AVO New Zealand

This e-mail address is being protected from spambots. You need JavaScript enabled to view it
Product Manager

Click-On-Fault (COF) testing, a Megger innovation that was first introduced in 1995, is universally recognised as being of enormous benefit for protection engineers involved with testing distance relays, as COF capabilities not only make testing easier, but are also big time savers. COF testing has come a long way since those early days and, with this in mind, let us examine some of the latest and most comprehensive of COF implementations has to offer.

Defining operating characteristics
With the best of current COF implementations, users are able to define the operating characteristics of distance (impedance) relays simply by selecting from either a list of generic characteristics or a library of relayspecific characteristics. Generic characteristics range from traditional MHO, to Half MHO, IEEE MHO (with blinders), IEEE Quadrilateral and European Quadrilateral.

In addition to the generic characteristics, there is a library of characteristics for specific relays manufactured by ABB, AREVA, GE, Siemens and SEL. The user is presented with defined nomenclature for each of the relay settings and simply enters the known values. The software then creates a relay operating characteristic ready for testing. Other optional inputs include setting values for load encroachment (forward and reverse), and up to five user-defined protection zones. Once an operating characteristic has been defined, it can be saved as a template for reuse. Defining fault simulations and test points After defining an operating characteristic, users can then select phase-earth, phasephase or three-phase fault simulations. To make the test setup even easier and faster, the user can choose Quick Test for the selected zone to be tested, for all defined zones to be tested, or for all defined zones and all defined faults to be tested.

Other powerful test options include testing in accordance with IEC 60255, multiple test points through the origin, and shot test points inside and outside the operating characteristic. The user is free to define the test search lines or test points using the instrument’s touch screen, or with a mouse when working the PC version of the software.

Up to ten search lines or test points may be defined for each phase, each type of fault, and each zone. The relay operation may be determined using linear ramp, pulse ramp, or pulse ramp binary search methods. When using the shot test method, test points are easily defined by touching the screen or using the mouse to set points inside and outside the tolerance bands of the operating characteristic of the relay under test.

The pulse ramp binary search provides the capability to determine an unknown operating characteristic. With up to ten test points per fault simulation, even difficult operating characteristics like load encroachment are easily discovered.

Other user-selectable test-defining inputs include constant test voltage, constant test current and constant source impedance models. For phase-earth fault simulations, users also have the option of entering K_N or Z₀Z₁ compensation factors. For testing accelerated tripping characteristics, the user may also define the pre-fault condition including load currents and angles. When selecting pulse ramp, the pre-fault condition will be supplied to the relay for the specified time prior to returning to the next incremental fault values.

One-touch multiple fault simulations for multiple zones
After defining and setting multiple zones and tests, the user simply presses the play button and the software performs the tests in an orderly progression while automatically recording the results, including pass/fail data. If operating times are included for each zone, the tests will automatically record the operating time for each fault for each zone and display the results.

Viewing tests in real time
The right side displays the moving test vector, indicating a moving test point showing the impedance value and angle in real time. On the left hand side of the display are the test vectors showing the phase-to-phase test voltages and currents being applied to the relay under test. The user also has the option of viewing the positive, negative and zero sequence vectors instead of the voltage and current vectors. When the test is completed, all test results for each fault simulation and each zone may be selected for immediate viewing and/or saved to the internal results file for subsequent downloading.

Priceless
The COF functionality described here greatly simplifies the task of testing distance relays and is undoubtedly of great value. In fact, software that provides this functionality is currently being offered by at least one supplier at a price in excess of $9,000 per licence, with a separate licence required for each test system. Megger has taken a rather more customer-friendly approach by providing software that includes all of the features described at no additional cost with all of its new SMRT relay test sets. The benefits of COF technology are now, therefore, affordable and readily accessible to all relay test professionals.

High pot versus insulation - part 3

This is the third and last instalment of articles looking at the differences between insulation testing and hi-pot testing. The first two instalments, which are still available on line, looked at insulation testing and introduced industrial hi-pot testing. This instalment provides further information about industrial hi-pot testers and also explains how cable hipots differ, as well discussing some of their applications.

Industrial high-pots have two primary controls, one to set voltage and the other for trip current. Current levels are based on accepted values relevant to the reaction of the human body. Not all studies, and not all experts agree, but generally 0.5 mA is considered perception level, 5 mA provides a “shock”, and 12 mA is approaching “let-go” current, where the victim can no longer release the source. Therefore, high-pots are typically adjustable across a similar range. If leakage current exceeds the set level, the high-pot must both visibly and audibly “trip”. Test time is commonly one minute, but in the interest of practicality, a one-second test at appropriately higher voltage may also be allowed.

