Operating Large Power Transformers in Very Cold Weather 7

[freethase]

Hi All,

I'm researching how various power plants operate their large power transformers in extremely cold weather. Particularly interested in transformers that are rated in the hundreds of MW and 1000 MW range.

With extremely cold weather, issues arise with oil temperatures becoming too low. This is especially true with OFAF types. I'm looking for:

[li]specific operating procedures[/li]
[li]when and how are cooling fans switched off (what oil or winding temps?)[/li]
[li]how are transformer temperatures monitored?[/li]
[li]user experiences[/li]
[li]transformer user manuals where the issue of very cold operating temps is addressed[/li]
[li]etc. (any insight into the topic is appreciated)[/li]

Thanks!

 

[cranky108]

I would hope you know that it dosen't get that cold in much of world. And if the plant is operating, the transformers won't get cold.
If it gets cold enough to freeze the cooling fans, then we probally don't need them anyway.

 

[freethase]

But it does get cold enough to bring oil temps below what they should be. For example, we have procedures in place to stop some of the cooling fans if oil temps fall below 43 C. If this happens AND if winding temperatures indicate < 67.5 C we then stop "automatic" operation of fans (as they should not all be running when it's that cold).

I realize these procedures vary from transformer to transformer, but I'm interested in how others deal with this and I'm having a hard time finding reference material online.

 

[111R]

What is your ambient temperature?

 

[freethase]

Lately we've had nights where ambient temp drops to around -7 F and -20 F.

I'm finding a lot of information on operating procedures for high temps but there doesn't seem to be much in the way of cold temps.

 

[ScottyUK]

In the UK where we get moderately cold winters - we get down to -10°C fairly regularly, but rarely down to -20°C - I'm not aware of any particular problems with cooler banks, nor of any special efforts to limit the capability of the coolers in cold weather. The additional cooling from a low ambient is usually beneficial on a GSU transformer where the top oil temperature is often high. At the very least it saves running hours on the fans and/or pumps if the weather is cold.

 

[marks1080]

Typically operators like cold temps for their transformers... Conversly, air blast breakers are scary things to operate when the temp gets too cold.

If you have a transformer that has been out of service in the cold and you want to bring it online there will be some amount of time (24hrs?) that you want the unit on pot (off-load) to warm the bank up, before you put it fully in service.

 

[freethase]

The main issue is that when the oil temps fall below 40 C, there's the potential for static electrification, which can completely ruin a transformer.

I can see how this wouldn't be an issue for most of the world, but for areas where it gets really cold, there have to be some documented procedures in place to keep the oil within safe limits.

 

[cranky108]

Marks, you still have airblast breakers? Most companies have switched to gas.

We do not have static problems with transformers. However we use arresters for lightning, and most power transformwers have a low tempeture that the fans turn off (it's automatic from the factory that way). We also ground the case of the transformer, and the windings are connected to some voltage, with a ground reference (except the few delta connections). And we do have some high isocronic levels, and we don't see static problems. But as I said almost everything has a ground reference.

 

[DRWeig]

Here's a thought from GE, not sure if it's applicable: Old GE Book



Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[DRWeig]

And here's a nice IEEE Paper.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[ScottyUK]

Interesting to see the problems those with more extreme climates than ours have to deal with. Thanks.

 

[freethase]

Appreciate the responses everyone. This is probably going to take a little more digging on my part . DRWeig, the second one was a pretty interesting read--thanks for that.

 

[Sn00ze]

Cooling and Heating schems of a 18/24/30MVA designed to operate in -65 to 45 Celsius (ambient temperature)

Our XFMERs here usually have Fans/Heaters running continuously and some with a thermostat as the oil needs to be a certain temp for it to operate. Hope it helps.

Cheers

 

[Sn00ze]

Heating

 

[marks1080]

cranky108:

Yep. Quite a few 500kV air blast breakers still in service. We have to man the sites when the temp gets too cold. Slowly switching to SF6...

 

[JaPaBo]

Can anyone provide more information on this comment from freethase?

 

[crshears]

Hi there freethase, we have to be more PC than your utility, so we are obliged to staff our air blast breaker stations in cold weather ;o)

And SF6 breakers don't handle the cold at all well; in our jurisdiction it seems nigh impossible during periods of sustained cold to provide enough heat to prevent gas liquefaction...overnight we once had all of the SF6 breakers in a switchyard trip themselves out of service one at a time on low SF6 density, interrupting a number of radial ckts and the customer load that was on them, and we were powerless to do anything about it until the daylight hours when the sun warmed the breakers, the alarms and trips cleared, and we were able to return the breakers to service.

