IEC 60076-7:2005 - Loss of Life of Transformers 2

Kevin,

The following should be considered when reading the Loading Guide:
1) The loading guide assumes that the transformer's life is determined by failure modes based on chemical reactions. This is true for many of the important failure modes, e.g. paper degrading, but not true for some other failure modes.
2) The temperature dependence of reaction rates for chemical reactions is explained by the Arrheniu's equation. The equation gives reaction rates based on absolute temperature, therefore the loading guide uses the absolute temperature of the transformer (and not the temperature rise above ambient).
3) There is no such thing as a normal life expectancy for a transformer, it "depends".
4) How good are the transformers that have been operating perfectly fine for 60 years? Can they withstand a short-circuit without a tank rupture? Determining the "end of life" can be tricky.

Best regards,

Jens

Hello Jens,
Thank you for your reply. Again I lay emphasis on Table 4 of the IEC in relation to the loss of life. Am I right in my understanding of it?

Thank you
Regards
kevin

It depends mostly on the hot spot temperature of the transformer. In many older transformers, this is unknown entirely. A transformer with a more uniform temperature gradient through the winding may be able to handle well beyond recommended overload situations. On the other hand, a transformer with poor cooling duct design or excessive localized stray flux points may be within rated top oil rise, but well beyond the limits of the paper insulation in a few areas. This can lead to accelerated aging or failure despite the gauge showing a reasonable temperature.

It also depends on overhead in the transformer design. I've seen 65° C rise transformers range anywhere from 45° C to 66-68° C rise (penalty applies) at rated MVA during heat run tests.

Although publications such as IEEE C57.91-1995 suggest that a life span of about 20 years for a power transformer, today there is common belief that an oil-filled transformer is designed for a life expectancy of 30 to 40 years. As mentioned in the original post, there is however transformers that has gone beyond this age and is still in operation.

Perhaps the longevity of power transformer could be attributed to advance in insulation materials, overvoltage protection with metal oxide surge arrester, protective relays, better manufacturing QA/QC process, etc. Design safety margin on the other hand, has been reduced in the pass decades do to competitions and advance computer tools.

According with a FERC report, the average age of installed large power transformer (>100 MVA) in the United States is approximately 38 to 40 years, with 70 percent of being 25 years or older. While the life expectancy of a power transformer varies depending on how it is used, aging power transformers are potentially subject to an increased risk of failure.

Fundamentally, the major components of a transformer are windings with associated solid insulation, and the steel parts e.g. core, frame, and tank. The steel parts can last for many years. This is true also for the copper, from which the windings are made. This is however not the case with the insulation system.

IEC 354 thermal model for transformer hottest-spot calculations is very similar to the IEEE Std. C57.91-1995 classical model. Should be noted that
transformer failure could be best characterized as a stochastic (probabilistic) model rather than a deterministic approach. For illustration, below are a few graphics and picture from various source showing basic probabilistic tools to consider in failures analysis in power transformer. Hope this help with the discussion.

It is worth to point out that several utilities, which adopt the N-1 security criterion, rarely have an average load higher than 50% on their transformers. When the average load is low, thermal ageing is almost negligible and mechanical wear due to short circuit stresses is the most significant fault cause: this is tipically the case of EHV/HV transformers and autotransformers.

Si duri puer ingeni videtur,
preconem facias vel architectum.

Kevin,

Regarding table 4:

Kevin1990 said:

Table 4 gives you the limits of Normal cyclic loading, Long time emergency loading and short time emergency loading. My understanding is that these limits values, if ever exceeded, will cause permanent damage to the transformer. However, it implies that the loss of life will be 'accelerated' compared to normal loss of life for a transformer operated at rated loading.

Click to expand...

I share your understanding of the guide; exceeding the limits in table 4 can lead to permanent damage but when operating within the limits of table 4 the relative aging calculations apply.

Rgs.
Jens

Your understanding is correct. Within those maximum limits, the ageing will double for every 6C rise above a hot spot of 98C (see Table 2) But there is another parameter ie moisture content in paper. In a new transformer it will be less than 0.8 % by w/w, but within a few years it may reach 2% or more. For the same temperature of 98C, life is 15 years with new transformer, but will be only 0.8 years with 2 % moisture (Table 1. So moisture will increase ageing drastically and be careful when overloading old transformers. It can also increase chance of bubble formation

Thank You All,
Highly Appreciated

Cheers
Kevin