Re-commissioning old oil filled transformers after 20 years - Is there anything to look out for? 1

[bdn2004]

We are planning to re-use some old 10,000kVA oil filled transformers. The transformers have not been energized for probably 20 years.
They are annually gas-oil tested and have shown no issues. But that is the only testing done, as they aren't connected to a live bus.

There are many other transformers like these on site of the same vintage that are in service right now.

Is there anything we should be worried about? additional recommended testing? Would you change out the oil?

 

[shamsdebout]

Based on what I saw, a transient and low-frequency dielectric tests would be appropriate due to the inactivity of the described unit.

I would think a megger test would also be appropriate.

I would think testing the connections and conducting a turns ratio test would be appropriate.

It is good the transformer oil as been tested annually.

 

[bdn2004]

Shams,

That's a good answer....although I wonder about the turns ratio test. Would that change?
How about the oil ? Would you change that out even though it tests good?

These are still connected to these old GE IA electromechanical relays that we want to re-use. These drift out of calibration right?

I'm trying to capture all potential costs in a realistic estimate.

 

[oldfieldguy]

No, the turns ratio should NOT change, but still I'd have a reputable testing agency perform a complete battery of tests: Power factor, turns ratio, winding resistance, oil samples (standard physical properties and dissolved gas analysis).

If the tests return acceptable results, I'd put the transformer into service, monitor it closely, and resample the oil in a couple of weeks or thirty days of normal service.

I would not be changing the oil in any likelihood. If it tests poorly, several companies can remediate, i.e., dry, degas, add inhibitor, etc.

The electromechanical relays, with a little servicing and testing, will serve as well as they ever did, but there are many new products that will make management of your transformer much easier. The costs of these things has come down sharply. Upgrading your protection is not too expensive.

old field guy

 

[shamsdebout]

bdn,

Based on what I understand the voltage ratio between any two windings should be within + or -0.5% of the value indicated on the nameplate. I wasn't sure of what environment the transformer was in or if there was some fault on it while it was in service some years ago. I certainly think the windings could have degraded may due to an earlier fault and there could have been a partially shorted turn. Doing a turns ratio test would give us some insight.

I am not too up to speed on gas-in-oil analyses or oil test data. On one hand if the test have been conducted annually then by comparing the results we should be able to trust the result of the test. I am not familiar with the standards for the oil test data as mentioned in ASTM D1816-12. I imagine the service life to be 25-35 years. So I would think there is still some life left out of the transformer and if the oil checks out good with the lab conducting the test then it should be ok.

I agree, the relays may probably need to be adjusted. I would think those would probably better served being replaced though, who knows which piece could be defective, that is of course assuming they are really old.

 

[waross]

Is your oil tank sealed? If there is a chance that the transformer could draw in and expel air as the temperature changed and the oil expanded and contracted I would be concerned about the possibility of moisture in both the oil and in the paper insulation. This is sometimes an issue with distribution transformers.
If there is a possibility of air changes above the oil then do some research on moisture in transformers. It has been a few years since I had a concern with moisture issues and my memory is a little fuzzy.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rterickson]

Further to Bill's advice, I would get fresh oil samples and have them tested for moisture in oil before energizing transformers that have been sitting that long.

 

[prc]

The major question is how is the oil preservation system of thetransformer-Conservator with silicagel breather? Or air cell? or Inert air? Are there any leaking joints in transformer (then chance for water entry is very high)How the silicagel or inertair was maintained?

Most prominent threat for such non energised transformers is large amount of moisture accumulation in insulation of transformer.You can find it out from two tests- do the power test on winding insulation. It shall be less than 0.5 %. Do the relative saturation of water in oil samples from tarnsformer.It shall be less than 20 %. Even if values are Ok, better to do a site dry out by hot oil circulation for 12 hours-drain out oil-fill with dry air-measure dew point -apply vacuum for 12 hours- oil fill- hot oil circulation cycles -2-3 times depending on water content.

Check the condition of bushings by power factor test. Operate tap changer one end to other end several times to wipe the contacts and make sure mechanical moving parts are working well.No need to replace oil.

Do the conventional commisioning tests before energising- ratio and rsistance at all taps, low voltage excitation current measurement,insulation resistance, oil tests and reference DGA.

 

[stevenal]

I think PRC meant "power factor test" on the winding insulation rather than "power test." I fail to see the wisdom of gas-in oil testing on a deenergized unit. Where would the combustible gases come from? Oil dielectric testing is the one to do, since moisture would be the biggest concern.

 

[racookpe1978]

I would torque check the fasteners and mechanical features: This thing has been sitting outside in what basically amounts to a parking lot for that long with no attention. Look for rust and corrosion, cleanliness, and general condition BEFORE the meggar tests and hookup. Actually, do the first meggar before you move it. Restore the wheel and tracks first: Standard things (grease, paint, remove rust, etc. since the "violence" of breaking it free from the old rust and debris under and around the tracks may crack old rubber insulation and crack insulators made brittle by tens of years of exposure to UV rays and heat cycles.

Restore and clean up the original tracks and their switches as well - you don't want this thing shaken to pieces as you move it around the transformer yard crushing gravel and rocks and concrete rather than rolling. Check that all tracks are actually still level: Areas of one 50 year old power plant in FL were 6 and 8 inches below their "original built plane" when we made mods and extensions recently near foundations holding up the 40,000 volt primary ducts out there.

Open and look into the "vented" parts of the circulation piping and coolant tanks: the areas above the former liquid level will have deteriorated (rusted, flaked off paint, been corroded through or been deposited with coolant distillates) much differently than those areas assumed to be covered with liquid continuously.

 

[ScottyUK]

After this length of time you should consider using a vacuum processor to extract moisture from the core and winding assembly. For a transformer this size it may be more economical to bring a mobile processor to the substation with a crew and generator. I suspect you will have enough moisture there to make a new charge of oil worthwhile.

 

[prc]

Stevanal,you are right,it is insulation power factor test.It is true, there is not much sense in doing DGA after a hot oil circulation. But some times slight content of acetylene noticed later may cause concern.So it is a usual practice to take bench mark DGA measurements just before energisation.Oil dielectric test may not detect slight moisture content(more sensitive to particles)But water content measurement (ppm) as a percentage of maximum water soulbility at the temperature of oil sample will give an indication of moisture level in paper.

Definitely you have to get a mobile processor.At least 50 litres of water can be expected inside that transformer after 20 years of idling.It should be remembered that out of 50 litres of water,only 0.5 litre will be in oil and balance will be in paper and pressboard. 98-99% of water will be in paper and balance in oil. Water content will be more in an idling transformer than in an energised unit.Hot oil circulation by mobile processor will remove only 0.5 litres of water.Water from paper will come out very slowly may take several weeks.Hence the resaon for the dry out process explained in my earlier response.No point in changing the oil as water is mainly in paper.

 

[bdn2004]

Attached is a picture of the transformer(s) we are going to put back into service (on the right).

 

[bdn2004]

One of the things we are concerned about is the fans...They haven't been energized for a long time. This is a midwest plant and it gets hot in the summer.

 

[shamsdebout]

Thanks for the picture. They don't seem to look too bad to me.

 

[Overvoltage]

Couldn't you just take a power source out there (even a portable generator) and test the fans?