Fire in VCBs: Faulty Excitation System or Manufacturing Defect?

[I_am_learning]

I worked in a project in an old and small hydro power station where we replaced existing MOCBs with VCBs. In the second day from installation there was fire in the VCB when the breaker tripped. Everybody guessed it was manufacturing fault and there was air leekage to the VCB. A spare VCB was used. But unfortunately it also caught fire. Later it was discovered that the excitation system of the 500KW generator was faulty; the field breaker wouldn't trip when the main breaker tripped.
The generator is rated at 6.3 KV but is stepped up through a transformer to 11KV. The VCB is at 11KV at which synchronization is done.

Can the blame be placed on faulty excitation for the fire in the breaker?
I thought VCBs should be designed to withstand worst case condition of generator terminal voltage, which occurs when the generator is unloaded and maximum excitation current that the exciter can supply is supplied.

The blame can't be put on the system side, because other two similar units on the same 11KV BUS were functioning normally.

Thanks.

 

[Bobbywhy]

I_am_learning,

Without a circuit diagram it is difficult to visualize exactly where in the circuit the Vacuum Circuit Breaker (VCB) is located. A simple diagram would be useful for troubleshooting this. 11kV should not be excessive for the VCB: they typically are used up to 36 kV. As you speculated, there may have been a manufacturing defect that allowed air into the VCB. But a second one, same result? Not likely.

The power frequency current passing through the contacts in the VCB crosses the current zero point 100 times in a second (in your area). It is always desirable to interrupt the current when it passes the zero value. The control circuitry which commands the VCB to open should activate at exactly “zero crossing” of the voltage so that minimum current flows at the time of the opening of the contacts.

At the instant of opening if there is any flowing current it will cause a “hot spot” on the point of contact. Because it is in a vacuum, the metal of the contact is easily vaporized and creates a conductive path for an arc. The arc will continue until the next zero current crossing occurs. If the current is below 10kA the arc would be distributed over a large surface area of the contacts and metal vaporization would be at a minimum. Current above 10kA, however, causes the arc to constrict by its own magnetic field, and therefore it contracts. This causes extreme heating and metal vaporization on a small surface area of the contact, and, most likely, fire.

Since the VCB should open its contacts exactly during zero current crossing to minimize arcing, the problem may be in the sensing/control circuit, commanding the VCB to open at the wrong time. Can you test the VCB control circuits?

 

[jim hardy]

...
I thought VCBs should be designed to withstand worst case condition of generator terminal voltage, which occurs when the generator is unloaded and maximum excitation current that the exciter can supply is supplied.

The blame can't be put on the system side, because other two similar units on the same 11KV BUS were functioning normally.

Thanks.

I would have thought the same.

VCB's Achilles heel is inleakage. What kind of vacuum monitoring device do yours have?

Also - breakers have "interrupting current rating". How do your VCB's rating compare to the breakers they replaced ? Some switchgear includes intentional impedance to limit fault current so one can use less expensive breakers.

 

[Windadct]

For draw out or metal clad VCB I have never see a zero crossing circuit - it make the breaker mechanism way too complicated - at 500KW and 11KV you have very little current. Also - while a VCBs are used up to 36KV - they are still rated by voltage, at 11KV I'll bet this is 15KV Class breaker, like this ABB 1200A / 15KV - Probably just 3 "bottle" type contact assemblies - with no vac monitor.

My first suspicion is the gen over excited and then over voltage at the VCB cause a flash over phase to phase Gen side - should be pretty evident from the damage. The bad news is that this could have very likely caused damage in the Step up transformer.

Since it is known that the exciter / breaker malfunctioned - that aspect need to be looked at.

Lastly - Gen breakers are the worst for having big blowups - because, the arc ( fault) does not shutdown until the generator stops moving, there is no protection or system to stop the energy floe to the fault. The Step-up will limit the fault current, but these are often very low impedance transformers (2-3%), to minimize losses, since they are typically run at or near max-capacity all of the time - so their current limiting could be > 20x full load current.

One more thing - since there was some type of a control malfunction ... could the breaker have been closed in out of phase...- yep - that would blow it up. Were the PT / synchronization relay touched - AT ALL, when were they last checked. Confirming phasing and phase relay system on a live generator is incredibly dangerous - do not attempt. ( The the one of the most dangerous jobs I have done - even at low voltage - 480 V Generators ). Also - if the breakers contacts bounced when closing - that can cause a failure(you create a voltage boost due to system inductances), this can be seen on a breaker analyzer - with the breaker out of the circuit.

