Trends in transformer failure analysis
Part three of this series on Trends in transformer failure analysis will report on some recent progress on standards body projects that were described in...
byWallace Binder
Part three of this series on Trends in transformer failure analysis will report on some recent progress on standards body projects that were described in the first issue. It will describe some failures which have evolved recently. It will also identify many things that are not found in the IEEE Guide for Failure Analysis and describe the reasons why they are not included.
Recent development in failure analysis and reporting
In my first article on the subject of failure analysis [1], I described work underway in the IEEE/PES Transformers Committee to revise the IEEE C57.125 Guide for failure investigation, analysis, and reporting for power transformers and shunt reactors [2] and parallel work in CIGRE WG A2.45 Transformer failure investigation and postmortem analysis. What I neglected to mention is the IEEE effort included incorporating the process by which any entity can establish a database of a population of similar transformers, determine the differences and statistically come to some conclusion about how the population was behaving. The original work on developing the database was titled IEEE C57.117 Guide for reporting failure data for power transformers and shunt reactors on electric utility power systems.
We have merged the rules for statistical analysis, the appropriate definitions, and the suggested forms which we hope will become a part of the data describing the transformer so that its identity and pertinent rating information can be entered into a database. An Electrical Power Research Institute (EPRI) project took that effort to the pilot stage but the effort died for lack of funding or lack of interest. Fortunately, another U.S. company has stepped up to host a database for clients who wish to participate and have sufficient interest to proceed. Each participating company’s data has been converted into a “standard format” and the queries are developed to calculate various failure rates of interest to the group. The host is currently updating everyone’s 2013 data; adding in failures that occurred during the year plus adding in newly installed transformers and noting transformers that were retired so that the database maintains accurate population data.
The CIGRE A2.45 Working Group has met six times, the latest meeting being August 2014 in Paris. Their early meeting minutes recognised the C57.125-1991 as state of the art. We share some common interests although we are not actively coordinating the work between the two groups. The A2.45 WG is making progress toward a brochure which, based only on my reading of their minutes, appears it will address many of the same issues that are addressed in the IEEE Guide.
I am happy to report that as of this writing, the effort by the IEEE/PES Working Group has progressed to ballot of a final draft of PC57.125 and if all goes well, that ballot should be concluded by the time this issue of Transformers Magazine reaches you. The final draft contains tables of tests whose interpretation can help with the diagnosis of problems. I mentioned some of the new tests which I included in the “Alphabet test” category in the last article on Trends in transformer failure analysis contained in the second edition of Transformers Magazine [4]. There are other documents to which one can refer to obtain values where industry norms have set limits. It still remains up to the individual to analyse the test results and compare them to previous readings to detect trends or to detect when limits indicate that either a failure has occurred or is imminent.
While we have been revisiting the document, we have been made aware of several failure modes which have revealed themselves. These and perhaps others which manifest themselves in the future will have to become part of the next revision since there is a time limit on completing a revision within the cycle required by the sponsoring body (in this case IEEE-SA).
To read the article, subscribe and choose the option which suits you best. We offer both free and paid options, and the registration takes only a minute.
Subscribe to Transformers Magazine
