Withstand Capability [7].
3.3.3 RF Signals and RF Interference RF signals from nearby transceivers have interfered with some electronic relays and also with electromechanical relays with solid state electronics. Manufacturers and users are referred to [8] for guidance. 3.3.4 Non Electrical Influencing Factors Altitude, temperature, humidity, and seismic shock are other influencing factors on relay performance [9]. These factors are not discussed here. 3.4 Integrity Manufacturers design integrity tests to verify the condition of the relay using simple test signals. Some users add tests of their design to verify certain relay functions, which they have found important by experience. The majority of test signals consist of steady-state sinewaves which check relay setting and operation (3.2.1.2.) Dynamic state test signals are used to measure relay response time. (3.2.1.3) Test signals with various harmonic content may also be applied as integrity tests (3.2.1.8). Harmonically restrained differential relays are the most common types. Transient test signals may also be used for integrity tests; this technique is growing in acceptance as suitable equipment becomes available. Generator synchronizing and load-shedding frequency relays require rate of change in phasor quantities, to verify performance (3.2.1.7). Since integrity tests are routinely performed in the field, test signals which can be produced by portable, robust, and easily operated equipment are required.
3.5 Application Application test plans are intended to verify that relay performance is appropriate for a specific application, or for general applications in a power system. Application test plans may utilize all of the signals described in 3.2.1. Steady-state test signals verify the capability of the relay to accept the required settings. These tests are usually supplemented by other test signals for complete application tests. Dynamic state test signals establish the speed and transient response of the relay [10]. Ensure that dynamic state tests are suitable for the type of relay being tested; the high rate of change in test signal values may giveunexpected results. [11] If dynamic state tests give unexpected results,
the relay performance should be verified by full transient tests, or the results discussed with the manufacturer. Since they closely simulate actual fault conditions [12] transient tests signals are required to fully assess relay performance. Transient tests need include only those system conditions relevant to the application. Harmonic contaminated test signals may also be applied in some application test plans, particularly where the measured signal is also contaminated with harmonics. Environmental test signals may be applied in application test plans, particularly where the manufacturer's specification is silent about an environmental factor which may be present in an application.
4.0 TEST EQUIPMENT 4.1 General Equipment used for testing protective relays spans a wide range of technologies. Virtually all prior test equipment developments remain in use today. Earliest relay test equipment included general purpose passive components, interconnected to provide the necessary test signals. Variable resistive, capacitance, and inductive elements were used to produce voltage and current test signals and to create phase angle relationships between test signals. All outputs were metered during the test process to provide a continuous indication of actual values. This method used the power mains as a signal source. Later, test equipment manufacturers assembled these passive components into packages, organized specificallyfor relay testing. Control circuits, signal switching, and metering were incorporated to simplify operation. More recently, electronic test instruments synthesized test waveforms and incorporated direct reading digital controls for amplitude, phase angle, and frequency. Voltage and current outputs were regulated and unaffected by variations in mains voltage or waveform, or changes in relay burden. From earliest periods, analog model power systems were used by relay manufacturers and research laboratories. There were power systems in miniature having three phase generators driving lumped-constant RLC lines. Relays switched interconnections to produce fault conditions. and PT provided analogs of voltage and current representing the model systems' performance. Graphic recorders were used to provide a record of system response to disturbances. Initially voltage and current outputs were capable of limited burden. Voltage and current amplifiers were used to increase CT and PT voltampere output sufficiently to drive protective relays.
cr
Later, higher powered model lines were capable of driving relays directly. Interaction between relay and system model is an important feature of the analog model power system.
- 13 -
Today, computer based electronic simulators of various types have become available, and are used in increasing numbers by utilities and manufacturers alike. The most basic playback simulators use records of actual faults from DFR or system waveform information produced by mathematical models. Real time digital simulators provide the interactive flexibility of analog model power systems. They compute initial, balanced system conditions and the changes in system conditions due to disturbances, all in real time, e.g.: protection system response modifies the system parameters and conditions are re-computed in real time. Section 4.2 briefly reviews the characteristics of the passive test equipment, and presents criteria for its application. Electronic test instruments are used extensively today. They are described in more detail to provide an overview of their broad applicability. Signal generation, parameter control and amplifier requirements are discussed. Model power systems are described in sufficient detail for the user to comprehend concept and basic applications. Refere nces are extensive. Users must be aware of the limitations in all test equipment and in various methods. Even the most sophisticated test equipment is not capable of compensating for poorly organized test protocols or incorrectly applied signals. Intelligent planning and observation, coupled with effective documentation and appropriate instrumentation are mandatory adjuncts to all test sessions. 4.2 Passive Electromechanical Test Sources The oldest form of relay test equipment used passive electro-mechanical components. Load box, variac, phase shifter, and measuring instruments were interconnected to provide the range of signal amplitudes and power required for the particular relay. Single or three phase mains power sources supplied commercial quality sinewave signals and energy to drive the relay burden. 4.2.1 Voltage Signals Adjustable autotransformers commonly known as "variacs'' are used to provide secondary voltage and current
test quantities. Variacs can be utilized in single phase, three phase open delta or three phase wye configuration. They modify the mains potential, producing a source of variable voltage. Outputs are not isolated from the power mains and are a shock hazard. Relay inputs must be connected appropriately to prevent incorrect signal paths. Variacs can be a source of undesired harmonics. Variacs are also utilized as an integral extension ofeither the load box or the phase shifter to vary the voltage or current test quantities. 4.2.2 Current Signals A current test circuit typically consists of a variable autotransformer, a multi-ratio transformer, a series impedance, and an ammeter, as shown in Figure 9. When supplied with a distorted waveform, relay operation may be out of tolerance even with correct metered values. The discrepancy may take the form of different pick up values and/or different operating times.[2,13] The relay burden is often a non linear saturating inductance. The non-linear burden of the relay will cause distortion of the test signal if the current test circuit presents a low source impedance. A linear series impedance is used to increase circuit impedance and so reduce waveform distortion due to changes in relay burden. To be effective, this series impedance (often a load box) should be at least six to ten times the impedance of the relay under test. [13] Distortion in the power mains contributes to distortion in voltage and current test signals; unexpected relay performance may result. Load Boxes.
