Statistics show that bushings are one of the main contributors to transformer failure. Their failure therefore has huge impact on reliability as well as safety of the system and operators managing the asset. With this in mind, major utilities in southeast Asia embarked on a joint bushing reliability survey and also jointly formulated mitigation measures to improve bushing performance. The survey, discussed in this contribution to INMR by engineers S. Gobi Kannan, Chitapon Jedwanna and Henny Ika, respectively at Tenaga Nasional Berhad (TNB) in Malaysia, the Electricity Generating Authority of Thailand (EGAT) and Perusahaan Listrik Negara (PLN) in Indonesia, summarized failure and root cause at each participating utility between 2001 and 2015. The different practices among members were discussed with a view to improving selection, specification, design, condition assessment, storage and handling of bushings. These changes were deemed necessary to have real impact on safety and performance of transformers and the grid as a whole.
Existing maintenance practices, standards and bushing technologies are all well established in the industry, although each have relative advantages and disadvantages. Drawbacks in practices are due mainly to differences in existing maintenance strategies and in acceptance criteria to address bushing degradation. There are also non-standardized dimensional requirements in terms of ease of maintenance and strategic spare requirements. For example, the disadvantages of an oil-impregnated paper (OIP) stye bushing are mainly related to fire risk and environmental problems. In the case of resin-impregnated (RIP) bushings, risk of moisture uptake during storage is a major issue. These disadvantages must be addressed individually to formulate an action plan to improve bushing performance. Looking to the future, some member utilities also embarked on new technology solutions including resin-impregnated synthetics (RIS) bushings and RIP moisture barriers technology, still under evaluation.
Cases of Failure
This section reviews several cases of bushing failure for purpose of compilation of lessons learnt and improvements made by the utility members to reduce risk of similar accidents.
Failure #1: Copper Migration Phenomenon on 300 kV Bushing Manufactured in 1996
Root Cause: Site and laboratory examination with historical failure records suggest that the dielectric insulation breakdown were likely due to one or a combination of the following:
• Poor stress control design of the bushing, resulting in high electrical stress at certain locations along the paper/foil insulation;
• Poor manufacturing quality as seen in lab examination in some paper/foil insulation layers leading to uneven distribution of electrical stresses;
• Development of electrical treeing along the insulation as a result of copper migration and accelerated by high electrical stresses.
Failure #2: Test Tap Problems
Surveys indicate that 10% of bushing failures at ASEAN utilities relate to test tap problems. In this incident, it was found that the test tap measurement and tap bushing had no continuity. Insulation resistance measurements conducted indicated an open circuit. Further disassembly of the test tap indicated the following:
Based on utility experience, the majority of test tap failures within the region have been attributed to spring-loaded design versus permanent connection types, as illustrated in Figs. 15 and 16. Poor spring contact pressure over years and possible damage during high surge current flow were concluded to be possible root causes. Failure leads to floating potential and arcing at the last layer of the condenser design bushings. This can cause contamination of insulating oil at bottom side of the bushing and eventually a dielectric failure breakdown.
Failure #3: Draw Lead Bushing Fast Transient Issue in GIS/RCT Sub
Root Cause: Loss of dielectric strength of the main condenser insulation due to heat generated along the bushing’s aluminium tube as a result of fast and very fast transient phenomena in the system during switching.
Measurement of Power Factor and Capacitance on the bushing prior to the failure did not indicate any abnormalities, as illustrated in Table 7.
The phenomenon of fast transients on draw leads of bushings is still not fully understood in the field. Evidence of arcing/sparking between draw lead and aluminum tube at common potential raises concern, especially in regard to reactive compensator or GIS substation installation failures. By way of improvement, the specification has been revised such that for bushings with draw lead connections, the transformer manufacturer shall ensure the draw lead cable within the tube is insulated using thermally upgraded insulation material with a minimal thickness of 1 mm. This, without causing any significant rise in hot spot temperature of the bushing to prevent arcing between draw lead to aluminum tubing during any systemic fast transient phenomena.
