Corona in oil transformer

[unclebob]

Hi! One of our customers have an oil transformer and we do a dissolve gaz analysis annually. We found that the transformer have a corona (partial discharge) problem. The key gases are hydrogen (1150 ppm)and methane (110 ppm). The level of the other gases is insignificant. The progression
of H2 is about 200 ppm/year.

Now that we know this, is there a way to find where corona is occuring? We have made a visual inspection, from the main hole and found no grey-to-black powdery traces.

The transformer is:

Northern Transformers
1 500 kVA
25 000 - 600/347 V
sealed type
1984
ONAN

Any comments?

 

[alehman]

See page 32

http://www.usbr.gov/power/data/fist/fist3_31/fist3-31.pdf

 

[GamezBeCJ]

You can do a sonic scan, usually you would use an ultra-frequency sensor and trace the magnitudes around the transformer. The partial discharges will make the readings go high in the area where they are ocurring. I have used this method to successfully pinpoint where the partial discharges are coming from.

 

[electricpete]

I assume you already checked for oil moisture content but if not that should be checked.

I agree with your diagnosis of corona but is it remotely possilbe this represents overheating. We had a transformer allowed to operate around 120-140C oil temp for several days and if memory serves me right the results were H2 and CH4 and CO and CO2.

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