A load box is a collection of switch selectable fixed impedances, usually resistors; these are arranged to provide a variable impedance. Load boxes are frequently equipped with a vernier adjustable impedance, an ammeter and a load shorting switch. When used in conjunction with a voltage source (they commonly plug directly into the domestic mains) they can be used to provide a source of test current.
- 14 -
Multi Ratio Transf.
Variac
Load Resistors
Tap Select
Variable Resistor
Load Shorting Sw. L..----(O
Output Terminals
Figure 9 Simple Current Test Circuit 4.2.3 Phase Shifter
•
Transformer coupled feedback amplifiers provide regulated voltages and currents. Variations in test signals due to fluctuations in the power source, or variations in relay burden are minimized.
•
Internal alarm systems eliminate use of external meters
The phase relationships between test signals can be developed by using a specially constructed three-phase "wound rotor" or with switched transformers. The output voltage from a special three-phase induction motor with a balanced and symmetrical rotor (primary) and stator (secondary) windings provides continuously variable phase relationships with respect to its input. A six-phase to three-phase tapped transformer, tap changing switches and a variac also provide variable phase angle output voltages with respect to its input. Tap selection provides coarse selection of phase angle; the variac is used for fine tuning. 4.3 Electronic Test Instruments 4.3.1 Common Characteristics Common characteristics of electronic instruments designed specifically for protection testing include: •
Internally generated sine wave reference
•
Variable frequency operation
•
Direct digital control of voltage, current, phase angle, and frequency
Electronic relay test instruments produce sine waveforms of voltage and current at appropriate levelsfor testing relays. Typical voltage and current maximums are 300V rms and 35A rms, at power ratings range up to 100 VA. Higher powered current sources for high current tests on electro-mechanical relays are rated around 150A and 500 VA. Advanced functions provided by newer digital designs include synchronous switchingof phasors for dynamic-state tests and playback of waveforms from external digital files for transient simulation tests. Use of external frequency references from satellites allows synchronization of remotely located test instruments for in-situ end-to-end tests on relays. D.C. coupled feedback amplifiers offer reproduction of the d.c, transient component of fault waveforms. P.C. based programs for steady-state, dynamic and transient test methods are in general use; automated integrity testing for periodic maintenance testing is a common application.
- 15 -
EXTERNAL REFERENCE MAINS POWER INPUT
~
VARIABLE DIGITALDIVIDER
PHASE LOCKED LOOP
FREQUENCY REFERENCE (GPS ETC.)
FREQUENCY REFERENCE
--------------DIGITAL ANALOG OUTPUT SAMPLE CLOCK
WAVEFORM GENERATION ERROR ALARM ___
-
-
--~__t
-
-
-
-
ERROR DETECTION
-
-
~
-
-
-
-
-
-
-
-
-
-
-
--
\V
------.......- .....----< FEEDBACK CONVERTIBLE VOLTAGE CURRENT DIRECTCOUPLED OUTPUTSTAGE
ANALOG INPUT
CONDITIONING AMPLIFIER
TRANSFORMER COUPLED CURRENT OUTPUTSTAGE
TRANSFORMER COUPLED VOLTAGE OUTPUTSTAGE
AMPLIFICATION AND OUTPUT STAGES
Figure 10 Electronic Test Instrument 4.3.2 Operation Frequency Reference
Test signal frequency is established using a high frequency clock. The clockmaybe derived froman internal temperaturecompensatedquartzcrystal or froman external reference. The powerlineisa convenientexternalreference for LineSynchronization. Geostationary OrbitingEnvironment Satellite (GOES) and Global Positioning Satellite (GPS) satellite receivers provide access to Universal Coordinated Time (UTC) for synchronization of remotely located instruments.
locked loop (PLL)synchronized to an externalsignal. For line synchronization the PLL multiplies a signal from the nominal 60 Hz line by 3600, producing216,000 Hz sample clock, lockedto the line. For synchronization to UTC the PLLmultiplies the one pulseper second(lPPS) signalfrom a GPSreceiver by216,000, locking the sampleclockto UTC time. The PLLincludes a variablemultiplier, thisprovision allows generating arbitrary clock frequencies for transient simulation waveforms which mayhavesamplerates ranging from ten to thousands of samplesper cycle. Waveform Generation
Under the controlof the instrument microprocessor, the An intermediate frequency sample clock is produced
using the internal crystal frequency, divided by a variable divider to provide variable frequency operation. Alternatively, the sampleclock can be derived from a phase
sample clock drives a binary address counter; this strobes a random access memory (RAM) containing waveform
co-efficients. Each clock pulse sequentially advances the waveform address strobe to read a newvaluefrommemory.