 

[I_am_learning]

For draw out or metal clad VCB I have never see a zero crossing circuit - it make the breaker mechanism way too complicated - at 500KW and 11KV you have very little current. Also - while a VCBs are used up to 36KV - they are still rated by voltage, at 11KV I'll bet this is 15KV Class breaker, like this ABB 1200A / 15KV - Probably just 3 "bottle" type contact assemblies - with no vac monitor.

Yes you are correct. There is no zero-crossing detection mechanism and no Vacuum monitoring.

My first suspicion is the gen over excited and then over voltage at the VCB cause a flash over phase to phase Gen side - should be pretty evident from the damage. The bad news is that this could have very likely caused damage in the Step up transformer.

Since it is known that the exciter / breaker malfunctioned - that aspect need to be looked at.

Lastly - Gen breakers are the worst for having big blowups - because, the arc ( fault) does not shutdown until the generator stops moving, there is no protection or system to stop the energy floe to the fault. The Step-up will limit the fault current, but these are often very low impedance transformers (2-3%), to minimize losses, since they are typically run at or near max-capacity all of the time - so their current limiting could be > 20x full load current.

The question here is, should 15KV generator VCB be designed to withstand the maximum voltage the generator can generate (which occurs when the excitation current is at max and generator is unloaded). Since CBs job is to protect the system, its only logical that they be designed to withstand that rare condition when excitation systems fail. What is the industry standard?

One more thing - since there was some type of a control malfunction ... could the breaker have been closed in out of phase...- yep - that would blow it up.

Were the PT / synchronization relay touched - AT ALL, when were they last checked. Confirming phasing and phase relay system on a live generator is incredibly dangerous - do not attempt. ( The the one of the most dangerous jobs I have done - even at low voltage - 480 V Generators ). Also - if the breakers contacts bounced when closing - that can cause a failure(you create a voltage boost due to system inductances), this can be seen on a breaker analyzer - with the breaker out of the circuit.

The generator was synchronized and after sometime, there was some fault and the breaker tripped. The incident occurred during tripping. So, the synchronization system is fine.

Thanks for the help.

 

[I_am_learning]

I would have thought the same.

VCB's Achilles heel is inleakage. What kind of vacuum monitoring device do yours have?

Also - breakers have "interrupting current rating". How do your VCB's rating compare to the breakers they replaced ? Some switchgear includes intentional impedance to limit fault current so one can use less expensive breakers.

I terms of current rating the new VCBs are greatly overrated. But Voltage rating was just normal 11KV. There is no intentional impedance introduced. Further, the incident occurred when the breaker tripped (it was generator overloading trip. the overload relay was quite faulty).

 

[jim hardy]

500 kw at 6.3 kv is only 46 amps
and at 11 kv it's only 26 amps.

so do I understand both your breakers got a false trip signal and failed while just trying to interrupt normal load current?

You are suspecting an overvoltage transient.
At the instant of breaker opening there's substantial energy in the magnetic field that has to go someplace.
Presumably the generator is equipped with a surge suppressor across its field to help with that - I would find a way to check that device.


That's just one of the things to look into, I mention it only because it's a passive device that one doesn't usually think about.

Good luck with your troubleshooting !

old jim

 

[I_am_learning]

500 kw at 6.3 kv is only 46 amps
and at 11 kv it's only 26 amps.

so do I understand both your breakers got a false trip signal and failed while just trying to interrupt normal load current?

You are suspecting an overvoltage transient.
At the instant of breaker opening there's substantial energy in the magnetic field that has to go someplace.
Presumably the generator is equipped with a surge suppressor across its field to help with that - I would find a way to check that device.


That's just one of the things to look into, I mention it only because it's a passive device that one doesn't usually think about.

Good luck with your troubleshooting !

old jim

Yes the incident happenned when when the breaker tripped when carrying normal full load current.

Since the field circuit breaker failed to operate, the "surge supressor" , a resistance used to dump the field energy, didn't get chance to operate.

 

[jim hardy]

In my plant we had permanent high speed recording equipment on bus voltages and currents so we could see just what went on in such events - I guess you are not so blessed ?

knowing nothing of your setup...
of course you'll verify grounding etc

this is a long shot - I'm wondering might your generator voltage have grown large enough to saturate that stepup transformer - saturated transformers will make quite high voltage spikes at line freq, every time flux flips.

in the Nuke plant to test for that would take a mountain of paperwork. And it'd be risky to stress the equipment.
Perhaps you can get a generator open circuit curve and some help from the transformer supplier for a calc.

Could you hang a 'scope and harmonic analyzer on transformer primary current & secondary voltage ( via ct & pt of course) and run an open circuit test to mild overvoltage? You'd look for peaks in primary current and spikes on secondary volts .

ahhh,,, I got to stay away -
- but it's hard . Old firehorse instincts. I'm sorry for the bother.

Good luck !

old jim