re-commission of 11/6.6 kv transformer

[ukzn]

I need some help in determining whether or not to re-commission a transformer or not, has we are not 100% sure that the transformer had or had not seen a through fault. The transformer feeder did not trip but the incomer to that board did trip on Timed- earth fault(51G), no other protection on that transformer did pick up, ie transformer diff. on the day of the assumed fault we did megger all 3 phases with the red phase going straight down to earth, but the blue and white giving me 2.3 M ohms of 5kv, but this to me is incorrect has the transformer is a Delta/star and I Meggered the delta side. The following day I deiced that I will do a ratio test, sensitive resistance test and Hi-potential test. But to my surprise the transformer came out clean. Even when re-meggered the values were 21 M ohm between all phases and earth. But from DGA I found high levels of Acetylene ( 89 ppm) and that shows the transformer is not operational. Can I get some other advice. I have already taken the transformer out and sent it for refurbishment.

" THINK BIG "

 

[Marmite]

High levels of acetylene could point to a very high temperature fault such as arcing, but you would also expect to see ethylene. I think you cerainly need to do a bit more detective work to determine what the fault is, before concluding the transformer is faulty. From what you've said, apart from the DGA you have nothing to indicate a faulty transformer. If the DGA was dodgy then you might be undertaking expensive "repairs" for nothing. The protection operation if genuine only indicates an earth fault somewhere downstream. Why didn't the differential protection operate? Why didn't the transformer feeder protection operate? Is there REF protection? Is there a Bucholz relay? Is the transformer cable connected? I would re-do the DGA.
Also carry out insulation tests core/tank, between windings and earth, and between HV and LV windings. Carry out winding resistance tests on highest lowest and nominal tap positions and compare with the manufacturers data. Also carry out magnetising current, ratio and tapchanger continuity tests and compare with the commissioning results.
Regards
Marmite

 

[oldfieldguy]

A transformer may be damaged by a through fault in such a manner as to be difficult to detect. High current from a through fault may overstress internal blocking and insulation resulting in turn to turn shorts.

The normal turns ratio test may appear normal because it is a low-voltage test and the turn to turn voltage differential is not enough to "buck" the resistance of the short between turns.

Unless the damage encompasses insulation between the high and low-voltage windings or winding to ground insulation, it will not likely show up on the standard power factor and insulation resistance (Megger) tests.

Also, the standard winding resistance test will appear normal, since the fault has not affected the continuity of the winding. The test current takes the path of least resistance through the shorted turns and therefore appears to exhibit normal resistance.

The presence of acetylene seems to point to the arcing associated with a possible shorted turn or turns. I've also seen the shorted turns cause a transformer to draw abnormally high current under 'unloaded' conditions as current circulates through the shorted turn.

old field guy

 

[ukzn]

guys i have all what marmite said, and i am pondering about the same things that he is. it does have Bucholz and even explosion vent protection. but the transformer is out and i will inform you all of the findings

" THINK BIG "