When does a transformer become a bomb? 5

[miketech59]

Could anyone please advise me ? Currently we are having problems with a Parsons 1963 generation transformer rated at 59MVA with a conservator . The oil is water cooled .The problem seems to be the load Tap changer which for the last ten years has be only changed of-line .Dissolvee gas analysis found high levels of ethylene in the oil. The unit was inspected and no slack connections found .The oil was then filtered . On its return to service the buchoz alarm was triggered after two days .A DGA was done every day on the unit . The results showed Hydrogen 784 at 89.48 ppm per day ,Methane 126.53ppm per day, ethane50.67 ppm per day,ethylene 262.18 ppm per day, acetylene 2.34 ppm per day. The unit cannot be replaced before 2007. What can be done ?

 

[electricpete]

Based on mike's comment that ethylene rather than ethylene is the dominant gas, I now believe that an actual hotspot is a more likely explanation than my previous proposed explanation arcing type LTC.

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

 

[electricpete]

I meant ethylene vs acetylene

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

 

[alehman]

The extremely high level of ethylene production along with hydrogen and acetylene indicates an overheating element (e.g. TC contact) which is beginning to arc.

What subtech said!

 

[RRaghunath]

rcwilson,

contents of the post are interesting. Thanks for sharing.

 

[stevenal]

Pete,

Please reread the OP. Tap changes only take place when unit is off line. No arcing should occur from any style of LTC. The arcing/heating is most likely from carbon build up as ScottyUK first stated and my Cooper rep seconded.

 

[electricpete]

Steve – I did read the original thread. I am not trying to be a wise guy but I'm not sure you have read or understood my responses. I was very explicity addressing the possibility that tapchanging while energized had recently begun (perhaps the original poster was unaware or didn’t think that part relevant). I have made 3 previous posts this thread. I excerpted a portion of each one below for your reference:

“If you have an arcing style and have only recently begun to cycle taps while energized, that could be one possible explanation for what you are seeing. (Or it could certainly be a real deteriorating condition).”

“I feel pretty strongly the results listed would be normal for older arcing LTC's [/b]assuming it had just begun cycling while energized[/b]; and abnormal for newer non-arcing LTC's.”

“Based on mike's comment that ethylene rather than ethylene is the dominant gas, I now believe that an actual hotspot is a more likely explanation than my previous proposed explanation arcing type LTC. “


rraghunath - you had some comments about leaking between tank and LTC compartment. I assume all gas measurements here are in LTC compartment. But maybe deserves some clarification from Mike. Also is the Buchholz on the LTC or main tank?

Mike - fwiw I agree with Steve that lack of cycling is a known cause of coking. Many utilities whose LTC’s do not cycle over the full range require periodic manual cycling which includes forcing the LTC to excercize the reversing switch.

I had some direct experience with gassing of a Westinghouse type UVT LTC on two occasions 10+ years ago. I don’t remember the gases but I’m pretty sure ethylene was hightest. Levels I think got around 2,000ppm at highest. I also remember 8C rise from bottom to top of LTC tank by infrared. Found massive damage on one of the selector switch contacts. The first time we attributed it to infrequent cycling. We instituted periodic manual cycling requirements. Very similar symptoms recurred a few years later and again found severley degraded selector swtich contacts. We identified two additional possible causes: 1 – high moisture in LTC oil up to 75ppm; 2 – contact material. We upgraded our dessicant breather to a better style and installed silver-plated contacts from high-voltage engineering. Problem has not recurred since then.

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

 

[miketech59]

Good news . The unit has been removed from service and an attempt will be made to by-pass the load tap Changer compartment with all its assoiated problems .We will then degass and run back up . This should take about 7 days . I know that this time is insufficient to properly dry out the insulation of the transformer but we can get some megawatts until the next oppertiunity to dry .
I think the e-mail i sent to my immediate manager which was copied to the CEO did the trick .

 

[ScottyUK]

You can save a fair bit of time on the oil processing if you can run the processor between two tankers while you're on with the nuts and bolts job on the tapchanger. You'll still have to degas under vacuum once the oil is in, but your gas levels seem high enough to warrant an early start.

Glad your management listened - sometimes wish mine would! If you find the origin of the problem let us know so we can all have the chance to learn.


----------------------------------

One day my ship will come in.
But with my luck, I'll be at the airport!

 

[electricpete]

Sounds like a good solution if you don't need to switch taps while energized. Will you still have the capability to switch while off-line without draining the oil? Will be interested to see how well it works for you and how difficult to accomplish. We are in not too different a situation with an LTC that is not needed.

rraghunath brought up good questions about gases in the main tank. I am pretty sure the buchholz is in the main tank, right? So how do we explain the buchholz actuation? Maybe in your case the LTC is right in the main tank? Common conservator? If none of the above, maybe we suspect leak from LTC compartment to main compartment. That probably wouldn't be a huge problem but you may want to review rraghunath's comments to satisfy youself on the subject and see if any other investigation/repair of seals might be necessary when you have the unit drained.

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

 

[miketech59]

The spring contacts in the tap changer was replaced by a solid copper bar . Any output would be comntrolled by the exciter on the turbine .There was some sludge build up on parts of the contacts but the oil from the main tank keept leakoing into the drained tap changer section ( did not drain transformer ) no time to repair seals . The oil in the LTC was degassed and the combustible gases are well within their limit . Although the methane is rising i feel that this is a function of migration from main tank