 |
|
|
|
|
|
|
 |
|
|
|
Published by Megger
Oct 2010 |
|
| The industry's recognised information tool |
|
ELECTRICAL
TESTER |
|
|
|
|
|
|
|
In this issue |
|
|
|
|
The next step in insulation diagnostics |
|
|
Moisture content detection in oil immersed current transformers |
|
|
Magnetic losses, finances and environment |
|
|
|
The next step in insulation diagnostics |
|
|
|
|
|
|
|
|
| Matz Ohlen |
| Director - Transformer Test Systems |
|
 |
|
| High-voltage insulation testing is an essential tool for condition assessment of almost all major items of electrical power plant, including transformers, bushings, circuit breakers, cables and rotating machines. Many of the insulation test sets currently in use, however, have significant shortcomings, which means not only that they are less convenient to use than they should be, but also that the results obtained are less comprehensive and less reliable. |
|
| A new generation of instruments is now available which addresses these shortcomings. To see what these versatile products have to offer, let´s take a look at one of the latest 12 kV insulation diagnostic systems. |
|
| The first thing users are likely to notice is that it is much more portable than its older counterparts – its two-piece design weighs a total of 36 kg, making it probably the industry´s lightest power factor test set. Another useful benefit is provision for fully automatic tan delta/ power factor and tip-up testing, which is a big time saver. Facilities are also available for manual testing – including the option to increase the test voltage during the test – to allow special testing requirements to be accommodated. |
|
| The ability of the test set to generate its own test voltage, which can be varied in frequency over the range 1 to 500 Hz not only increases its versatility, but also ensures that dependable and repeatable results are obtained even when the instrument is fed from a poor quality supply. |
|
| Another benefit is automatic voltage dependence detection. If the instrument detects, for example, that the dissipation factor of the test object varies with the applied voltage, which suggests there is a problem that requires further investigation, it instantly provides a user alarm. |
|
| The inability to correct accurately for temperature differences when calculating results is a weakness of many insulation test sets currently in use. The new generation instrument overcomes this by using a novel method to apply individual and accurate temperature compensation for the actual test object. This is based on carrying out an additional DFR measurement and |
|
|
 |
|
| mathematically converting data at different frequencies to data at different temperatures. |
|
| Results analysis is a key aspect of insulation testing and, in the new instrument, analysis is facilitated by allowing immediate comparisons to be made between the current results and stored data sets. Comparisons of results of obtained at different voltages and frequencies can also easily be made. |
|
| The instrument is supplied complete with powerful industry-standard acceptance and maintenance test data software, which not only offers extended test automation options, but also provides comprehensive facilities for archiving, analysing and reporting results. |
|
| The instrument described in this short item is Megger´s recently launched DELTA4000 series 12 kV insulation diagnosis system, which is supplied with the powerful yet easy to use PowerDB software package. Developed after careful analysis of user requirements in relation to high voltage insulation testing, this innovative new test set is a significant step forward in high voltage insulation testing technology. |
|
|
|
|
|
|
Moisture content detection in oil immersed current transformers |
|
|
|
|
|
|
|
|
| Diego Robalino, PhD |
| Applications Engineer |
|
|
|
| Part I. Measurement Set Up |
|
| As discussed in the last issue of Electrical Tester, frequency domain spectroscopy (FDS) is an invaluable tool for determining the moisture content of the solid insulation in combined oil-paper insulation systems. In particular, work carried out in North America and Europe has verified the effectiveness of FDS as way of determining the moisture content in instrument transformers and, more specifically, in high voltage current transformers (CTs). This article describes FDS testing of CTs. It includes configuration data and a step-by-step test procedure using the Megger IDAX 300 test set. Part 2, which will appear in the January 2011 edition of Electrical Tester, discusses analysis of results using MODS software and gives recommended values for new and aged units. |
|
| The dynamic properties of dielectric materials can be measured in the time and/or frequency domain. The fundamentals of dielectric response functions and the theory of dynamic properties of dielectrics are well described in several publications and, in particular, in a very detailed way in two articles by W S Zeangl, which appeared in the IEEE Electrical Insulation Magazine Vol. 19, 2003. |
|
| The traditional dissipation factor testing technique allows identification of the deterioration process of the insulation by measuring the changes in the dielectric properties of the tested unit. This approach involves measurements of capacitance and dielectric loss quantified by the loss or dissipation factor. This type of testing is part of many manufacturing quality control procedures but it is normally carried out at power frequency only. With FDS, however, a wide frequency range from 0.001 to 1000 Hz is used, and this allows the determination of the moisture content in the solid insulation. |
