MODERN RELAY TESTING Javier Palomino* and Eduardo Marchesi.

EuroSMC, Spain

ABSTRACT
The extraordinary development of base technologies like computing and communications, the consolidation of standards and the outbreak of new approaches and methods in the field of protective & control process automation, are pushing test equipment designers and manufacturers to maintain a permanent attitude towards innovation, mainly in terms of accuracy, simplicity and efficiency. The present scenario of electrical protections testing is described here with emphasis to key technological and methodological aspects, as well as realistic answers to the most outstanding questions about bringing our testing procedures and tools up to date. Important factors from staffs experience and training to test technology to tools design are depicted to help the reader at understanding and facing the need for innovation and adaptation. Finally, the discussion materializes in a proposal of proper selection of modern test equipment.

RELAYING TECHNOLOGY
Electromechanical relays can still be found in many switchgear installations all over the world. Not surprisingly, these devices are commonplace especially in countries that pioneered the distribution of electrical power, i.e. technologically advanced countries. The reasons why they are still there are ELECTRICAL MAGNITUDES INPUT diverse, the most obvious being the big cost of replacement. However, brand new units have already vanished out FUNCTION FUNCTION FUNCTION from the mainstream control & 1 2 3 automation market, so failed electromechanical relays must be repaired or replaced by a functionally equivalent SINGLE OR MULTIPLE CONTACT (BINARY) OUPUT refurbished unit. On the leading edge of protective deployment, microcomputers found their Fig. 1: Functional diagram of a traditional relay place in every piece of equipment, including relays, more than one decade ago. The elementary protective elements that were represented by formulae to describe the relays operation in the past are now actual

algorithms coded inside the non-volatile memory of an IED1. This means that the operation of a modern relay is greatly determined by its programming, rather than its physical construction. As a result of this revolution, the socalled digital relays are no longer relays, as understood in the traditional sense. Instead, these sophisticated, compact and versatile microprocessor-based devices are replacing complete protective and automation systems. They feature the electronics, control and computing resources needed to implement a full scheme rather than a few more or less coupled protection functions. The advantages of using digital relays greatly compensate the cost of replacement in the medium term and, naturally, determine the obvious choice for any new installation. Reliability, versatility, accuracy and manageability are just a few and, for the sake of this discussion, economy in testing and maintenance ultimately define digital relays as the best pieces of technology at the core of any modern protective system.

ELECTRICAL MAGNITUDES INPUT D/A CONVERSION MEASUREMENT & RECORDING

BINARY INPUT

TRIP LOGIC

SETTINGS FORMULAE FUNCTIONAL PROGRAMMING

SINGLE OR MULTIPLE CONTACT (BINARY) OUPUT

Fig. 2: Functional diagram of a digital relay

However, electromechanical and solid state relays have long time conditioned the way in which every elementary protective function is accomplished and implemented, posing a heritage that is still evident in the name we continue to give to (and the way we still handle) substantial relay dynamics like reset time and maximum torque. The resulting transitional stage will persist in old power grids, because the tight dependencies between the schemes design and the working characteristics of traditional relays make direct, one-to-one replacement an almost impossible task.

TESTING
In its most comprehensive meaning, relay testing involves complete, realistic fault simulation. Simulators combine powerful computer and signal generation resources for accurate real-time synthetic production of, and data acquisition from, an electrical fault in order to fully check an IED for correct operation. Though the use of simulators is mandatory in some of the implementation and design stages, most testing is carried out using simpler equipment, typically portable devices, and sometimes fixed test appliances that are installed in automation racks and panels along with the relays themselves.
1

Table I: testing traditional and digital relays Traditional relay Difficulty of type testing Difficulty of commissioning Frequency of routine testing Frequency of specific testing Computer-based test functions required Required operator skills Medium Medium High High Recommended Medium Digital relay High High Low Medium Yes