Failure #4: Monkey Damage to Polymeric Insulators (Still Unresolved)
As initiative to reduce risk of explosion and deal with pollution concerns, some member utilities have begun application of polymeric insulators as means to reduce such risks. However, there have been cases of polymeric-housed bushings being attacked by animals such as monkeys that chew the insulators. Discussions have been conducted with suppliers looking into the possibility to inhibit this through use of repellents but no commercial solution is yet offered in the marketplace for transformer bushings. In recent years, sheds damaged this way require urgent replacement to prevent external flashovers and avoid possible moisture ingress into the main condenser layers. Up to now, utility members have not reported any flashovers given that damaged bushings were detected during routine substation inspection and replaced before such problems could occur.
Failure #5: Hygroscopic Issues on Resin Impregnated Paper Bushings
Resin Impregnated Paper (RIP) bushings are vulnerable if exposed to high humidity during storage or handling and this hygroscopic property needs to be addressed. Long-term exposure under an outdoor environment with high humidity can result in the bottom portion of an RIP bushing absorbing moisture that can then affect performance. Change in colour of the bottom part of these bushings is an indication of this problem, which is a major concern among utility members due to the high humidity environment in this part of the world.
Measurements performed will not show significant change in C1 parameters, however there is a change in pF C2 values (after assembly) due to humidity ingress at the bottom surface/core of the bushing, as shown in Table 8:
For RIP bushings with high humidity, drying in an oven has been found effective in removing the moisture ingress at their bottom surface. As shown in case study data in Fig 25 and Table 9 below, these issues are usually detectable by high pf C2 values after assembly, while typical values of 2-3% are the acceptable limits recommended by the OEMs. It is typical that PF for the overall bushing (HV terminal to flange) is near to C1 values for dry bushings. This is a useful parameter to monitor during the drying process since pf C2 measurement may not be accurate when the bushing is tested by itself without mounting.
Initiatives & Risk Mitigation Measures
While risk of fire in transformers due to bushing failures is relatively low, it is not negligible. Prevention of transformer fires and implementation of strategies to prevent loss of life, minimize loss of adjacent assets as well as loss of supply is therefore an essential part of managing transformer assets. The goals of the strategies or initiatives proposed below are to address and quantify the risk of such fires and provide guidance on how fire risk can be managed most cost effectively.
The following list of risk mitigation measures is based on discussions and best practices between utility members and deemed effective to improve the performance of bushings in the grid:
1. RIP Bushings are preferred instead of OIP for condenser bushing classes 52 kV and above;
2. Procure only from suppliers with proven designs. A product pre-qualification process is recommended;
3. Oil filling and sampling is not recommended unless performed by personnel with related competence or under OEM supervision. This is applicable only for OIP bushings;
4. Enhance maintenance practices by all member utilities, including
• Replace and monitor bushings exceeding acceptable criteria;
• Power factor and capacitance measurements are sensitive parameters to detect incipient problems;
• Thermal scanning at least once every 6 months;
• Visual inspection at least bi-monthly looking for leakage and severe external contamination.
5. Follow storage and handling guide for RIP and OIP as per OEM recommendation. Fig. 26 below shows proper storage practice on a long-term basis exceeding 12 months. Long-term storage is done with specially designed metallic covers filled with dry transformer oil.
Use of an RIP protection coating as a water barrier is under evaluation and requires further study on effectiveness in the field:
• One standard uncoated RIP bushing
• Bushing with water barrier protective coating
Fig. 29 below illustrates moisture ingress testing on a standard bushing and a bushing with the special coating currently being used by some member utilities. Test results are shown below.
Based on water ingress tests, it can be concluded:
• Effectiveness of a water barrier coating is still under field evaluation;
• Experimental analysis proves the barrier coat prolongs duration of moisture or water ingress into bushing layers by multiple of about 80 versus compared to uncoated RIP bushings;
• Increase in power factor despite use of this coating suggests that the coating is effective for handling but not for long-term storage purposes.
• No migration of moisture must be possible through the external insulation to the RIP insulation condenser body. All bushings rated voltage 52 kV and above shall have either fibreglass or an alternative proven material that acts as moisture barrier between the RIP condenser body and the external composite insulator. Fig. 30 illustrates such a fibre glass moisture barrier for RIP bushing designs. There are also OEMs that use pure resin or SF6 foam instead for similar purposes. This is acceptable among all utility members.
