Buzzing Capacitor Bank

[Robert789]

We have a pole mounted, 200 KVAR per phase, grounded wye capacitor bank on a 12.47 kV rural distribution feeder. During a recent line patrol, it was reported that an audible hum or buzz could be heard around this bank that was 'similar to, but slightly louder than that of a medium sized three phase distribution transformer' (500 - 1000 KVA).

When removed from service, the hum remained until the last unit was taken off line. The crew performing the line patrol changed out each of the capacitors with new units, and put the bank back in service, with the noise immediately returning.

A later check of the removed units with a capacitance meter indicated no problems.

Our primary line is 336 ACSR, with a bank of 437 Amp voltage regulators that is approximately 3 miles upline of the capacitors. Loading is residential and agricultural, with no loads that I would consider unusual. This is the only capacitor bank that I know of on the circuit.

Noone here has ever before encountered a noisy capacitor bank, and an online search has turned up nothing. I thought of harmonics and resonance, but I would expect blown fuses and damaged capacitors to be the symptoms of either of these situations.

Any thoughts on this situation would be appreciated.

 

[busbar]

Any current in grounded-neutral conductor?

 

[buzzp]

This isn't the problem I wouldn't think since you replaced all caps anyway, but I had a similar experience only on a smaller scale. We checked them with a cap meter and found nothing. We put them back in the circuit and still had problems. Turns out a capacitor was breaking down under load. The morale of the story is a cap meter works fine for testing if you include a voltage test(apply rated voltage across terminals and see if it breaks down-we used a hi-pot tester for this since it has built in protection). Good luck.

 

[electricpete]

Possible corona at a terminations? (just a guess). In that case you might hear the highest noise when air is very humid or when dew is forming.

 

[Robert789]

Thanks for the responses, I still haven't figured out what might be happening. I've never encountered corona on 12.47 kV equipment before.

I had some amp readings taken at the location yesterday. It was a very mild afternoon, so loading was light. Without the capacitors, we found phase currents of 7,20,26 Amps, and current on the grounded-neutral of 11 Amps. With capacitors in service, we found upline readings of 29,31,38 and 13(neutral); and downline readings of 8,18,25 and 10 (neutral). About 10 minutes or so probably elapsed between the downline and upline readings.

We will probably just keep an eye out for blown fuses or bulging tanks, and leave the capacitors in service. The buzzing is not coinciding with an any apparent operational problems.

-Robert

 

[electricpete]

You're right, we don't see the same type of corona on open-air connections at 12kv that we see at higher voltages. But possibly if there is some kind of defect in the connection method it would arc/sputter across the intermittent contact? OK, it's a stretch.

Anyway, you might ask your infra-red guy to take a close look at it next time he's in the area to see what comes up.

Also... do you have anything you use to check for corona or leaks? For instance Ultraprobe UE2000 from UESystems ultrasonic device can help you localize exactly where the noise is comming from.

 

[Robert789]

Infra-red testing is done each fall, I will put this on the list of things to look at then. An ultrasonic device is a little too fancy for us to have. The buzzing does occur as soon as the first capacitor is placed into service, and continues until the last one is removed. It doesn't appear to be on any particular phase, the only common thing that these units would each be connected to is the neutral/ground.

 

[electricpete]

I'm not too familiar with capacitor configurations myself. Is it possible there is a reactor up there for limiting switching current?

 

[Robert789]

This bank has no reactor or switches. Just one capacitor and one expulsion fuse connected in series per phase. The pole hardware is an RUS unit C1 or similar (tangent three phase, pin type insulators).