- 16 -
For sinewave generation, 3600 sine co-efficients are stored in memory, in this example; one value is stored for every0.1°, from 0° to 359.9°.The address counter wraps around afterevery3600 counts and returns to the start point to produce a continuous waveform. Phase control is achieved byoffsetting the start point of the address counter. The phase angleisset by front panel controlsor byexternal computercontrol. Sincephasecontrol iscompletely digital, thisdesign does not create phase errors when the sinewave frequency is changed. When reproducing transient waveforms, the stored coefficients are those of the desiredwave. Coefficients may represent any digital number within memory depth (typically 16 bits) and any number of samples within memorycapability. The clockfrequency isset to the desired samplereproductionrate; the addresscounterruns through a sampleset once, then stops. UsingEMTP data at 10J,lSec. per sample for example, a transient event of one second requires 100,000 samples or 1666/1667 samples per 60 Hz cycle. The numerical output of the waveform memory drives a digital to analog converter(DAC)providing instantaneous analogvalues for each co-efficient. The sequentialsamples producethe desiredwave at a smallsignal referencevoltage, typically around 10volts peak. That referencewave is fed into a multiplying digital to analogconverter, (XDAC). This scales the sinewave amplitude according to the digital value set by the front panel controlsor externalcomputercontrol. When the XDAC is set to full scale, scaled amplitude is equal to the reference wave amplitude. The analogoutput signal drives poweramplifiers to attain VAlevels suitableto drive relay inputs. Amplification and Output Stages
The smallsignal output from the waveform generator is amplified by negative feedback amplifiers to produce relaying level voltages and currents. Because relays contain saturating electromechanical components, the amplitude and phase of the relayburden can change by orders of magnitude. This mandates amplifier stability under rapidly changing load impedance. The amplifiers maybe voltage, current,or convertible types capable of being switched between voltage and current modes. The amplifier output is fed to the output terminals byan output stagewhich conditions the feedback, and rangecontrol relays. Transformer coupledsources include a variable ratio output transformer. This maximizes amplifier power transfer to the load by impedance matching. DC coupled sources drive the relay directly. Feedback is derived from a current shunt for current sources and from a resistorfor voltage sources. Negative feedback around the amplifier compares the output withthe input,correcting errors in the
output causedby changesin the load impedanceor the line powersupply. The output is monitored by error detection circuits and an alarm issoundedifthe output containsamplitude or total harmonic distortion errors exceeding test instrument specifications. 4.4 Transient Simulators
Descriptions of the designs and operationalcharacteristicsof the three mostprevalentformsof transientsimulator available today are included in this section. Transientsimulators accurately representthe steadystate and the transientpowersystem waveform. Earlysimulators were analog, and used scaleddownor electronicmodelsof power system elements. These model power systems tend to be large and expensive and have been confined to manufacturers, larger utilities, and research laboratories. Playback digital simulators using off-line EMTP solutions and conditioning amplifiers allow stored digital waveforms to be fed to relays. These simulators are less costly and smallerthan the ModelPowerSystem (MPS), but cannot acceptinput from the device under test and interact. The MPS has the advantage of runningin real time and can be used in an interactive mode with the device under test. Real time digital simulators are nowavailable. Running EMTP in real time with parallel processing computers which driveconditioning amplifiers, combines the smallsize of the playback simulator with the real time interactive capability of the MPS. They are more expensive than playback simulators. 4.4.1 Analog Simulators
Lowand high power analogsimulators or MPSutilizing lumped parameter models have been used to simulate power systems for many years. These provide analog waveforms fortestingrelays understeadystate and transient conditions. The lowpower MPS requirespowerconditioning amplifiers. These convert the signals to levels suitable for driving the relayunder test. Amplifiers are described in Section 4.5. The highpowerMPS operates at signal levels which allow direct connection between the MPS and the relayunder test. See Figure 11.[14] The MPS provides facilities for representationof all the important elements of a power system.[14] These include transmission lines, shunt reactors, series compensation capacitors, circuit breakers, and generation sources. cr and CVT models are also available. Fault location, fault type and point-on-wave fault initiation are controllable to research wide ranges of power system operation. Relayoperation and the MPS waveforms are measured and recorded to provide information describing how the relaywould performon the actualpowersystem. The MPS usesac contactorsto simulatecircuitbreakers;these maybe operated from the trip contacts of the relay or recloser where appropriate.
- 17 -
"
LOW ANALOG MODEL POWER POWER SYSTEM HIGH POWER
~~
.........
.........
I---
........
CONDITIONING AMPLIFIERS
JIIlIlIII""'"
RELAY SYSTEM UNDER TEST
I
--.
