Gas in oil analysis, high acetylene, how dangerous?

[PowerDawg]

Just received a dissolved gas in oil analysis on a 2000 kVA 25kV/480V three phase transformer. Acetylene is 102 ppm and total combustible gas is 212 ppm. Having another sample pulled tomorrow. How dangerous a transformer do I have? Should it be de-energized immediately? From what I have been reading on the internet, this looks real bad. Anyone with any experience with a transformer with that high an acetylene level?

 

[rcw retired EE]

We had a 13.8 kV auto transformer that had high levels like that. When it was untanked, we found a 12" ball of carbon around the manual tap change mechanism. Only 2 of the 10 fingers that gripped the tap point were making contact.

I would guess you have arcing inside.

 

[Marmite]

The first thing I would say is avoid jumping to any conclusions based on a single sample. Repeat the analysis, and if it confirms the initial sample was correct then look at the previous history of the transformer. When was the last sample taken? What was the result? If the transformer has been in service for a long time with a steady change in gas levels, then a sudden change should be taken seriously. The other important aspect to be recognised is that some hydrocarbon gases are present in the oil of most power transformers. Those which have been in service for some years and have been operated for considerable periods at or near to rated load can have levels of many tens, maybe even a hundred or more parts per million of some gases and still be healthy. Because ofthis, and because of the many other variable factors involved it is not generally possible to obtain an indication of the condition of a particular transformer, or of its life expectancy, simply by carrying out a dissolved gas analysis. The most reliable indications are those obtained when a dga history has been maintained and a step change in an established pattern is suddenly observed. The ratios of the dissolved hydrocarbon gases are often used to point to potential problems rather than absolute values in ppm. However, if you repeat your analysis and the acetylene evolution is increasing over time then you probably have an arcing fault.

Regards
Marmite

 

[OffTap2020]

If it retests AGAIN at 102 ppm of acetylene. Switch it out immediately.

You have a bomb on your hands.


OT

 

[PowerDawg]

We have tested again... acetylene is down to 85 ppm. I know its still very high but it is coming down. We have scheduled another test I believe next week to see if we can establish a downward trend. The plant that the transformer serves is planning some work that may force us to move the transformer in the future. However, we don't have a date for this plant work. We were hoping the transformer would make it until the other plant work... but we will change it out if it becomes necesssary.

 

[Eam1]

There is definetely arcing inside that transformer. Check the loading and correlate it to the dga being taken. Is the acetylene generation load dependent? If it is, you may have to limit the load on the bank until it is taken out of service for inspection, repaired or replaced.
Marmite above has very excellent points to take into account. Any past dga's on this bank? Check the pattern.
Still if you check IEEE dga guideline for DGA that bank is deteriorating/failing as we speak.

 

[electricpete]

Do you happen to have an LTC in compartment adjacent to the main tank?
We have had gases show up apparently due to small amounts of LTC oil leaking past barrier board into the main tank.

=====================================
(2B)+(2B)' ?

 

[PowerDawg]

As I've indicated at the beginning of this thread, this is a small 2000 kVA transformer. We usually do not pull dga on transformers this small. We had a small leak after someone made contact with one of the cooling fins last year. We pulled the dga after a request in response to a quote for repair. Otherwise, we would have not pulled the dga. Therefore, there is not LTC associated with this transformer. I also forgot to mention the total combustible gas has decreased from 212 to 188 in the second test. Could the initial high acetylene concentration be from an internal arc and the decrease in acetylene be an indication of no further arcing?

 

[VTer]

Try to find out the position of taps and if anyone change the taps while the transformer was enrgizied. Units this size typically have to be de-energized before the taps are changed.

"Throughout space there is energy. Is this energy static or kinetic! If static our hopes are in vain; if kinetic — and this we know it is, for certain — then it is a mere question of time when men will succeed in attaching their machinery to the very wheelwork of nature". – Nikola Tesla

 

[PowerDawg]

I'm not positive this transformer has taps, but if it does, I am certain that no one has changed taps (energized or otherwise) without myself or my employees knowing about it. I agree that if that were done that it would possibly create acetylene in the oil.

 

[matanuska11]

If the Total Combustible Gasses (TCG)% is higher than the Safe Handling Limit (SHL)% that unit should be taken offline, deenergized, and purged with nitrogen. Make sure you are not using anything that creates an arc; vehicle, pump, lighter and so forth. If so installed on the XFMR, the pressure-vacuum relieve valve maybe blocked. TJ H2B has a procedure for this issue, I would contact them or whomever does your DGA. Once the acetylene is verified no longer present, I beleive that is when you could perform the work and determine what caused it.
Just my opinion

 

[stevenal]

How exactly does one get from a percentage of gas dissolved in oil (TDG) to a percentage of gas in air (SHL)? There should be very little air in the transformer. If you expose the oil to air, some gassing might occur; but it would be a pretty low rate. I don't see how one percentage can be compared directly to another.

 

[matanuska11]

The blanket is the space between the oil and the top of the XFMR and varies per size(depth of head) and manufacturer. Could be 10" or 3", nevertheless, most gasses are lighter and will rise to the top in oil and this would be the compartment that accumulates the 'air' and technically exhaust it out. You will still find these gasses in the oil, they are preserved in it. A history of dga results can help identify what is taking issue with XFMR...Hydrogen created by corona discharge, Carbon dioxide & Carbon Monoxide overheating of paper insulation, Methane&Ethane - high heat, Acetylene- arcing. These are general guidelines of course. But overall, the specifications of the XFMR and past dga's will give you the percentages of TDG and SHL.

 

[stevenal]

When preserved by a gas blanket, nitrogen is used. SHL is not applicable since there is no unsafe value of combustible gas in N2. I've never seen SHL in a dga report. Which labs provide this, and how do they come up with a value to even report?

 

[Jebb]

There is a way to approximate the amount of gas in the head space of an xfmr based on the ppm of gas in the oil. The Ostwald Solubility Coefficient is used to determine the partitioning of gas between the oil and head space. I’m not a chemical engineer so I probably can’t properly explain how it works exactly. If you use the Lower Explosion Limit (LEL) for a particular gas in air multiplied by the Ostwald Solubility Coefficient you can determine when you might have a possible explosive situation based on the amount of dissolved gas in the oil. But the best way to determine if you have an explosive mixture is to measure using a combustible gas meter to measure the gases in the head space.

Some time ago we had people injured in an explosion and fire due to the accumulated combustible gas in a de-energized xfmr while preparing it for shipping to be sent out for repair. Luckily there were no serious injuries, just some dirty underwear. But this issue is not to be taken lightly….

 

[PowerDawg]

Update... We got the third oil analysis back last week. Acetylene is staying pretty steady at 89 ppm. We've decided to change out the transformer. We are looking at getting it changed sometime next week. Thanks for everyone's input.