Generator Stator Damage incident 1

[elham02]

Hello,

Background:

During a scheduled unit SD, the generator rotor was taken out and the stator goes under rewedging and retaining rings replacement process. In the mean while the old protection panel was replaced with a new numerical one.

Incident:

When we synchronized a 153MVA generator to the grid, the generator tripped on split phase protection. The generator have a double winding each winding is 20 turns that is there will be 380V induced voltage for each turn for 13.8KV LL terminal voltage. If a single turn is shorted approximately 570A Circulating current will flow between the Paralleled windings. The split phase CTs are 3500:5 A and the setting is 40% inst.
What I am puzzled about is that, all three phases of the generator is showing a split phase differential current.

Is it normal or just something terrible gone wrong inside the generator?

After the incident the hire-ups come to the site and a lot of bla bla bla and at the end…the really mull operated or as they said too sensitive and they gave the clearance for machine rooling,

Once filed breaker is closed, I observed the differential current in the split phase input, it was below the pick up and for all phases. The chief decided to add a time delay of 60ms on the split phase element. At this moment there was abnormal humming sound from the generator, after the time delay the clearance have been given for synchronizing…..and then with a thundering noise the generator tripped but this time, the Buchholz relay of the main transformer operated on the second stage. Generator differential and overall differential element operated, when they open one end to inspect the generator, cracks have been found on the stator winding insulation.

My guess about what happened:

I think the problem exist before the first synchronizing attempt since the rotor was maintained by one contractor and the stator was done by another contractor.

The buchholze relay operation in the second time was because of the high inrush currents from the grid toward the generator, according to the relay records it reaches up to 60KA.


Any thoughts about what happened

Is it really possible to have split phase differential for all phases?

thanks

 

[davidbeach]

How about posting COMTRADE files of the relay events for the differential trip. Probably just a setting/configuration problem.

 

[elham02]

Dear David
This is not the first unit. 10 units are in-service without any problem.

And if it was setting problem how do you explain the Buchholze relay operation?


Posting the events… sure…but I think the osc.is better since it will show the currents and voltages waveforms. There was a severe voltage dip at all phases…you will need EenerVista to view the files…I shall post them in the coming days. I whish everybody enjoy working on this case as I do.

elham

 

[davidbeach]

By relay event I did mean the oscilography. More people will be able to open a COMTRADE file than a manufacturer specific format; I'll look at a COMTRADE file but not an EnerVista file for instance. If GE can't save as COMTRADE that would be an additional reason that I'm not much impressed with GE relays. There is much that could be learned from that file. Although at this point, rereading the first post plus your second post, it sounds a lot like an out of phase closing. Differential tripped because one set of CTs saturated differently than the other set of CTs. But an event file would help determine that also.

 

[elham02]

Thanks for your thoughts

Out of phase closing, I have thought about this, will the out of phase closing cause a circulating current in the double winding of each phase?

Regarding GE, I would prefer to stay neutral in my opinion about all manufacturers...

And for the synchronizing scheme used, I have no single doubt about the circuit; it is working fine for all other 15 units what shall cause it to fail in this unit. You may say loss connection or relay setting. We have performed the test for the 25 element in the G30 relay and the result was according to the settings and to sort out the possibility of wiring problem, a voltage is applied from the field at the generator PT terminal all the way to the relay. The PT input is also monitored via the PT fuse fail element.

Thanks

 

[waross]

From the information given so far I would investigate the wedges to see if the wedges were driven in so as to short and/or ground a turn on each phase. Also check the wiring on the synchroscope. The connections may have been changed or rolled after the service work.
Each time I reread your description of the event I think:
"Winding damage done by over driving wedges".
More damage done by first and second sync attempt. Possible sync angle error also.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Pikoplatt]

What were your Generator Stator Winding Insulation Resistance and Surge Test results prior to and after the event? Were generator stator cables disconnected & re-connected during the repair? If yes, was synch check done across Generator Breaker terminals with breaker removed from cell, using a kV meter/hot sticks, while observing the synchroscope?

 

[elham02]

Hi all,
Sorry for the delay, I could not access any of the data, the case is under investigation by the OEM.

Mr.Pikoplatt

The IR values are in order before and after.
You mean by stator cables the termination? No it was not removed.

Generator Breaker: what do you mean by removing the generator from the Cell? The Synchronization is done across the 230KV breaker that is after the main transformer. I think you mean some systems where there are two different breakers: Main (HV) and Generator (after the generator).

thanks for your thoughts

Back to the case;

After removing the rotor, the extent of the damage become clearer, the cracks in the glass tape insulation is observed only at the cross over areas. The peeling of the varnish paint was all over the windings. Also some cracks in the strain blocks were found.