USER INTERFACE
"
MONITORING EQUIPMENT
Figure 11 Analog Model Power System Because of size and cost considerations, the high power MPS is limited in the extent and number of power system components which are modeled. It generally models a smaller part of the power system to simulate the most important conditions for relay testing. The analog MPS is very flexible providing rapid testing over a wide range of parameters. Load flow, source to line impedance ratio, fault incidence angle and evolving faults are typical simulations. Many cases can be modelled in a short time period, efficiently locating the worst case condition for the relay under test.
4.4.2 Real Time Digital Simulators Advances in computing technology have now made it possible to run transient simulation programs in real time on digital simulators.[15][16] Instead of scaled down physical models or electronic models of system elements, digital simulators solve the equations representing the behavior of the system elements. The hardware does not change from one study to the next, only the software. The simulator can model any system element for which there is an accurate software subroutine. The initial digital simulators [15] were based on parallel processing techniques and used many high speed digital
signal processors. Parallel processing means that the computations required in a particular time step are shared between several processors. Each processor performs its calculation independently of the other processors. Information exchange between processors occurs at the end of each time step. By sharing the computational load in this way the real time simulator overcame the inability of a single processor to solve the system equations in realistic time steps of 50J,ls. Some of the newer Reduced Instruction Set Computer (RISC) workstations are now (1993) capable of handling the computations for a power system of sufficient size for testing relays [16]. Both the parallel processing based simulator and the RISC based simulator use extensive rewrites of the established electromagnetic transients programs in order to meet the constraints of running in real time. The simulator is generally configured for the application using graphics based software for system diagrams and data entry. This software performs the same task as "patching" the R.L.C. models in an analog simulator. Software allows the user to set up steady-state prefault conditions by altering set points on voltage regulators or governors, taps on transformer banks, etc., then applies the fault on request.
- 18 -
~, REAL TIME SYSTEM SIMULATOR
.......... ......
D/A CONVERSION AND CONDITIONING AMPLIFIERS
.........
.......
RELAY SYSTEM UNDER TEST
~~
--..
USER INTERFACE
"
MONITORING EQUIPMENT
Figure 12 Real Time Digital Simulator Typicalphysical connections to the simulator are;
•
between the Digital to Analog (01A) output ports and the test current and voltage conditioning amplifiers
•
between the ac test signals, the logic signals representing breaker states, and a digital fault recorder for monitoring
•
between simulator input ND ports or logic gates and the relay trip contacts or the output of any other apparatus under test. Examples of other apparatus include a metal oxide varistor energy monitor in a series compensation application, or a power system stabilizer in a power swing study, etc.
Real time digital simulators occupy less space than their analog counterparts and are less expensive. Digital simulators can be interfaced to existing analog simulators via DIA and ND connections creating a "hybrid" simulator.
waveform and logic signal data of a real system event from a DFR. It may also contain waveforms calculated off-line using an EMTP simulation program [17]. Playbacksimulator hardware is essentially that described in 4.3for electronic test instruments, many of which provide playback functions. Hardware elements include RAM storage for both power system signals and logic signals. These data are synchronously clocked to DIA converters which produce low level analogs of the waveforms. Conditioning amplifiers produce cr and PT secondary level test signals at voltampere levelssuitable to drive relays or relay systems. Simulator test waveforms and relay outputs are monitored by digital fault recorders or equivalent instruments. The resultant profile of test and relay signals vs. time documents the results for analysis. Graphic software running on PCs or workstation preprocesses the data before downloading to the playback simulators.
4.4.3 Playback Digital Simulators A playback digital simulator is much simpler than a real time digital simulator. The essential difference is that the action of the device under test has no effect on the simulation. The playback simulator provides real-time playback of voltage, current, and logic signal waveforms which have been stored in' a file. The file may contain
- 19 -
Editing functions include: •
Selecting the files required to run tests
•
Converting files to one sampling rate for synchronous replay
•
Scaling data to establish secondary levels
cr and PT
•
Identifying coefficients on various waveforms at which protection timing begins
•
Converting files to binary for downloading to waveform memory
•
Storing test results
For DFR and EMTP files, extending the prefault data period is an important editing function. Sufficient prefault is required to allow relays to normalize operation before disturbances are applied. Post-fault periods may require
extension as well, if reclose functions are involved in the test plan. In the event that tests using actual DFR records are insufficient to establish relay performance, complementary EMTP studies will be required. Power system and relay application experience will dictate the level of complexity required for playback simulators. In comparison to todays real time simulators and model power systems, the off-line EMTP runs can represent very complex power systems.
EMTP SIMULATION DFR OR RELAY RECORD
D/A CONVERSION AND CONDITIONING AMPLIFIERS
DATA FILES
RELAY SYSTEM UNDER TEST
"
"
MONITORING EQUIPMENT
USER INTERFACE
Figure 13 Playback Digital Simulator
4.5 Power Conditioning Amplifiers All electronic test and simulation systems require power amplifiers to convert their output signal levels (typically 10V from a high impedance source) to the secondary levels of the main CT and PL The nominal CT secondary level is SA or 1A; and the nominal PT secondary level is 70 or 120V
..±.
Voltage Signal For voltage sources, two to three per-unit values are required. For 120~ RMS phase-to-phase quantities, two per unit values reach 336V-pk.
(2 x
f2 x 120V =
336V pk.)