|
| Water significantly accelerates the ageing of cellulose. Oil analyses by means of Karl Fischer titration (KFT) have traditionally been used for the evaluation of moisture content, assuming existence of equilibrium in distribution of moisture between oil and paper/ pressboard. In reality, the analysis only reflects the moisture percentage in the liquid insulation and equilibrium curves are required to estimate water content in the cellulose insulation. The equilibrium curves are a useful reference but their reliability is still a topic of discussion. Moreover, continual sampling for DGA and water content analysis is not recommended for CTs due to the small volume of dielectric oil they contain. “A chain is no stronger than its weakest link”………. [1868 L. Stephen in Cornhill Mag. XVII. 295] |
|
| Failure of a substation-type oil-immersed CT can lead to a high-energy release and thermal runaway, very possibly ending in an explosion (see Figure 1). Because of the difference in thermal expansion ratio between the metallic housing of the CT and the relatively fragile porcelain insulator, mechanical stress builds up, resulting in a blast where fragments from the porcelain insulator may reach up to 50m from the location of the unit. Loss of this important device results in phase to ground fault that will trip the substation, shut down operation and possibly affect other electrical components in the vicinity. |
|
 |
|
| Figure 1 - Failure of a CT in a substation |
|
| The internal construction of typical oil-immersed CT is shown in Figure 2. The high voltage/high current primary winding is a conductor, a bar or a set of conductors passing through the window of the toroidal core. Engineers or technicians testing the unit should verify with the manufacturer the construction of the primary conductor, as there are applications where a single conductor is wound inside the CT to provide multiple turns. The core is surrounded by the low voltage winding, which is evenly distributed all along the toroidal core. This is completely covered by solid insulation, which wraps the secondary winding and the core in multiple layers of paper insulation. The core, the primary and secondary windings and the solid insulation are fully immersed in liquid insulation (mineral dielectric oil). |
|
|
 |
|
| Figure 2 - Cross Section: CT windings, core & insulation. |
|
| In some applications, the core housing can accommodate up to six independent multiratio cores feeding protection relays, or cores feeding a combination of relays and meters, requiring up to 30 secondary leads. More details of the construction of oil-immersed CTs can be found on manufacturers´ web sites and literature. |
|
| A very real concern |
|
| The issue of exploding CTs is currently a very real concern as, in Europe and North America, there have recently been several catastrophic failures of hermetically sealed units in sub-station applications. Initial investigations carried out on CTs similar to those that have failed, using dissolved gas analysis (DGA), have revealed that the failures are due to moisture ingress, and that many CTs still in service are at risk of similar failures. Regular CT testing is, therefore, increasingly seen as essential but, as has already been noted, tests that involve oil sampling, which includes DGA testing, cannot be used regularly on CTs because of the small volume of oil they contain. FDS testing, which eliminates entirely the need for oil sampling, is therefore establishing itself as the preferred approach. |
|
| Performing FDS tests on an oil-immersed current transformers |
|
| First make a visual inspection of the unit and its surroundings. Ensure that local safety procedures (tag-out/lock-out) have been observed and that the test area is properly identified and is free of obstacles on emergency evacuation paths. |
|
| The procedure for testing a HV CT uses the same configuration as that used to perform power factor/dissipation factor tests. Thus, the unit must be isolated from the power system (primary as well as secondary winding) and discharged. If the test is to be performed after a through fault, it must also be demagnetized. As a general rule, when a series of tests is to be carried out on an HV CT, DC tests should be performed last. With a clear area of operation established, confirm that good connections have been made to the unit under test (UUT), to the substation´s ground system, and from the test instrument to the same grounding point. Typically, the ground terminal of the test equipment is connected to the same ground terminal on the secondary box of the CT. The CT is tested using the standard method. This implies energizing the primary winding and measuring the secondary connected to ground. The recommended test setup is grounded specimen test (GST) as shown in Figure 3. |
|
 |
|
| Figure 3 - GST measuring setup for HV CT |
|
| Normally, there is no need to short circuit the primary winding but if, as mentioned earlier, there are several units built in with multiple turn arrangement of primary conductors, these need to be shorted. Bar-type and single-conductor primary windings do not require short-circuiting of the P1-P2 terminals. |
|
| The UUT should preferably be at thermal equilibrium. The average temperature of the insulation should be measured or estimated and recorded. One option is to measure on several positions on the outside of the CT core housing with a pistol grip laser target temperature gauge or, if the CT has not been in recent operation, the insulation temperature can be assumed to be the same as ambient temperature. |