But I digress.

Pretty well all of our transformers, some as high as 750 MVA FCR, typically have all fans shut down by the time the oil temp drops to ~ +40 C.

Most of our stepdown transformers are equipped with under load tap changers for output voltage regulation. One sometimes very unwelcome protective feature on these banks is that the ULTC's lock out if the oil temp gets too cold - - which is understandable, since you wouldn't want taps moving when they're submerged in insulating oil that's halfway to being treacle.

We have specific instructions to follow during very cold weather; a transformer low oil temperature alarm is one thing, but a lightly loaded bank may annunciate with low conservator oil level due to shrinkage, which means rousing some poor soul out of bed in the wee hours to confirm the bank hasn't developed an oil leak.

Some of us old guys will deliberately skew the tap positions on paralleled transformers to deliberately produce circulating currents just for the warming effect; often enough this is enough to clear the low level alarm.

Anecdote:

At a large hydraulic generating station where I worked we once had an 86 MVA single-phase bank of 135 kV L-N primary develop a fault during a particularly severe and sustained period of cold weather. Maintenance crews came in and swapped it out for our spare bank. When it came time to place the spare forced oil, water cooled transformer in service, our supervisor obstinately insisted that the cooling water had to be in service on the bank since an instruction stipulated that this type of transformer could not be placed in service without cooling water.

Us lowly operators protested that since the oil in the transformer was at about -20 deg C, the first thing that would happen was that the water side of the heat exchangers would freeze solid since the oil would cool the water, not the other way around, and that it would be better to leave the cooling water out of service until the oil temperature was on the plus side of the freezing point of water...but we were overruled, and being subject to discipline if we did not comply, we did as directed.

Result? Both HXRs pronmptly froze solid; we knew this was the case because return cooling water from the HXRs only flowed into the tundishes inside the plant for about five minutes. Since maintenence personnel were still on site, we had them remove the HXRs from service one at a time and melt the ice out of the water boxes with a steam cleaner [there was no need to de-ice the tubes, because without any cooling water in service that transformer's share of the 250 MW of generation being stepped up to system voltage soon enough melted the ice in the tubes] ...but this took time, and the oil temperature in the transformer had risen to between 75 and 80 deg C before we managed to get both HXRs back in service.

We operators figured it was our supervisor who should have been disciplined...

A sage gentleman acquaintance of mine told me of an axiom he had once heard early in life and had applied often during his thirty years as a police officer. The axiom was: "Instructions are written for the guidance of the wise...and for the blind obedience of all others."

CR

 

[freethase]

crshears, thanks... that's exactly what I was wondering. So, you guys would turn all fans off before oil temps hit 40 C. Your comment about how they'd "skew the tap positions," on parallel transformers, that's really interesting. The kinda stuff that only comes from experience, I take it? (heck, is that safe?).


JaPaBo-- so, "static electrification" happens (and has happened quite a bit) when a potential builds up across the oil (for example, the oil nearest the coil builds up a positive charge and the oil near the casing builds up a negative). When the potential difference becomes too great, a path for current can be created. It can be affected or caused by a variety of things, including contaminants, oil pump speed, temperature, and other things that affect the dielectric breakdown of the oil. I've come across a few papers on it on the web describing how it happens. But basically, worst case scenario, it can cause a short between phases.

 

[crshears]

Hi, freethase, I can't speak to static electrification of oil, never having heard of it; but I do wonder, can this happen quickly enough @ 50 or 60 Hz to cause such severe damage? The only scenario in which I can conceive of this happening would be if an electrically 'floating,' viz., ungrounded internal component were to build up a standing DC electrostatic charge...then again, I had best leave discussions of this topic to those who know something about it.

As to skewing tapchangers for the heating effect: in my view it isn't intrinsically unsafe, per se, but it is a calculated risk. It's not a good idea when there are high winds about, or ice accretion is ongoing, or when other circumstances prevail that could interrupt one of the two supplies in a dual element spot network configuration. Deliberately putting a large spread on the taps does after all force large amounts of reactive power to flow into that bus from one HV circuit and out through the windings of the other transformer into the companion HV circuit, and if a contingency occurs while such a spread exists there could be a minute or two of over- or under-voltage on the LT bus or busses until the operator can re-adjust the taps.

But if it's a calm, clear night with temperatures getting down into the -25 dec C range and a low transformer oil annunciation is received, it can be surprisingly effective to create some artificial circulating current in the transformer windings and generate just that little extra bit of warmth needed to clear the alarm.

CR