Thermal limit for Class A insulation Vs limits in IEC 60076 for transformers

[RasikaC]

Dear All,
For "class A" insulation the thermal limit is specified in standards as 105 degrees of Celsius. Distribution transformers are made with "class A" insulation. But,IEC 60076-7 standard for transformers, takes 110 degrees as the temperature which corresponds to a relative aging rate of one. What is the disparity between 105 and 110? Kindly help to understand the difference between the two.

 

[Gr8blu]

Rasika: Think about how the temperature is actually measured. For most rotating machines, the conductor is covered in electrical insulation - which acts as a thermal "blanket". This means the detector is seeing an attenuated (i.e. lower than actual) value. How much is the difference? Depends on the thickness of the insulation, as well as the insulator(s) thermal properties - and to some extent, exposed surface area.

Transformers, on the other hand, tend to have bare copper (or at most have the conductor covered by an insulating varnish) for their windings. This means the difference between what can be measured (on the surface of the varnish) is a lot closer to the actual conductor temperature.

Finally - consider the way temperature is measured in terms of "device vs conductor resistance". In the rotating machine, a lot of the conductor length has poor heat transfer characteristics in some areas (because it is surrounded by the lamination steel) and other sections have (relatively) good heat transfer (because the end turns are exposed to moving air) - which in turn means the average temperature determined by resistance is going to be wrong in both sections. A transformer coil, on the other hand, tends to have fairly uniform exposure to the coolant which means the average temperature determined by resistance is going the be a lot closer to the actual value.

Bottom line: it's all about "margin of error" between measured value and actual value.

Converting energy to motion for more than half a century