Intermittent Generator Differential Fault

[BigJohn1]

I have a 480V 1MW utility-paralleled generator that is tripping on phase differential.

It will go months between events where it operates fine at all levels of load, but then a couple trips will happen in close succession. There doesn't appear to be any common factor between trips events.

The generator protection is always picking up ~200A of differential between the generator neutral and the leads on "C" phase. It sounds like a ground-fault, right? The problem is that this unit has a high-impedance ground with 59N and this element has never picked up. So if there is a differential, where is the current going?

Additionally, I've done insulation-resistance and winding surge testing on the generator and the leads at 2kV and everything looked fantastic, where if there was a ground-fault I would really expect it to flush out at 4X the RMS operating voltage.

Unless there's a problem with the 59N circuit or the differential CTs, I'm sort of at a loss to explain this. And if it were a metering problem, I wouldn't tend to expect it to be intermittent.

Any ideas?

 

[Hoxton]

You are lucky to have differential protection on this relatively small unit!

I think that you have it in the last paragraph....

Check the CT and relay wiring connections, look for a loose screw.

Secondary injection of the relay circuit may also be useful, check that the relay operates similarly on all three phases.

we may get more suggestions from others

 

[DTR2011]

You may also want to check if the CT's are connected correctly. For example, if you have multi ratio CT's, they need to be connected to the same taps, for equal ratios. There may be a different CT. An easy way to deck (deenergized primary circuit & secondary disconnected) is to apply some convenient AC voltage, say 100V. Measure what you have applied, then also check all of the other taps (X2-X5, X3-X5, ETC). If the CT's dont match (A,B,C Phase and Neutral end) in voltage measurements, based on the connections, this may be a reason for 200A constantly appearing on the C Phase.

 

[motorspert]

Hi
A quick check (if you are unable to complete the checks suggested by DTR2011 or Hoxton) would be to swap the CT's between B and C phases. If the 'fault' then swaps to B phase, you know its the CT's, if it stays on A phase then it's either the winding (unlikely from your other checks) or the relay.

But checks and secondary injection would be the ideal thing to do

 

[BigJohn1]

Visually the CTs were identical and the connections appeared mechanically sound for what that's worth. Obviously this isn't the same as the results of a primary injection test, which may be the next solution.

I like the idea of rolling the CTs between phases because that would be a pretty darn effective test. But I'm concerned about possible repercussions of changing the current phase-sequence, I honestly don't know what that will do to the metering or any other protective functions?

 

[motorspert]

Hi John,
Goof question, and I'm not 100% sure, but normally diff CTs just feed a differential relay and phasing shouldn't matter. But for a multifunction relay, I would imagine they should be ok. If you swap all 6, the phase sequence will be the same.

 

[BigJohn1]

If it's any help, here's a waveform capture from the relay. I'm definitely thinking CT testing, this looks for all the world to me like one of the transformers is saturating:

Thoughts?

 

[catserveng]

Had a site that looked similar last year. The neutral side CT's mounted in the generator had loose hardware, were cracked and damaged.

Replaced and remounted all three CT's to repair.

This is from an SR489, correct?

Mike L.

 

[panter]

Big John,
Maybe you can with some conductor of the corresponding section bridge generator winding phase by phase and let the test current through current transformers of same phase . If you have difference current in the relay then definitely there is a problem with some current transformer .
Good luck

 

[bb3030]

Look at your C phase neutral waveform. It is distorted. Neutral current from ct should cancel phase current from ct (180 degrees out of phase)

 

[pwrengrds]

It looks like phase C neutral CT certainly has problems. Testing would likely reveal if it's failed.

Keep in mind that most of the protection functions are from the neutral CT's, not the high side CT's.