Current sources must be capable of ranges from 10 per unit up to a maximum of 50 per unit for some power systems. Typical 20 per unit, SA secondary current values are:
20 x
f2 x SA
=
141A pk.
Power levels for current sources are much higher than required for voltage sources due to the higher per unit values required to effectively simulate power system conditions. Factors such as cost, portability and relay burden influence the choice in VA rating of conditioning amplifiers. For a relay burden of 5VA at rated tap, the power required for ten per unit simulation would be 500VA, (5VA 10 * 10)since VA is proportional to the current squared. For 20 p.u., 2kVA would be required. For a 25VA relay burden, the VA requirements for 10 and 20 p.u. would be 2.5 kVA and 10 kVA respectively. Recent field simulation tests have simultaneously driven both primary and backup relaying. Selecting and applying conditioning amplifiers for these high burden loads requires investigation of relay burdens for all parallel voltage and series current inputs. Minimal specifications for power conditioning amplifiers: Amplifier performance must conform to the requirements of both the relay burden, and the simulation quality. Higher DFR sampling rates and typical, 50J..Ls per sample EMTP calculations place significant burdens on amplifier
- 20 -
design. Amplifiers formerly suitable for use with analog model power systems have less dynamic range and frequency response than required by today's simulators. Frequency response, output amplitude range and accuracy are well known criteria. They conform to specifications and standards for instrument transformers and with power system performance. As noted previously, peak voltampere rating must be considered in light of both per-unit simulation values and relay burden. Amplifier stability under widely changing relay burdens has been mentioned. Load power factors from unity to nearlyzero lead and zero laghavebeen observedand should be considered limits for amplifier specifications. Stability under rapidly changing signal conditions should also be consideredwhen specifying amplifiers. Large signalslewing rates of a few amperes per microsecond are commonly observed in EMTP simulations. A change in peak current from -7A pk (SA RMS), to 210A pk is based on 140 A pk at 20EU. plus 70A pk(50% offset). Toaccommodatea fully offsetcurrent signal,twice the peak current of a symmetrical signal is required. The slewing rate is approximately 0.5NJ..Ls (2l7NlOOJ..Ls). Lightning strikes represent thousands of amps per J..ls in primary current - requiring slewing rates of tens of amperes per J..lS secondary current. Voltage amplifiers
•
Input voltagerange + 10Vpeak (high impedance source) --
•
Output voltage range.::!:. 300V peak
•
Power output: 75 - 150 VA rms (150 - 300 VA peak)
•
Frequency response (small signal) de to 10 kHz + 3dB
•
Noise -80 dB of full scale
•
Transfer Gain Error @ nominal frequency.::!:. 1%
•
Common Mode rejection -60 dB of full scale
Current amplifiers
•
Input voltagerange.::!:. 10Vpeak (high impedance source)
•
Output current range.::!:. 160A peak
•
Power output 0.5 - 3kVA rms (1- 6kVApeak)
•
Frequency response (smallsignal)de to 10kHz + 3dB --
•
Noise -80 dB of full scale
•
Transfer Gain Error @ nominal frequency..±. 1%
•
Common Mode rejection -60 dB of full scale
5.0 TEST METHODS 5.1 General
Previoussectionsof thispaper havediscussed varioustest signals and test equipment that are available to the relay engineer for both integrity and application testing. This section is a general discussion of how that information may be used to plan effective integrity and application tests. It should be recognized that protective relays come in a wide variety of complexities, from simple single quantity overcurrent functions to complete distance protection schemes: therefore this paper is not a comprehensiveguide on relay testing. Section 6 of this paper includes examples of integrity and application tests and should help illustrate test protocols. Care is required in applicationof multiphase test signals to ensure accuracy of the test signal. For instance, the voltagesduring a phase-to-phase fault maybe simulatedby applying twophase to neutral phasors withthe proper phase relationship. In cases when the phase-to-phase voltage is low, very smallerrors in the phase relationshipof the twotest phasors will cause large errors in the magnitude of the test voltage. On a 120volt system, a phase fault with a source to fault impedance ratio of 10 to 1 would result in a faulted phase-to-phase voltageof 12volts. In a wyeconnected test systemthe phase-to-phase fault voltagewouldbe produced by two phasors with equal magnitude of 35 .157Vdisplaced by 9.83 0 • A total phase difference error of plus 10 (10.83 0 ) would result in a phase-to-phase to voltage of 12.6 V, a voltage error of + 5%. A phase angle error of 0.50 producing12.3 voltsisa 2.5%error. Duringsuch tests,direct application of a phase-to-phase signal is preferable if possible. For any given test, the relay engineer must decide what type of input signals and what type of test equipment are required. In some cases, compromise may be necessary because of practical considerations such as test equipment availability. In the choiceof input signals, the engineer must be guided by his experience and knowledge of both the power system and the relay.[18] In many cases,the test will be similar to those suggested by the relay manufacturer. For these tests, decisions are generally straightforward, depending primarily upon what equipment is available to meet the manufacturer's requirements. If, for example, the test involves verificationof the time curves on a time overcurrent relay, the engineer may conclude that steady-state current signals will produce the desired results. This choice will allow the use of several different types of test equipment. However, when a relay is being tested for a specific application, or as the result of some specific power system condition,the manufacturer's instructionbook method may
- 21 -
not be sufficient. Recent examples haveproven the need for simulation tests on overcurrent relays, one in an industrial protection circuitwith highlydistorted waveforms, another where an instantaneous element was driven by a saturated CI: For these tests the relay engineer must determine the appropriate inputs and test equipment. Since all relaysare applied on power systems, a power system simulator will generallyprovide proper current and voltageinputs for any type of test. However, power system simulation tests are usually expensive and time consuming. Between these two extremes are many other tests of varying degrees of complexity. The selection of the input signals and the type of test equipment is not directly dependent upon whether the test is an integrity test or an application test, but rather on how close to actual power system signals the test signal must be. Since many relays filter out all components except the fundamental from the measured signal, dynamic state simulation is a powerful method of application testing, but it has some important limitations.[10] •
The improper representation of transition from one state to another can sometimes cause incorrect relayoperation, especially in high speed relays.