And one more issue, in case of out of phase closing the maximum current that will flow from the grid toward the generator is approximately 50% of the rated machine current. The current observed in the second attempt was almost more than 10 times the rated machine current.

Before closing there was already a current circulating in the split phase arrangement less than the pick up value of the element. After closing the element immediately operated.

In any case, let us wait for the OEM investigation results, there are some pictures for the damage, if get my hand on them I will upload them.
Thanks

 

[ScottyUK]

If you are able to do so please post the OEM's findings, I think they will be quite educational for us. Thanks for the update.


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

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

 

[racookpe1978]

How long will the re-repairs be? Can you tell us?

 

[waross]

"And one more issue, in case of out of phase closing the maximum current that will flow from the grid toward the generator is approximately 50% of the rated machine current."
I find this hard to accept.
"The current observed in the second attempt was almost more than 10 times the rated machine current."
I have seen the generator breaker trip on instantaneous (10 times normal rated current) with phase angle error on closing on smaller sets. As the phase angle error increases the current increases.
Specifically on a 350 kW set with a miswired synchroscope:
30 degree error, successful closing.
More than 30 degree error, breaker trips on 10 time instantaneous.
Much more than 30 degree error, coupling shears the key before the breaker trips.

"There was a severe voltage dip at all phases."
Consistent with out of phase closing.

There may be more than one issue.
Possibly there was damage done to the stator that resulted in differential currents.
Possibly the synchroscope was miswired, possibly by the same crew that damaged the stator.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[edison123]

In the mean while the old protection panel was replaced with a new numerical one.

Did the panel include a new synchroscope ? Possible it was wrongly wired creating a phase angle error ?

Events and outcomes indicate out of phase synch. I would check the differential CT's too for any error or internal failure.

 

[Pikoplatt]

Do you close the breaker based on a synchroscope indication or based on a synch check relay (25) interlock in the breaker close circuit?

 

[elham02]

Thanks mr.Waross for your contribution, what I was trying to point out is as follows:

There are two situation of out of phase synchronization; the generator ether leads or lags the grid. The synchronizing circuits differ, some circuits if you reverse two wires you will end up with an error of multiple of 60 degrees. However, some other circuit's uses single phase from the grid and single phase from the generator side, here a reversal between the phase and the reference ground means 180 degree phase shift.

Mr.Waross, you are so lucky to witness an out of phase closing and observe the effects of such case. What I have said about the maximum current from the grid is 50% is based on dynamic simulation of the system using MATLAB. In actuality, the current in the first few cycles will be more than or almost equal to 10 pu. In case the generator is lagging, the stator current will decrease to the value corresponding to the prime mover input. The time of the decay depends on the phase angle, the larger the more it takes to be back to normal, generator protection should trip the breaker in this situation. However, in case of generator leading the grid, the stator current will shoot up in first few cycles up to 10 or almost equal to 10 pu, but will decay to 50% of the rated machine current. Machine rotor will experience change in acceleration from (Zero Value = fixed speed) in either lag or lead cases but it will be more severe in the leading case (Rotor will accelerate and the acceleration never back to zero at least for the entire simulation time).

I don’t know wither my simulation is right or wrong, but what I find hard to explain is that, as Mr.Waross point out, the 50 element shall operate but in this case the Split Phase Differential (50SP) is operating in the first attempt. In the first attempt the currents is almost 2 to 3 pu which is quite abnormal by itself. In the second attempt, the 50SP was delayed (based on what I don’t know) and the 10 times story come in picture.

Mr.Pickoplatt, the breaker is closed based on the synchro-check relay output contact in the closing circuit.

 

[esee135]

...little bit of a late reply, however, yes you can have split phase current in all three phases simultaneous, and yes I did experience this first hand.

My issue was sourced to uneven air gap and it is a function dependent on the physical winding circuit arrangement and parallel path connections (paralleled circuits of different voltages caused from the air gap flux difference).

I experienced 50% of the rated circuit (not rated stator current) current at open circuit where Is should be zero. I also experienced 100% rated circuit current at partial load. I wrote a report which shows the metering of say T1A 180 degrees out of phase in comparison to T1B.

 

[QBplanner]

I am not sure Matlab will do the job. It depends on how detail the matlab machine model is and other factors related.
Anyway, interesting case and keep posting