IED: Intelligent Electronic Device

2
High

IEDs also differ from traditional electromechanical relays in the way they can and should- be tested, and different test levels and goals must be distinguished, too, according to different test scenarios: type (application) testing, commissioning (startup), specific (diagnostic, corrective) testing and routine testing. Table I illustrates the main differences between traditional and digital relays in the context of testing activities, required tools and necessary skills. The persistence of the aforementioned ancient terms is just one of the signs revealing the urgent need to incorporate the new technology to our general thinking about protections in a greater extent. And this lack of adaptation becomes especially evident when we talk about relay testing. The reasons for this persistence are numerous: slow update of the training personnel is possibly the biggest, but this belongs to a different discussion. We must stop thinking of relay testing in the traditional way. Instead, scheme testing is the appropriate term. When IEDs are used as the central components of protective systems, the engineering process is centered on their working parameters and their interaction with other elements in the system or, ultimately, with other interconnected systems in the automation plan. As a consequence, testing must be understood in a different way now. Pickup quantities, operation time and trip restraints are no longer topics that one can include in the test task in an individual fashion for two reasons: 1) they are difficult to visualize and manage individually in an IED, and 2) testing them will not guarantee the expected operation of a scheme as a whole. IEDs are designed using computers, are setup using computers, and must be tested with the help of computers. Type Testing Type Testing is called before a new relay type enters the protection system, and basically consists of assessing the intended protective application(s) and the relays capacity of interaction with other components in a scheme. The primary goal of type testing is validating a
Table II. Fault Simulator Characteristics (subset of IEEE recommendation) VOLTAGE AMPLIFIERS Peak Output Voltage Peak Output Current Continuous Output Power Frequency Response Accuracy Power Bandwidth Slew Rate Output Impedance (0 3 kHz) Worst Case Load Impedance Number of digital inputs Digital Input specs Number of digital outputs Digital Output rating Number of analogue outputs > 10 V/s < 0.5 70 (S/C protected) 300 V 1A 150 VA 0 10 kHz < 1% (0 1 kHz), < 3% (1 3 kHz), < 5% (above 3 kHz) 0 10 kHz > 2.5 V/s > 250 5 k (stable in open-circuit) CURRENT AMPLIFIERS 50 V 100 A 2,500 VA

16 (per terminal) Optically coupled, 50 100 V dc. / 10 mA 16 (per terminal) Optically coupled, 50 150 V dc. / 200 mA 8 (per terminal)

products suitability to a given protective function in the context of a particular scheme. Closedloop test procedures typically associated to type testing may involve the use of complex, 3

specific methodology and highly specialized equipment like fault simulators, as mentioned before. Simulators are dedicated, expensive equipment, usually rack-mounted and engineered to stress the design characteristics of protective IEDs in a laboratory environment, to where IEDs are brought for testing. A brief look at the IEEEs recommendation for fault simulators characteristics (see Table II) will give a good idea of what closed-loop fault simulation is all about. Type test execution goes well beyond just applying a discrete series of static CT and VT secondary magnitudes to the protective terminals inputs. The need for high power values, awesome I/O capabilities, complex modeling, intensive parametric calculations, real-time generation of, and data acquisition from line-level faults and logical events precludes the use of commonplace portable test sets for this job. Commissioning Commissioning tests are conducted on the installed equipment right before the electrical facility or any of its sections comes into service for the first time or after carrying out significant changes. This definition leads to two important considerations that determine the characteristics of needed methods and tools Portable equipment is required for on-site testing IEDs are tested as an integral part of a multi-function protective assembly that includes interconnection cables, auxiliary devices and other relays, among other components.

The following is tested during commissioning: 1) The IED as such a device (measurement accuracy of analogue inputs, proper initiation and reset of I/O features, relays settings and overall observed operation compared to planned response, etc.), when it is assumed to be in its final location and setup. Computer-originated mistakes in settings (like 32 instead of 3.2), faulty trip logic and other last-minute errors are targeted and unveiled in this phase of testing. 2) The IED as an integral part of a protective scheme with its surrounding auxiliary switchgear, cabling, signaling and control elements, following the definition of semiclosed-loop testing. This testing is fairly conclusive because it is carried out at the cell level rather than at the device level. Numerous I/O features are required in the test equipment, as well as sufficient power, accuracy and stability characteristics because electrical performance will be affected by the added load upon application of test values that have been previously calculated for solid assessment of the IEDs digital and analogue response. 3) Communications (end-to-end testing). A pre-recorded fault is simultaneously played back into protective cells at both ends of the line. Two separate test sets and an accurate time reference (typically GPS clock signals) are required. The test sets must be able to initiate the fault play back automatically at the programmed time. All the components in charge of communicating both ends, along with the resulting signaling and coordination response are tested against specified values. Commissioning tests usually reveal faulty cabling, wrong relays settings, polarity-related issues or a mixture of the three, making isolation of failure causes a tedious and lengthy task. Technical experience and appropriate tools are essential to safely reduce protective commissioning duration to a minimum. Specific testing Specific or diagnosing, corrective testing is conducted after a protective component has failed to operate as expected during a real fault. Accurate fault playback is one of the most efficient techniques to isolate the causes of repetitive relay failure in these cases. Digitized fault information recorded in a quantization file at a high sample rate is streamed into the test equipment to be played back on the suspicious relays