Finally, the specifying agencies want no risk of human error. Unlike an insulation tester, high-pots are often operated by minimally trained personnel on the factory floor. The instrument should do all the work. The operator just changes the test item and pushes buttons. Often the supervisor will set the controls at the start of the run. From the viewpoint of the agencies, the idea is to have uniform testing, with the tester either passing or failing the item. Human judgment, which is absolutely critical in insulation testing, is taboo in industrial high-potting.

Industrial high-pots can be used in troubleshooting applications where their voltage is sufficient, and some industries like aeronautical and electro-motive use them to break down marginal equipment so that it doesn’t go into service. This usage bridges the gap to high-voltage dielectric test sets that are used for commissioning, maintenance testing and troubleshooting of transmission cable and transformers, motors, bushings and other equipment that operates at elevated voltages. Typical test voltages are values like 70 to 160 kV. The design of such models is a careful balance of breakdown power, measurement reliability, safety and practicality.

Test voltages are multiples of common transmission voltages in order to fully accommodate stress tests. They are used for both proof and maintenance tests. Proof testing is similar to factory high-potting. It looks for breakdown under stress, and is used as an acceptance test upon commissioning and when repaired equipment is returned to service, as in a cable splice. Maintenance testing looks for a value that makes a determination as to equipment’s condition and future service. Current is the preferred unit of measurement rather than conversion to the reciprocal resistance because it is directly observed and easier to work with. Testers commonly operate at around 5 mA, although much greater currents are available in some models. Although 5 mA approximates the output of a megohmmeter, the much higher voltages produce a quantum leap in power.

In keeping the testers as safe as possible, 5 mA is about the lower limit that will still produce an effective high-pot. For added safety, the power supply is typically a separate module from the control unit, so that the control can be limited to line power only while the high voltage is kept a safe working distance away. At these voltages, arcing becomes a sinister threat, as personnel may not anticipate danger from an airborne arc. Industry-standard safe working distances must be maintained, and keeping the control module separate makes that possible.

Portability is served by having air-insulated power supplies that keep the weight down, while a dc test aids practicality by providing the necessary power in smaller, lighter units than could be accommodated with ac.
High-voltage dielectric withstand testers therefore complete the trio by combining critical aspects of both the insulation tester and the industrial high-pot. Depending on the operator and the application, they can be used as measuring devices for maintenance regimes or as breakdown sources to remove equipment from the field.

Back to Top

The dielectric breakdown voltage test is commonly performed on new oil before it is used to fill equipment such as transformers, switches and circuit breakers. It is also used for the acceptance of new oil deliveries. Sometimes, however, Megger’s Technical Support Group receives questions about breakdown test results for new oil that seem to be anomalous. Here are the answers to some of those queries.

Q: I’ve just carried out acceptance tests on some brand new oil, and it’s failed the breakdown voltage test. I find it hard to accept that the new oil is defective, so does this mean that there’s a problem with my oil test set?
A: Not necessarily. Of course your test set could be faulty, but that’s actually one of the least likely reasons for new oil failing a breakdown test. It’s much more likely that the problem is caused by contamination.

Q: How can it be that new oil is contaminated?
A: It may not be – it’s perfectly possible that only the sample you tested is contaminated. It’s essential to ensure, for example, that the test vessel has been thoroughly cleaned prior to use, and that the sample is taken under suitable conditions. Getting a good sample when the relative humidity is higher than 50% is very difficult, and when the ambient temperature is very high, contamination from perspiration can be a problem.

Q: The sample was taken very carefully in appropriate conditions, and the test equipment had been properly cleaned. However, it still failed the dielectric test. Are there any other possibilities, other then instrument error?
A: Yes. If the possibility of a contaminated sample has been eliminated, it is necessary to consider whether the oil itself may be contaminated, even though it is new and was almost certainly in good condition when it left the supplier’s factory. Often oil is delivered in drums that are stacked on site in readiness to fill a transformer. In hot weather, the drums heat up during the day and as a result, the pressure inside them rises. At night, the drums cool and the pressure falls. This constant temperature and pressure cycling is sometimes too much for the seals, and they effectively begin to “breathe”, drawing moisture into the drums. It takes only a small amount of water to enter the drums in this way for the oil to fail the breakdown test.

Dynamic frequency response (DFR) testing of power transformers is increasingly acknowledged as the most dependable and most convenient way of determining the condition of the insulation in power transformers. Crucial benefits of DFR testing are that the tests can be carried out at any temperature, and that it is able to evaluate the condition of the paper insulation as well as the oil and also estimate the temperature dependence of the dissipation factor, all in one single unit.

Because of this, DFR testing gives a much more comprehensive and dependable indication of the transformer’s overall health than other test techniques. In addition, with appropriate equipment, tests take only a relatively short time to complete.

