Moisture in Liquid Filled Transformer

Status
Not open for further replies.
R

rpg

Member
Location
Kailua, HI, USA
We are installing several 1500 KVA medium voltage substations. They were shipped with a N2 blanket, and most of them had a blanket pressure of approx. -2 lbs. About two weeks ago, the person who removed oil for initial testing opened the vent to atmosphere so he could get the sample out, thus introducing air/moisture. The transformers have not been energized yet and I'm wondering what damage this may have done and if purging and re-pressurizing with N2 will be a sufficient fix. I know you would need a lot more information to do an analysis, but generally speaking, could the oil have been contaminated at this point?
 
Little if any. And you cannot pull samples Lea most of the time with a vacuum. In addition pulling samples with the transformer not under load is a total waste of tome: the oil is not circulating and will stratify. Especially with corn oil there is a lot of gassing the first few months so an N2 blanket is kind of pointless. You also often jostle the bushings a little during install. And no matter what the insulation is never fully cured and continues to release a little moisture for a couple weeks after startup: Dissolved moisture in oil or insulation is harmless except for degradation. Liquid moisture above 2-4% lowers your BIL. There is no right answer but before that point you can hook up a filter system to clean it up. Some plants just filter for the first month after startup to remove anything remaining from initial manufacturing.

In a transformer the cellulose in the insulation naturally breaks down as it depolymerizes. This produces water. Moisture attacks the oil causing I to break down producing acids. The acids in turn attack the cellulose: this chemical reaction limits the ultimate life. The theory behind nitrogen blankets is to limit oxygen availability to slow down the reaction but my observation is you get 30-50 years total irrespective of whether it has a nitrogen blanket or conservator or not. By that time between mistreatment, gasket failures, and tank rusting it doesn’t matter anyway. FM Global recommends replacement after 30 years. Moisture is always present and far more is absorbed in the insulation than the oil. Some people interpret results based on

On GSUs with full instrumentation and oil filled bushings they use conservators for the most part, not N2. SPX manufacturers them and they are a customer. So if they aren’t doing it for 10+ MVA, it sounds like a solution looking for a problem.

When I pull oil samples I alert the customer to the issue and only bleed vacuum until the oil flows. If they don’t want to pull samples that’s their call.
 
rpg said:
We are installing several 1500 KVA medium voltage substations. They were shipped with a N2 blanket, and most of them had a blanket pressure of approx. -2 lbs. About two weeks ago, the person who removed oil for initial testing opened the vent to atmosphere so he could get the sample out, thus introducing air/moisture. The transformers have not been energized yet and I'm wondering what damage this may have done and if purging and re-pressurizing with N2 will be a sufficient fix. I know you would need a lot more information to do an analysis, but generally speaking, could the oil have been contaminated at this point?
Click to expand...

Don’t see a problem here at all. What happens when you bleed it off to remove a bay-o-net?
If your that concerned pull a vacuum and re-introduce N2.
the little moisture you could have possibly introduced by pulling the vent ring is nothing.

Look at a regulator and oil filled reclosers for example. These devices are free breathing for the most part
 
rpg said:
Again, thanks so much. As a dumb construction manager, I find the expert feedback in this forum most helpful.
Click to expand...
If your a construction manager, dumb is not something you are...😃
 
Status
Not open for further replies.
Top