Generator Differential Trip Mystery

[dinocu]

I was recently called out to a hydro site that had an 8 MW, 6900V unit trip on A,B,C differential (Basler BE1-87G).

Within the neutral and phase CT's is a tap to a Static Exciter Transformer. Outside of the CT's, but still on the generator side of the Generator breaker is a surge pack, arresters, and PT's. On the buss side of the Generator breaker is a breaker for the Station service transformer(50/51 protection), a breaker for the tie to the grid through a 6900-11500v step up xfmr, and also a feeder breaker feeding a small camp load and aprox 5 miles of line to a headgate(50/51 protection).

All loads were being fed from the grid while generator was offline.

Assuming a generator problem since it tripped on differential, I tested the generator, CT's, and CT loops back to the relay. All was good.

Attempts were made to sync on to the buss, however it tripped instantly on 87, all three phases. It was then decided to try shedding all the load and disconnecting from the grid and energizing just a dead buss. This worked, and we were able to energize station service, and sync to grid. However when the feed to the small camp load and 5 miles of line was energized, the unit tripped on 87 all three phases.

I guess my question is how could a fault on this line trip all the way back to the differential protection of the generator? Also why were no problems apparent when this feeder was being energized from the grid.

Thanks for any ideas.

 

[slavag]

Hi.
1. Are somebody provided stability test to 87 (diff)protection on commissioning time?
2. Are this generator first time connected to grid?
3. If isn't first time, maybe is after some maintanace?
4. You need test also this relay with secondary injection.
Not so much options for this type of trip :
1. Reversed connection of star point of one of CT's set.
2. Relay fault.
3. one of CT's set with secondary circuit shorted.
Regards.
Slava

 

[dinocu]

No this unit was online when it tripped, has been in service for years.

I tested the Diff Relay and it tested fine, none of the CT's are shorted, loop checks were performed on all CT circuits, and the CT's were ratioed and sauration tests performed.

 

[slavag]

Yeh, it start as mystery.
Try check this relay directly from CT's.
Something starnge or I'm missing something.
Are You check loop also with megger?
Regards.
Slava

 

[slavag]

About line trip.
If CT's not choiced right, possible unwanted trip by outzone fault. What is CT size?
Regards.
Slava

 

[dinocu]

Ct's on the differential zone are 800:5, relay set at 0.4.

CT's on the feeder are 200:5, relay set at 1.5 amps sec trip level.

 

[521AB]

Is it a numerical relay or not?
I am looking for internal disturbance recorder information. Do you have any?

 

[slavag]

Hi.
It's IEEE type or IEC type? You wrote A,B,C and PT, I think you at the ANSI land. What is CT: C100, C200, C800?
What is a distance between CT and protection panel, what is a wiring size?
Setting is 8%, is OK.
Regards.
Slava

 

[GTstartup]

I guess my question is how could a fault on this line trip all the way back to the differential protection of the generator? Also why were no problems apparent when this feeder was being energized from the grid.

Seems to me that it's not that there is a fault on this remote line but when it is connected there is enough current to exceed the relay setting with one set of CTs not seeing any current. You could apply slightly less load (maybe station service) and measure all CT currents and phase angles. If they are all equal with correct phase angle then I would say, since you already tested the relay, you might have a generator internal fault

 

[GTstartup]

But my guess is that if you do measure the CT currents you will find some wiring or CT problem.

 

[dinocu]

This relay does not have event recording capability, the Diff CT's are C100, the distance between the CT's and protection panel is ~10 meters, and the CT wire size is 10AWG.

 

[dinocu]

Hey GT

The unit is now running with rated load on it. It was just when this line was in the circuit that the trip occured.

 

[GTstartup]

Hi Dino
Well in that case, you're right it should not have tripped when you connected the remote load. Still, if you attempt to do it again maybe you could connect the load in increments while measuring CT currents

 

[GTstartup]

BTW... sucks to work on Sunday doesn't it?

 

[dinocu]

Well I was able to fly out yesterday when we got it back on line, but off again tomorrow morning!

 

[jghrist]

Attempts were made to sync on to the buss, however it tripped instantly on 87, all three phases. It was then decided to try shedding all the load and disconnecting from the grid and energizing just a dead buss. This worked, and we were able to energize station service, and sync to grid. However when the feed to the small camp load and 5 miles of line was energized, the unit tripped on 87 all three phases.

When the unit tripped, was the camp load still energized from the grid? If so, what was the load? If not, can it be energized from the grid with the unit off line?

Can you determine whether there was a fault on the camp line or what the load current was?

Maybe there was fault current or inrush that was sufficient to cause a trip because of CT wiring errors.

 

[dinocu]

Yes when the unit tripped, the camp feed stayed energized via the grid. The load on the feeder was less than 50 kW. There is overcurrent protection on this feeder, but nothing operated.

 

[slavag]

C100 is about 25VA 10P20 ( sorry, I must translate it for myself),you must check Basler's requerements.
It seems as CT errors.
Regards.
Slava

 

[slavag]

Sorry, I back again after speedy read of Basler manual.
with min setting 0.4A you must check CT performance.

 

[dinocu]

Hi Slava

Thanks for the input, I did test the ratio and saturation curves on the CT's, with no apparent problems.