analogue and digital I/Os. Relays recognition of the fault parameters and its consequent response in terms of operation time and I/O actuation are then compared against expected operation characteristic for a fast, conclusive location and correction of the erroneous setting(s). Routine testing The self-test feature available in many digital relays, combined with communications, remote management and reporting capabilities, dramatically simplifies and reduces the frequency of routine testing by a factor of 100 when compared to a traditional relay. This is true to the surprising conclusion that the more you extend your routine test intervals, the less subject to failure your digital relays will be. Built-in self-test routines check the IEDs hardware integrity at fixed intervals by examining its core functional components: power supply, microprocessor, analogue-to-digital conversion elements and information (settings and recordings) storage. When any of these critical parts fail to pass the self-test program, an alarm message will be issued and, where required, a preventive contact operation will be initiated that triggers other relays action or the tripping of a circuit breaker. This great technological step has practically removed the routine testing of IEDs from the planned maintenance boards. Instead, shutdown times and/or other human-assisted maintenance activities are used to perform very quick (usually just checking the correct measurement of voltage, current and phase angle) tests in selected, mission-critical relays, using basic, straightforward steady state functions in the test equipment.

NEW INTEROPERABILITY STANDARDS

The enormous flexibility of microprocessors has derived into a huge variety of products from IED vendors. Whereas this is an obvious benefit for users, integrators are having a bad time at providing solutions that can be understood, used, expanded and maintained safely and efficiently at a reasonable cost, due to the wide disparity of protocols, methods and even technical terminology used by devices that offer similar functionality but come from different manufacturers. The new IEC 61850 is now reaching a maturing stage to alleviate this situation. The technical foundation of their specifications is, undoubtedly, the digital nature of IEDs, as well as the availability of solid, proven technologies in the field of processing and data communications, including Ethernet and XML. Separating form and function, or process and communications, has been the key step towards successfully have products from different makes and vendors working together. The evolving changes imposed by IEC 61850 and related sub-standards are not only affecting the design and features in the new IEDs, but also the characteristics of the equipment and methodology needed to test them.

DESIGN GUIDELINES OF TEST EQUIPMENT

Growing specialization of the technical staff in charge of testing, added to progressive sophistication of protective devices, demand the use of microprocessors and human interfaces that streamline the process from the lowest element-by-element test capabilities to the highest possible degree of automation. The tools must be able to adapt themselves to different test situations and requirements, including unexpected device behavior, less-than-optimal test conditions or limited operators skills. Physical characteristics and capacity for adequate fault simulation are evident aspects of available test equipment, which should be viewed as an interface between the method and the device under test.

As a result of this consideration, the methodology sets the rules for good design, which in terms of technology means good software design. The test equipment must ultimately be evaluated in terms of human interface, i.e. its ability to match the users capacity and the test scenario to the biggest possible degree. When erroneous behavior results from commissioning tests, fast, accurate targeting and isolation of the failing component(s) is needed. The available tools must then be able to provide the operator with elementary functions and resources that let her/him suspend momentarily the automatic procedure and continue manually without disrupting the whole process. To achieve this, the test equipment must be designed with this casuistry in mind. Minor changes in connections, as well as immediate use of basic test and measurement features must be possible until the problem is located and corrected, and any pending automatic steps can be resumed if necessary. In fewer words, the goal of sophistication is to simplify the job, making it friendly, safe and easy to learn. These considerations lead to the following key guidelines when designing latest-generation relay test equipment: 1) Applicability. Fault simulation must be accomplished without modifying the relays settings. The output capacity in terms of electrical power, dynamic range, accuracy and operation logic processing must be able to face any IED installed in the system.

