Transformer cooler controls

[Bung]

We have a bit of a debate going at the moment about the temperature "gradient" between oil temp indicator settings (OTI) and winding temp indicator settings (WTI) on power transformers using oil pockets and load current driven heaters to emulate the winding temperature. We have a number of transformers for which the data is lost in the mists of time, and we are trying to recalibrate the controls.

Does anybody have any typical figures they use for their transformers? I have heard figures of anything from 15C to 25C at full load. Shorting out the WTI (seting the gradient to 0) seems a bad idea - you would delay the switch on of fans and pumps (usually controlled by WTI). But setting it too high could cause you to trip a transformer unnecessarily when it is heavily loaded on a hot day, just when you really need it. Present consensus is to go for a compromise of 10 deg C.

Any thoughts or suggestions gladly accepted! (Figures in Celsius or deg Fahrenheit OK, just be specific about the units used!)


Bung
Life is non-linear...

 

[electricpete]

I don't know the answer (I think I probably knew it once but it's been 10+ years since I worked on transformers).

As you know the factory would select these values based on test data.

Is the winding temperature indicator supposed to represent hot-spot winding temp or average winding temperature? To the best of my memory you are trying to display hot-spot winding temperature.

For a 65C transformer, the rating basis assumes a maximum 15C difference between the hot spot and average winding temperature ("hot spot allowance").

So a conservative approach would be to take that 15C and add some additional allowance for rise of average winding temperature above oil. I'm thinking that may not be more than 5C, so 15+5=20C would probably be a conservative number.

But as I said it's been a long time and I'm going from memory which ain't as good as it used to be.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[stevenal]

If the manufacturer is not around to help, Qualitrol might be able to supply some typical values.

 

[electricpete]

Just to clarify my previous post fwiw I meant 20C as the upper bound for the difference at full load. "conservative" was probably not a good choice of words.

I'm remembering from somwhere "typical" difference hot-spot to average winding is 8C and "typical" difference average winding to top-oil around 0C. Might be wrong. Same disclaimer as before.

Good idea to call Qualitrol or possibly Wechsler (sp?) who made those microprocessor transformer monitoring gadgets if you can't get hold of the transformer manufacturer.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Last comment and then I'll really shut up. What I called typical was based on FOA GSU transformers.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Bung]

Thanks electricpete - every litle bit helps. I did find a bit in the old Westinghouse book in the chapter on transformers, but it still requires a knowledge of more about the transformers than I already have. And if we are going to guesstimate, I would like to at least know the size of the ball-park.


Bung
Life is non-linear...

 

[electricpete]

One thing that would be pretty easy to do is walk out to some operating transformers and read the load, winding temperature, and oil temperature, along with nameplate info. I'm not sure if you have any transformers available. If you'd like me to do it let me know... my desk is only a few hundred yards from the transformers at our power plant.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[jnims]

The 1981 version of the loading guide has the following assumed characteristics(rise above ambient) at rated load for a 65C transformer:
OA FA FA
Hottest-spot 80C 80C 80C
Top-oil 55C 50C 45C

Values would vary for each transformers specific design, but I have seen published temperature rise tests that show the difference between top-oil and hottest-spot at about 22C.