7. For purpose of interchangeability and standardization, oil-air bushings of rated voltage 52 kV and above shall comply with standard bushing requirements in terms of dimension and rating as shown in the sample bushing dimensional standardization (Fig. 31 below). This has already been successfully implemented by one member utility.
8.For existing units in service in terms of replacement or refurbishment, it is recommended that the bushing be supplied to exactly match the existing dimensions. This to ensure bottom clearances are met as per initial design and to prevent substantial modification and retrofitting work at the site;
9. All type tests and routine tests shall be carried out according to the latest IEC or IEEE standard requirements. Type test reports and certificates shall be submitted to the user on delivery;
10. Partial discharge test shall be performed in accordance with requirements of the latest version of IEC 60137. Routine measurements to detect internal PD shall be, at maximum, virtually discharge free ≤5pC at 1.05 Um/ Ö3 and ≤5pc at 2.0 Um/Ö3, where the more stringent requirement(s) shall apply. If the bushing fails the PD test, it shall be rejected and not be reconditioned for later use in the system.
11. For bushings with draw lead connection, the transformer manufacturer shall ensure the draw lead cable within the tube shall be insulated using thermally upgraded insulation material with minimal thickness of 1 mm. This without causing any significant rise in hot spot temperature of the bushing so as to prevent any arcing between draw lead to aluminum tubing during any system fast transient phenomenon;
12. For fixed solid conductors, the transformer manufacturer will ensure a flexible lead is provided between transformer and bushing to avoid any vibration and dilatation being transmitted during operation;
13. The current rating of bushings to be supplied shall be inclusive of overload factor of at least 1.2 times maximum rated current for transformer rated capacity;
14. The test tap shall be manufactured from non-corrodible material and be sealed by means of non-corrodible screw-on dust cover during service. A reusable (oil, heat and UV resistance) seal shall be provided to prevent ingress of moisture and or other impurities from entering the test tap. Use of nitrile rubber with sufficient material hardness is preferred;
15. The internal connection of the test tap to the outmost condenser foil layer is by means of permanent electrical connection. Spring loaded type internal connections are not recommended due to previous poor service performance caused by loss of spring contact pressure over time period;
16. Air insulated cable boxes on transformers shall have arc venting. Such venting could prevent damage to the cable box, dislocation of the cables and breakage of bushings, as shown in Fig. 32;
17. Use of an online monitoring system is recommended for critical installations. It is the prerogative of member utilities to study the suitability of different technologies available to assess bushing conditions in the field. This handbook does not state any preference. Fig. 33 illustrates installation of a bushing monitoring on a power transformer and one sample case study findings are elaborated;
• Findings revealed comparison of offline and online monitoring information relatively agreed for bushings in a healthy state (Refer Table 10). The system can serve as a useful tool to monitor relative change in bushing condition. However, the benefits of such system have to be considered on a case-by-case basis, comparing risk to equipment of the system being monitored against the economics and reliability of such a monitoring system.
18. Using resin impregnated fibre technology for dry bushings to mitigate issues related to the hydroscopic properties of conventional epoxy resin impregnated bushings with conventional paper insulation. TNB embarked on a First Pilot Project in Malaysia in 2013.
• A Pilot project was conducted at 170 kV using new RIS technology. Results, as illustrated in Table 11, indicate superior performance compared to conventional RIP bushings.
Table 11 below summarizes the parameters measured and evaluation throughout the evaluation period against acceptance criteria by the project team:
19. Overvoltage Protection
• Metal oxide surge arresters are an important part of transformer protection in reducing risk of a transformer failure due to surges or transients in the system. All line terminals of the transformer should be protected by surge arresters.
Despite relatively low failure rate compared to the total population of bushings in the system, failure of bushings has a catastrophic impact on reliability, availability and safety of the ASEAN grid. This survey and development of the first ASEAN Bushing Guidebook offers significant benefit to standardize and review existing practices. This so that they will be in line with the best practice of international standards and also as outlined by several effective risk mitigation measures and improvements based on previous incidences of failure. The effective implementation of these proposed risk mitigation measures and continuous improvement will improve performance in general with reduction in bushing failures among members versus in past years.