SART AUTOMATION

ROUTINE 1 ROUTINE 1 ROUTINE 2 ROUTINE 2 ...... TEST INTERRUPTIO N

CHANGE CONTROL LEVEL

MANUAL MODE MANUAL MODE

PROBLEM PROBLEM ISOLATION AND ISOLATION AND CORRECTION CORRECTION

RESUME AUTOMATION

ROUTINE n ROUTINE n

2) Portability. Equipments size, weight END OF TEST and shape are a main concern. Designers must provide the best possible compromise between ergonomics and Fig. 3: The Adaptative Control Paradigm performance. A self-contained design is easier to move and is less prone to incidents with connections or misplaced peripherals. 3) Integral control provisions. The test equipment must be fully operable as is, without mandatory dependencies to external components that have not been designed to work under heavy-duty conditions. Basic control must be obvious for medium-trained operators, and graphical representation should be

extensively used throughout the built-in test intelligence, so that the user can concentrate in the task rather than in the equipments operation. 4) Dependability. The construction must be robust for in-field operation and must withstand heavy duty transportation and environmental conditions, especially high voltage influence and extreme temperatures. Key components like power supplies and output amplifiers must be built as plug-in modules that the user can replace easily.

ADAPTIVE APPROACH

Lets discuss here what could be a practical implementation of the above statements and conclusions. Physical characteristics of the test equipment must be comprehensive in terms of fault simulation, measurement and logic evaluation. The built-in computing platform will provide human interaction, simulation control, data management and communications. The human interface, i.e. the part of the software that determines the testing experience, is intuitive at any stage of testing, and adapts itself to different levels of control (see fig. 4) in order to maintain a perfect compromise between automation and flexibility to solve unexpected situations. Once the test equipment is provided with integral computing power, an enormous range of new capabilities and innovative approaches comes at hand. The electrical and logical features, including measurement functions, are the physical layer in this design. A second layer contains peripheral functionality like communications, storage and external interfaces, as well as general control of the hardware platform and interface to the computing resources. The computing platform implements a hardware abstraction that enumerates and determines the working rules of hardware resources like output channels and chronometers, viewing these as test services and providing a consistent API that allows the easy separation of the hardware functions and the control software. The visible computing functions live here. A basic level human interface displays an organized representation of the test resources which the user quickly recognizes and associates to the tested function. These resources include direct control of available voltage and current channels, adjustment of signal characteristics like frequency and phase angle, use

Quantities Pickup Definition I/O Activation Timing ASSISTED CONTROL Ramp definition Multi-state fault definition Test execution Events recording Data acquisition AUTOMATIC TESTING IED type & settings selection Element selection Routine execution Storage & Reporting

SPEED 7

DIRECT CONTROL

Fig. 4: Test Control Levels

FLEXIBILITY

THE

PROCEDURE AS PART OF THE TEST EQUIPMENT: AN

of timing resources and access to storage and retrieval of frequently used setups and test results. A set of pre-programmed tools that combine these elementary resources allow for instant testing of protective functions in one level, and for test automation and reporting in a superior level. The goal is completing the test in a simple three-step procedure: 1) Relay type selection and connection 2) Test procedure selection and execution 3) Results storage, management and reporting Any of these top-level steps can be broken down at any time into lower-level or elementary functions like, for example, definition of a new protective function or programming and applying a voltage ramp.

FINAL CONCLUSIONS
IEDs impose an obvious change in the way we look at protective device testing in an interoperable scenario. This change impacts our traditional thinking and practices in many aspects Scheme design, control features, practical implementation, operation and testing Design of suitable test equipment Updated training

The approach and ideas proposed in this document try to synthesize these changes in a brief overview of the impact of latest protective technologies present in modern electrical power utilities.

BIBLIOGRAPHY
Working Group I10 of the Relaying Practices Subcommittee. A Survey of Relay Test Practices 1991 Results. IEEE Power System Relaying Committee Report. IEEE Transactions on Power Delivery. Vol. 9, No. 3, July 1994 John J. Kumm, Mark S. Weber, E. O. Schweitzer, III, Daquing How. Philosophies for Testing Protective Relays. 48th Annual Georgia Tech Protective Relaying Conference. May 4-6, 1994. Working Group F-8 of the Relay Input Sources Subcommittee of the IEEE Power System Relaying Committee. Digital Simulator Performance Requirements for Relay Testing. IEEE Transactions on Power Delivery, Vol. 13, No. 1, January 1998. M. S. Sachdev, T. S. Sidhu, P.G. McLaren. Issues and Opportunities for Testing Numerical Relays. 2000 IEEE END OF THE DOCUMENT