LV Bushings: The Weakest Member of Inverter Duty Transformers

My introduction with an inverter duty transformer (IDT) happened way back in year 2012 when I was heading operations at an OEM. It was a 2 MVA, 2 winding unit. The first case of LV bushing failures at site I had encountered in circa 2013-14. And even after a decade the issue is still unresolved.

Recently our team has replaced hundreds of bushings at a 150 MW solar energy generating station. The pile of failed bushings was looking like a graveyard. I got a picture but do not have courage to post it here.

The issue is more than a decade old and still not resolved. It is irrespective of make of the bushings or transformers. They are failing without any bias. It is not that attempts are not being made to address the issue. From use of 3.6 kV bushings in place of 1.1 kV ones to making the palm wider to increase contact area to changing material from copper to aluminum to using compounds of higher temperature class to improving bus bar arrangements in the cable boxes. But still the issue persists.

I think the reason is the type of application of IDTs. IDTs have a unique loading pattern. Unlike other transformers, IDTs are subjected to cyclic loading, every day over all 365 days of the year. They are loaded at their highest when ambient is also highest during afternoon and lowest when ambient is also at minimum during night. This load pattern is unique to IDTs.

The LV terminals are of molded type. The metal part (flat/Bar) is encapsulated into epoxy compound to form the Bushing. The coefficient of thermal expansion of metal and epoxy are different. The mating surface of epoxy and metal undergoes through a cycle of mechanical stresses due to the cyclic expansion and contraction of these two materials. This happens at different rate and is of different quantum. This must be causing development of micro cracks at the joint and resulting into leakage of terminals. However, it is not confirmed whether this phenomenon is really happening.

To counter this, use of bushings made from compounds of higher temperature class was done. This compound works well even at higher temperatures up to 180°C. As maximum temperature at bushings is found around 120-125°C, this solution should work. It would be interesting to know that how effective this has been.

The good aspect of this issue is it does not cause loss of generation as replacement of damaged terminals are carried out during night when generation is off. However, the impact is felt by transformer OEMs. Being the supplier of transformer the onus of bushings failure comes on the OEM. Bushings’ replacement is costly as you require diesel generator, filter machine, skilled manpower besides new bushings. All these expenses have to be borne by OEM if the transformer is within warranty period.

Scientific research is required to find out the real reason behind these failures of LV terminals and work out the solution. Ideally, being the most affected party, the transformer OEMs must take up this research. However, they neither have the domain expertise nor the resources. On the other hand, bushing manufacturers are less acquainted with application aspect of their bushings. Would it be appropriate that both of them adopt a collaborative approach towards the solution?

Alternatively, Should OEMs also explore switching back to porcelain type bushings? Flip side of these bushings is they have more leakage points. However, if installed with right process and skill they are less likely to leak.

What more can be done to address this seemingly minuscule but potential havoc?