•
The effectof non fundamental frequency components such as harmonics is not included. It is important to includethese effectsin manyapplication tests. Somerelayssuchasharmonicrestrained transformer differential relays require harmonics forproper operation. Manyrelayscanbe adversely affectedbyharmonicsand other non fundamental frequencysignals; suchsignalsshouldbe present in the test if they are present in the application. However, dynamic state simulation tests may be useful to explore the expanded characteristicof a cross polarized mho or compensator relay in a timely and complete fashion.[19,20]
5.2 Integrity Test Methods The basis for most integrity testing on protective relays will be the manufacturers' recommended methods for the relay. These tests are designed to verify that the relay is functioningproperly,and that the correctsettingshavebeen applied.Generally,these are steady-state tests,but dynamic state tests are sometimesused for checkingtimingand relay memory action.
electronic test sources as well. However, the user must ensure that the test connections and phasor relationshipsof the electronicinstrument duplicate those of the passivetest equipment. 5.3 Application Test Methods When current and voltage signals representative of actual power system conditions are desired, some form of power system modeling is required. If dynamic state simulation provides an adequate level of simulation, load and fault phasor values can be calculated using traditional principlesor a computer based fault study program. If more realistictransient simulation is required, it may be possibleto obtain actual disturbance data from a digital fault recorder. Otherwise computer modelling with programs similar to EMTP are required. Transient simulation waveforms may be stored in either analog or digital form. These may be played back to the relays at a later time. Analog power systems may also be used. The type of model is often determined by what means are available. In a playbacksimulator,the relaycannot interact directly with the power systemsimulation. When the relay issuesa trip output, since there are no circuit breakers in the pre-recorded simulationto respond, there is no interaction. (See Figure 13) In a real time system, relay outputs can cause a change in the power system model, such as opening the circuit breaker when the relay operates. This is a "closed loop" system as shown in the block diagram of Figures 11 and 12. The majority of real time simulators use an analog model of the powersystem. This maybe a lowpower modelwhich requires current and voltage amplifiers. In high power modelsthe current and voltageoutputs can drive the relays directly without interposing conditioning amplifiers (see Figure 11). Recent advances in digital power system simulation have led to the development of real time digital simulation. This allows the relay to modify the simulation parameters (see Fig. 12). 6.0 EXAMPLE
In some instancesthe user maydesireto includedifferent methods in the integrity test plan for a particular application. These additional tests may be steady state, dynamic state, or transient in nature.
6.1 Introduction A Canadian utility was consideringapplication of a new typeof multifunctiongenerator protection relayon some of its hydraulicgenerators. A test program was developed to determine whether the relay would be suitable for general application on such generators. The relay generally performed satisfactorily, but some operations that were not apparent from the manufacturer's specification were discovered.
In many instances the test equipment used in the manufacturers literature is passive (load boxes, variacs, and phase shifters). The same tests may be performed using
The relay is capable of operating in two modes: fundamental frequency measurement, and total rms measurement. In the fundamental mode, all harmonicsare
- 22 -
filtered out. In the total rms mode, the rms value of the signal, including the effect of all harmonics is measured. Tests were made with the relay in both modes. An overview of the test program is presented. Details of numbers of tests at different settings are not included. These data will vary depending on the relay being tested, and purpose of the tests. It will be noted that a largevariety of tests are included in the following program because the relay is being tested for general application.
3.
Voltage and Current Measuring Elements - Steady-State Performance
1.
For allmeasuringelements,checkthe accuracy at min., reference and max. settings. Also check reset ratio. Use pure 60 Hz sine wave signals. (SS)
2.
For all measuring elements, check the effect of harmonic contamination on the accuracy. Use 60 Hz sine wavewith 20% third harmonic at variousphase angles with respect to the fundamental. Run checksin fundamental and total rms modes. (SS Harm.)
3.
For allelements,checkthe effectof frequencyvariation on accuracy. Use a pure sine waveover a range of frequencies from 30 Hz to 120 Hz. Run checks in fundamental and total rms modes. (SS Freq.)
6.2 Sample Test Plan
TEST PROGRAM FOR A MULTIFUNCTION GENERATOR PROTECTION RELAY General
1. For internal timing functions, check accuracy at minimum, reference, and maximum settings. Check overshoot (inertia) to determine whether signalspersisting for marginally shorter duration than the timer setting can still cause undesirable outputs. 2.
