Generator Differential Trip

[emviana]

Hello. I need some help.

In my area we have a 13.8kV, 74.3 MVA generator (wye connection) connected to a 80 MVA 13.8/69 kV (delta/ground-wye) transformer. The 69 kV side of this transformer is connected to a bus fed by a 100MVA 230/69 kV (grounded wye/grounded wye). The 230 kV side of this transformer is connected into a bus that feed other four 230/69 kV transformers. The industry is fed by this 230 kV bus.

The problem is: when a 100MVA is energized on 230 kV bus, the generator is tripped by the differential relay.

Does anyone have an idea? Thanks a lot.

 

[rbulsara]

Yes, any one or more of the following:

Improper settings - (Most likely culprit- check for harmonic restraint function and/or sufficient delay to allow transformer transients to die down.)
Improper wiring
Faulty relay, CT
Actual fault

Under what condition the 230 kV bus is energized? Dead bus and transformers or with synchronization on? What is in the protected zone? Gen alone?, gen and one transformer?

You may want to minimize transformer inrush if possible by staggered energization of the transformers.

At the end of the day, you need an experienced professional at the site, working with the project team.

(PS: I am not an expert on 230 kV systems, but just thoughts based on experiences on medium voltage systems)


Rafiq Bulsara

http://www.srengineersct.com

 

[emviana]

We have two differential relays:

1) One protecting only generator (this one trips on energization of 100 MVA tranformer);

2) One protecting generator + 80 MVA transformer (this one doesn't trip).

230 kV bus is always energized. The problem occurs when we energize one 100 MVA transformer at no load.

I´ve studied relays parametrization. Until now, I didn't find any discrepance.

Thanks.

 

[electricpete]

If the CT's approach saturation during the transient, then small deviations of magnetic and burden parameters can cause bigger differences on CT output.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[rbulsara]

This is very akin to the problem I had with smaller size set up. See thread thread238-125456

Along with what electricpete said and the lack of harmonic restraint are the prime suspect. In our case we implemented (GE has a application bulletin on it as well) dual settings. The one with longer delay was used during energization and then quickly reverted to the proper settings with a reasonable delay.

You may want to review the traces of waveforms if available and consult the relay mfr.

I would also expect folks more experienced than I, to chime in. Letting them know the relay mfr and model may also help.

I am wondering if the starting the gen & transformer as one unit would help as it will energizing the transformer on reduced voltage as the gen voltage builds up.





Rafiq Bulsara

http://www.srengineersct.com

 

[edison123]

Have you checked for proper CT polarities ?

Muthu

http://www.edison.co.in

 

[slavag]

Hi.
Usually you have some level of 2-nd harmonics for avoid maloperation of differential relay in energazing mode.
Posible..
1. level of second harmonic in your setting is high >15%.
2. different saturation curve of CT on the geneerator side and on the 230kV transformer side.
3. low setting of differntial protection base current (<25-30%).

Please attach SLD with CT disposition and CT data ( include wiring size of CT connection and lenght), please attach a setting of differential protection ( type of differential relay will be help too).

What is a log data of fault, disturbance recorder data?

Best Regards.
Slava

 

[emviana]

I'm showing current oscilography at instant of trip.

S1 = side of generator's terminal.
S2 = side of generator's grounding.

The CT's are of equal specefication on two sides (4000-5A).

Thanks.

 

[electricpete]

It looks like phases close to 180 apart? Seems to point toward connection / polarity problem.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[edison123]

The phase angles seem screwy.

Slava - reversed polarity ?

Muthu

http://www.edison.co.in

 

[electricpete]

"phases close to 180 apart"
I meant close to 180 difference between the two locations for each phase.
(180 difference for A, 211 difference for B, 157 for C)

Was this data recorded by the relay itself or some other device fed from same CT's ?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[davidbeach]

No, phase angles seem just about right.

Phase A - 1370A of restraint current and 9.6A of operate current.
Phase B - 1090A of restraint current and 787.6A of operate current.
Phase C - 1215A of restraint current and 504.4A of operate current.

Where restraint current = (|I_t|+|I_n|)/2 and operate current = |I_t+I_n|

Probably was the B-phase differential that tripped.

 

[electricpete]

I guess there is a polarity convention on how the device reports polarity of the current. I can see the logic you're assuming currents which are same polarity in the system are reported as opposite polarity by the device, since we would connect so that normal same-polarity currents create opposing flux in an electromechanical relay.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Also, if there was reversal one would assume the differential would trip more than just during energization of transformer... (provided loading level is high enough)

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[davidbeach]

Differential is always the sum of the currents into (or occasionally out of) the protected zone. Kirchhoff says that sum has to be zero and it doesn't matter how many inputs (restraints) you have. If you want to try to set up a differential that subtracts one current from another to get to zero, that's not too difficult for two inputs, but gets complicated fast as the number of inputs goes up. Add all the currents and get to zero works for any number.

 

[electricpete]

I rather thought that polarity convention and Kirchoff's law were two different things. But thanks for explaining Kirchoff's law.

As far as I remember, 3-phase differential relays are usually three single phase element, unless there is some kind of pilot wire scheme. I might be wrong... it's been awhile.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Sorry if my comments was in any way sarcastic. David you were after all only replying to my comment /question in a logical manner (and I hope you will continue to do so anytime I say something that sounds like an uncertainty or a question). I will shut up now since as I said my memory of differentials is faded and David and others can contribute much more.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[rcw retired EE]

Good analysis, David. I'm guessing that Phase C tripped and that the relay is set up to inhibit only individual phases when 2nd harmonic exceeds the setting.

Phase C had 41% operate current (100% x 504/1215) but only 13.3% 2nd harmonic if I read the data correctly. Phase B had 35.1% 2nd harmonic and was probably inhibited if the setting was the typical 20%.

The solution is to change programming to inhibit all phases if any one phase sees 2nd harmonic or to drop the inhibit level to 10-12%.

 

[davidbeach]

Since it was the generator only differential it might not have the harmonic restraint/blocking typical of differential relays intended to include transformers in the zone of protection. Probably need emviana to tells us what kind (make and model) of relays he has in use, and which one tripped.

 

[burnt2x]

Can't get a clear picture with just limited data. David is correct. OP should provide more info: relay specs, specific element tripping, and probably more.