Useful as the DFR testing technique undoubtedly is, however, it can sometimes be difficult to use successfully in areas where there is a high level of electrical noise, such as that commonly encountered in high-voltage substations. This is hardly surprising, as standard instruments carry out DFR tests with an applied voltage in the region of 200 V. This is perfectly adequate in normal environments, but in really noisy environments, even with good filtering, it is impossible to pick out the required signals for analysis with this level of applied voltage.

A recent development has, however, provided a complete and effective solution to this problem. It sounds deceptively simple – an amplifier to increase the output voltage of a DFR test set by a factor of ten to 2 kV. Unsurprisingly, actually producing a suitable amplifier that will operate accurately and reliably is technically challenging, but this has now been achieved.

The increased output voltage, combined with advanced digital filtering algorithms greatly reduces the influence of electromagnetic interference, allowing accurate and dependable results to be obtained under conditions where this would previously have been difficult or even impossible.

The best DFR test sets are also surprisingly fast in operation. With efficient signal acquisition technology and two separate current measurement channels, these instruments provide reliable true ac measurements without the need for dc to ac conversion. This means for example, that two separate measurements down to 1 mHz on a three winding transformer can be completed in a little over half an hour, whereas older instruments of more conventional design typically take around two hours.

It is also worth noting that the use of a highvoltage amplifier with a suitable DFR test set extends the capabilities of the instrument to include 50/60 Hz capacitance and tan delta measurements. The maximum capacitance of test object that the instrument can handle will depend on the current output capability of the amplifier but, with an amplifier capable of supplying 50 mA, for example, it will be possible to use the instrument with test objects of up to 80 nF at 50 Hz and around 67 nF at 60 Hz.

An excellent example of a highvoltage amplifier and DFR test set combination is Megger’s VAX020 amplifier used in conjunction with the company’s popular and efficient IDAX300 tester. This combination reliably and cost-effectively extends the invaluable technique of DFR testing into areas where its application was formerly difficult or impossible.

Christchurch, New Zealand, has suffered three major earthquakes in the past twelve months. The first, on September 4th 2010, was magnitude 7.1 on the Richter scale. The second, on 22nd February 2011, was less powerful at magnitude 6.3, but it caused more devastation, as it was right under the central city. It killed 180 people, destroyed almost all of the central business district as well as many thousands of homes, mainly in the eastern suburbs of the city. The third quake in June also registered 6.3 and destroyed yet more buildings, homes and infrastructure but thankfully there were no more deaths.

Peter Hockley’s neigbhour’ s house before it had to be demolished.

Christchurch has been left with a locked down central business district where some 70 demolition crews continue to knock down unsafe buildings. Demolition has just started on the city’s tallest building, the 26-storey Hotel Grand Chancellor. This building is leaning at a large angle and is a danger to surrounding structures. Many iconic Christchurch buildings have been lost forever; the most well known of these being the Christchurch Cathedral, which has lost its spire and the whole of its front entrance as well as suffering serious internal damage.

Peter Hockley General Manager, AVO New Zealand At the time of writing, the Rugby World Cup tournament had just reached its dramatic conclusion in New Zealand and games scheduled to be played in Christchurch had to be moved elsewhere because the AMI Stadium, which had only just been rebuilt, suffered immense damage to the stands and playing area. In addition, with most central city hotels either damaged or closed, there was not enough accommodation to house visitors for such a large event.

In many areas; roadways, sewers, water and electricity services have been severely damaged. Some suburban areas have been abandoned and it’s likely that more will have to be abandoned in the future. The electricity network has suffered badly. Earth movement stretched some underground cables by up to a meter and almost instantly caused more faults than would normally be seen in a decade. The February quake caused some of the most violent urban ground shaking ever recorded anywhere in the world, severely damaging substations, with one sinking two meters into the ground.

Trevor Lord, CEO of AVO New Zealand, handing over a PFL portable cable fault locator

Half of the city’s 66 kV underground network (30 km) was damaged beyond repair and two temporary 66 kV overhead lines were built to get power to the eastern suburbs. There is an ongoing programme of assessing, repairing and rebuilding of substations. Of the 2,200 km of 11 kV underground network, 330 km had over 1000 faults, each taking at least 12 hours to repair. The local network provider, Orion New Zealand Ltd, called on assistance to find and repair these faults and 13 crews from around the country have been working around the clock to restore some semblance of normal operation. Most of these crews are using Megger PFL portable cable fault locators and, thanks to the generosity of Megger, AVO New Zealand has placed four new PFL test sets with contractors working on earthquake recovery. To date 95% of faults have been repaired, but it is estimated that it will be a five-year job to get the network back to normal. Aftershocks continue and with them more cable faults are created. One year on, there have been some 8,300 after-shocks with 25 of these higher than magnitude 5.0. Scientists say it’s clear we are living through a major seismic event and they can give no guarantees about when it will finish. Aftershock activity will have to die away below normal level of seismicity before anyone can say that it’s truly over.

Kia Kaha (stay strong), Christchurch.