In allcases,observeand report on contact chatter when test signal approaches operating point.
3.
Sixdifferent typesof testshavebeen identified; theyare abbreviated as follows: SS
TIME SS Harm.
Steady-state 60 Hz tests. Unless otherwise indicated, this test should be applied using the fundamental frequency measuringmethod of the relay.
2.
4.
For all timed elements, check the accuracy of timers. (TIME)
5.
Check the relationship between elements which control other elements internally. For example, the voltage controlled overcurrent function is not allowed to start until the voltageis depressedbelowa settable level. Does lowvoltageon anyone phase allowall overcurrent functionsto start, or does only the overcurrent element on the associated phase start?
Current Measuring Elements - Dynamic Performance
1.
Check the dynamic performance of the inverse time overcurrent element. Does it emulate the integrating actionof an electromechanical inductiondiscwhen the fault current changes? (Dynamic)
2.
Check operating time of instantaneous overcurrent elements. Check effect of presence or absence of prefault load. (Dynamic).
Steady-state timing tests. Steady-state tests with harmonics superimposed on the fundamental frequency signal.
Voltage and Current Measuring Elements - Transient Performance
SS Freq.
steady-state tests at variable frequencies.
Dynamic
Dynamicstate simulationsof faultconditions using 60 Hz test signals.
EMTP
Transient tests using data derived from EMTP or other similar simulator.
AID Converter Accuracy 1.
Checkeffectof frequencyvariationof input signalfrom 30 Hz to 120Hz. (SS Freq.)
Using the indication facility, check the accuracy of all voltage and current input functions (magnitude and angle). Use accurateexternalmeters to checksignalinputs. Checkallvoltageinputsat three points, (1.0~ ref. V and max. V). Check allcurrent inputsat three points, (0.lA, 5 A and max. available current, keepingin mind the rating of the relay input circuits). (SS) Check the effect of internal self-ealibration functions. Look for sanity checks, or possibility of accidental rescalingcausingprotection function problems.
1.
Checkthe effectof transient decomponent on accuracy of instantaneous overcurrent elements. Run checksin fundamental and total rms modes. (EMTP)
2.
Check the effect of ct saturation on time and instantaneousovercurrentelements. Run checksin fundamental and total rms modes. (EMTP)
3.
Check the response of peak overvoltage elements to distorted waveforms present during simulated ferroresonant conditions. (EMTP)
Frequency Elements
1.
Check accuracy of functions. Use 60 Hz sine wavesignal. Check trip output maintained when frequency raised to 120Hz. (SS Freq.)
2.
Check reset frequency.
3.
Check min. voltage at which frequency measurement can be made. (SS Freq.)
- 23 -
4.
5.
Check whether trip timing is affected by application of ramped frequency change or step frequency change. (2 Dynamic)
79. Reconnect Time Delay 1.
This function will not normally be used, but will be tested for completeness. Check timer accuracy. Check how function is initiated and blocked. (TIME)
Apply third harmonic distortion such that there is more than one zero crossing per fundamental frequency cycle, and check whether measuring accuracy is retained. (SS Harm.)
System Tests 1.
Test the operation of the device with a variety of simultaneous and near simultaneous faults.
6.
Check response to EMTP simulations of ferroresonant conditions. (EMTP)
2.
7.
Check accuracy of timers. (TIME)
Simulate application of overcurrent and coincident undervoltage conditions. Set test conditions so that the inverse time overcurrent element is expected to operate just before undervoltage, and vice versa. (Dynamic)
3.
Simulate application of single phase current such that inverse time overcurrent function will operate just before negative sequence, and vice versa. (Dynamic)
4.
Simulate application of overfrequency and overvoltage conditions such that RMS overvoltage, Peak overvoltage, and overfrequency functions all operate near simultaneously. (Dynamic)
5.
Simulate application of near simultaneous neutral overcurrent and neutral overvoltage conditions such that the rms neutral overvoltage and inverse time neutral overcurrent functions operate near simultaneously. (Dynamic)
Negative Sequence Overcurrent Element. 1.
Check whether accuracy is affected by use of single phase current (12 = (1/3)applied current) or balanced three phase current rotating with negative phase sequence.
2.
Check accuracy of pickup. (SS)
3.
Check definite maximum time to trip. (TIME)
4.
Check time current characteristic. (SS)
5.
Check reset time by measuring operating time with various intervals between successive fault applications. (Dynamic)
6.
Check integrating characteristic - similar to test 1 of dynamic tests on time overcurrent element. (Dynamic)
7.
Check frequency response of pickup of function using a single phase test signal. Check effect on timing of off nominal frequency current. (SS Freq.)
8.
Check effect on pickup and timing of third harmonic current present in a single phase test signal. (SS Harm.)
9. Check whether function responds to balanced three phase current at 120 Hz rotating with positive phase sequence. (SS Freq.)
Directional Power Elements 1.
Check accuracy of reverse element using one, two and three wattmeter methods with various power factors. Use 60 Hz sine wave test signals at nominal voltage.