|
|
|
|
|
Magnetic losses, finances and environment |
|
|
|
|
|
|
|
|
| Dr Stan Zurek |
| Magnetics Technical Specialist |
|
|
|
| All electric energy in the national grid is transformed several times before it is delivered from the power plant to the final user. The high-voltage power transformers used are devices of considerable weight, size and cost. However, the cost of a transformer relates not only to materials and manufacturing, but also to the energy lost over its lifetime. Read on to find how magnetic losses affect economy and the environment. |
|
| All transformers dissipate some energy in the process of transformation. Operating principles dictate that the magnetic core is always magnetised to the same level, regardless the load. This means that even when transformer works with no load, the magnetic losses are almost the same as when it is operating at full load. |
|
| At full load, the efficiency of large power transformers can be very high – above 99% in some cases. However, because the magnetic loss is practically fixed, efficiency reduces with the load. Inevitably, under no load conditions any transformer operates with 0% efficiency – it consumes some energy to remain energised, but it delivers no energy to any load. |
|
| The electricity transmission grid and all the transformers in it are exposed to varying load. At night the load is relatively light, so the transformers operate at much less than nominal conditions, but they keep dissipating exactly the same energy in magnetic losses. |
|
| Remembering that a power transformer can have a life of 20 years or more, let´s now consider the hypothetical case of a 10 MVA (10 MW) transformer that dissipates 1% of the energy in magnetic losses. With a simple calculation, we can conclude that over 20 years 614 MWh of energy is lost and, if the price is 10p per kWh, the cost of the losses is £61000. This is just for a single transformer – if we consider the case nationally, not to mention globally, then we can see that the amount of energy lost, the money associated with it and the environmental impact are tremendous. |
|
| When transformers were first invented, magnetic cores were made from so-called “soft iron” . This material was quite lossy and, ever since, engineers and scientists have been striving to improve efficiency by introducing a number of technological modifications – to chemical structure, mechanical processing, annealing, postprocessing, etc. |
|
| The diagram shows how these changes have helped to reduce losses over time. |
|
 |
|
|
 |
|
| Nowadays, the most commonly used material for magnetic cores of power transformers is electrical steel, which is a descendant of soft steel. However, the power loss of the best grade of electrical steel now in use is around 50 times lower than that of soft steel. This has had a very significant impact on the overall efficiency of power transformers. |
|
| In some cases, medium voltage power transformers are still in use after 50 years of operation. Because these old transformers were made from inferior magnetic material, their losses are much higher and replacing them with modern units would most likely bring noticeable savings in the energy distribution network – a transformer made 50 years ago could lose 5-10 times more energy (for magnetic loss) than a modern equivalent with the same rating. |
|
| The production of the grain-oriented electrical steel is a very complex process (perhaps a topic for a future article). Further improvements are on the horizon, but currently are too expensive to implement. However, another approach is possible – different material can be used. |
|
| In the US many smaller distribution transformers are made from material called “amorphous” . Its operating flux density is slightly lower, but the losses are much lower than in the case of electrical steel. |
|
| Amorphous transformers are slightly bigger, heavier, and more expensive to make, but the energy savings of their lifetime easily justify the larger capital cost. Yet in Europe such transformers are still not very popular. |
|
| To ensure correct operation, several properties of transformers must be tested regularly: electric insulation, degradation of oil, turns and voltage ratio, etc. Higher losses in the magnetic core can influence to some degree the voltage ratio or the phase shifts between the test signals. However, this is usually not a problem because correctly designed transformer testers can cope with these effects, even for older transformers. |
|
| Study and research continue into the problem of losses in magnetic materials. Manufacturers of electrical steel and amorphous ribbon sponsor many of these studies, and there are a number of international “magnetic” conferences where the newest developments are presented and discussed. There is, for example, pressure from industry for making acoustically quieter transformers. Also, rapid development in Asian countries is making electrical steel one of the hottest items on the market, so that the manufacturers have serious problems in keeping up with demand, despite the recent economic crisis. |
|
| One thing is, however, certain. In future we can expect even better and more efficient magnetic cores. The improvements might not be ground breaking, but every single percentage point counts: for efficiency, for running costs and for the environment. |
|
|
|
|
|
|
|
News 