Cooling Fans on MV Substation Transformers

[johnnyball]

Is there any advantage gained by running the cooling fans on a distribution substation transformer that is normally loaded on peak to approximately 70% of the base rating? This particular unit has two stages of fan forced air cooling with ratings of 18/24/36 MVA respectively. To my knowledge 17 MVA was the most ever put on this unit, and that was during a contingency.

 

[ScottyUK]

Cooler oil and cooler insulation lasts longer. Whether the extension in life is worthwhile against the increased maintenance costs and the increased overall losses from running the fans is arguable. Personally I would say 'no' on gut feel in the absence of any firm numbers to base the decision on.


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

If we learn from our mistakes I'm getting a great education!

 

[jghrist]

I agree with Scotty. The gain in life of operating a transformer at a lower than design temperature is not as much as the loss in life of operating above the design temperature. I don't have any firm numbers to go on either, but if you look at the cost of running the fans full time, you may find it difficult to justify.

 

[Zogzog]

Just put them in auto, best way to go unless you are having high temps (Which could be caused by blocked radiator tubes), then maybe forced air can be justified.

 

[rcw retired EE]

I had old rectifier transformers running above their original rating. Our oil temperature measurements and the winding resistance measurements indicated that the kW to run the fans was more than offset by the kW saved by the decrease in winding resistance. We ran fans all of the time. It would be unusual to find a more modern transformer pushed to those limits where the cooling has a direct beneficial effect on the losses.

One problem was the 64G protection was driven by a CT that connected the transformer case to ground. The CT could not tell the difference between 13.8kV fault current and a 208V fan motor failure. A 1/2 HP fan motor ground fault would trip the 50MW rectifer station.

As the other experts recommend - just run in auto.

 

[prc]

This is an interesting topic.First of all as per standards,there is a relative gain in life when temperature falls down to the same extent as loss of life.As example,if winding hot spot temperature goes up by 8C more than reference temperature of 98C,then rate of ageing goes up by 8 times.When temperature goes down by 8C ,ageing goes down by 1/8 times.(Table 2 of IEC 60076-7 Loading Guide for Transformers.)So theoretically,working at lower temperature will enhance the life of insulation.


But running the fan throughout ie during ON cooling is not a very good idea.Rcwilson has brought in a new angle.The drop in copper loss due to temp drop may off set the cooler power consumption.Actual checking on some trfs does not justify such a line.One thing-though I2R loss goes down with temperature,stray losses goes up with lower temperature.At ON rating of say 70%load, the copper loss will be only half at full load.For a temp drop of 10 C due to fans,the reduction in loss will be only half of 3% of full load loss.In many cases cooler loss will be more than that saving.

A new trend is coming up-to use variable frequency drive for coolers ( some type of intelligent coolers)so that same oil temperature is amintained by changing the cooler speed to save energy.

Regarding 64 G Tank current protection.I have seen this only in France and other French influeced regions.But in such cases, it is a must that coolers,marshalling boxes and OLTC drive box shall be kept insulated from main tank and separately earthed to avoid mal operation of relay from auxiliary power break down to earth.So it has to be assumed that the rectifier trnsformer cooler mounting was not designed correctly.

 

[rcw retired EE]

Good comments, PRC. Just to clarify on our use of fans. These 7.5 MVA rectifier transformers were already used when they were installed at our facility in 1953. They were built by Allis-Chalmers in the early 1940's. Fans were added during an upgrade around 1965 when the units were pushed to about 150% of their original design MVA, creating higher I2R losses. None of the transformers had failed or been rewound when the plant was shut down and scrapped in 1998.

It would be very unusual to find a modern transformer where the I2R losses would be as high.

The 64G's main function was to detect winding faults on the low voltage(800V) side that would be back fed from the other 7 rectifer units via the DC negative bus. A sister facility did have a LV winding failure and the 64G unit did clear the line quickly.