7.0 CONCLUSION Relay performance tests address two questions. Is the relay operating as it was designed? Is the relay being applied properly? Integrity tests are intended to answer the first, and application tests the second. In the past, users have concentrated on integrity tests using relatively simple equipment in the field. Application tests required bulky and expensive equipment which was usually only available in relay manufacturers' plants, or in research laboratories. Historically, field test equipment was only suitable for steady-state tests and limited dynamic state simulations. Manufacturers and users developed comprehensive test plans to ensure the integrity of relays using passive test equipment. Integrity tests were also designed to be performed in the field using simple equipment. With common applications, the manufacturers application tests done during relay development will usually suffice to ensure correct operation under normal circumstances.
2.
Check effect on accuracy of max. and min. voltages for operation of reverse power element at unity power factor. (88)
5.
Check accuracy of timers. (TIME)
6.
With high speed setting, and ref. pickup setting, check effect on operating time of various amounts of power over setting. (Dynamic)
Development of electronic test equipment in the 1970's led to faster, more accurate steady-state tests and dynamic state tests. These are regularly used for application tests in the field. These tests allow users to answer some application questions which might not be addressed by the manufacturers' instruction manual. They also allow comparison of different designs for simple applications and some troubleshooting. While dynamic state tests are useful for a wide variety of applications tests, they have limitations which can only be overcome by full transient tests.
7.
Check effect of frequency variation on pickup setting accuracy. (SS Freq.)
Advances in relay test equipment have led to the availability of relatively economic and portable equipment
(88)
- 24 -
capable of doing full transient simulation tests. Analog simulators are the traditional means of transient testing and they are still widely used in laboratories and factories. Play back digital simulators are now being widely used in the field and workshop for "open loop" transient tests. Real time digital system simulators will allow relay users the ability to do complete "closed loop" application tests in their own premises. When the power system is properly modeled, and test equipment is properly used, virtually any proposed application can be comprehensively studied by the relay user or manufacturer. It is important to completely understand the signals that will be presented to the relay in its intended application, when devising a test plan to examine relay performance.
REFERENCES 1.
A Survey of Relay Test Practices, 1991 Results, IEEE Power System Relaying Committee Report, IEEE/ PES 1994 Winter Meeting, NY, NY, 94WM137-0 PWRD
9.
ANSI/IEEE C37.90-1989 "Standard for Relays and Relay Systems Associated With Electrical Power Apparatus"
10. C. F: Henville and J. A. Jodice "Discover Relay Design and Application Problems with Pseudo Transient Tests", IEEE Transactions on Power Delivery, October 1991,p.p.1438-1443 11. G. E. Alexander, E J. Lerley, and R. Ryan, "Comparative Testing Using Digital Simulation and an Analog Model Power System", a paper presented at the Doble ProTesT Users Group Conference, 1990 12. M. Kezunovic, et al "DYNA TEST Simulator for Relay Testing, Part II: Performance Evaluation" IEEE PES Winter Meeting, February 1991 13. LA. Thiem, "The Effect of Waveform Distortion on Overcurrent Relay Calibration", 49th Annual International Conference Doble Clients, Boston, MA, 1982 14. "General Electric Model Power System", GE publication GER-3225
2.
IEEE/PSRC Working Group, "Sine-wave Distortions in Power Systems and the Impact of Protective Relaying", IEEE Report 84TH 0155- 6PWR
15. EG. McLaren et al. '~Real Time Digital Simulator for Testing Relays.", IEEE Trans. PWRD, Vol. 7, No.1, January 1992.
3.
CIGRE SC 34 "Guide on Evaluation of Characteristics and Performance of Power System Protection Relays and Protective Systems"
16. M. Kezunovic et aI, "Transients Computation for Relay Testing in Real Time", IEEE Trans. Pwrd., Vol. 9, No. 3, pp. 1298-1307, July 1994
4.
ANSI/IEEE C37.111, Common Format for Transient Data Exchange (COMTRADE)
5.
IEEE Std. 519: IEEE Recommended Practices & Requirernents for Harmonic Control in Electric Power
17. J. Esztergalyos, J. Nordstrom, L Short, K. Martin, "Digital Model Power System", IEEE Computer Applications in Power, Vol. 3, No.3, July, 1990, p. 19
Systems 6.
C. F Henville "Type Testing of Distance Relays" 14th Annual Western Protective Relay Conference, Spokane, WA October 1987
7.
ANSI C37.90.1-1989 "Standard Surge Withstand Capability (SWC) Tests for Protective Relays and Relay Systems"
8.
ANSI/IEEE C37.90.2-1995 "Trial-Use Standard Withstand Capability of Relay Systems to Radiated Electromagnetic Interference from Transceivers"
18. K. H. Engelhardt, "Dynamic Performance Testing of Mho Relay Memory Action", 1982 Western Protective Relay Conference, Spokane, WA 19. W O. Kennedy, B. 1. Gruel, C. H. Shih, L. Yee, "Five Years Experience with a New Method of Field Testing Cross and Quadrature Polarized Mho Distance Relays", IEEE/PES Winter Meeting February 1987 20. R. Ryan, 'Automatic Testing and Plotting of Protective Relay Polarized Mho Characteristics", 59th Annual International Conference of Doble Clients, Boston, MA, 1991
- 25 -
