T'do Under Short Circuit with AVR Trip of Generator 4

[HenryOhm]

All,

I am on a project where the systems integrator is proposing an AVR trip for generator short circuit rather than generator OCPD trip. I am concerned about the implications for arc flash incident energies and the time it would take for:
1) the control system to recognize the fault, communicate it through to the AVR and for the AVR to trip its output,
2) the time it would take for the smaller field to decay on the external side of the brushless exciter and
3) the time for the internal field inside the brushless exciter to decay to translate to decay in armature voltage feeding the fault.
I believe that the first is the shortest and the last is the longest, so I am focused primarily on T'do. I am assuming that as the external field on the brushless exciter decays, that the rotating rectifier begins to shunt the decaying internal field somewhat similar to how T'do is determined by test with the armature open circuit. How would one estimate the voltage collapse under short circuit from such an AVR trip and would T'do still give one a sense for the delay in the armature voltage collapse?

Thanks in advance for any help.

P.S. I already have MG Say's "Alternating Current Machines" and Ion Boldea's "Synchronous Generators" if the answer is already right under my nose.

 

[RRaghunath]

You are talking of a captive generator without grid connection, I suppose. If true, it is more urgent to trip the field breaker and short circuit the field so that the flux in the generator rotor rapidly gets discharged is important. Of course, it is also important to trip the prime mover to allow coasting down of generator.
There is no other source that can feed the fault and hence opening the generator main breaker may not be critical and in fact, keeping the load connected could help dissipate the stored energy in the generator rapidly.

 

[HenryOhm]

RRaghunath, thanks. But, with the short circuit of the field, it will still be flowing. Won't that mean it will continue generating voltage and feeding the fault? Wouldn't opening a generator main breaker be the fastest way to interrupt that fault?

 

[waross]

Do you mean a short internal to the generator?
If so, you want to kill the excitation an the prime mover as fast as possible.
Opening the main breaker will not clear an internal fault, but must be done on paralleled sets to avoid a backfeed.
You will need some type of differential monitoring to detect an internal fault.
For an external fault, trip the main breaker.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[RRaghunath]

HenryOhm,
Bill has already answered the questions raised by you.
Some utilities break vacuum in the condenser (Steam Turbine Generators) to make the machine coast down faster.
Excitation panels have special provisions to quickly reduce the remnant flux by reversing polarity of excitation current input to the machine, before shorting the excitation.

 

[FreddyNurk]

RRaghunath, happy to be proven wrong, but my understanding was that your last statement does not apply to brushless excitation systems.

EDMS Australia

 

[RRaghunath]

FreddyNurk,
Agree that is touted as the biggest advantage with Static Excitation system (with Brushes), i.e. field can be quickly killed, minimising damage at the point of fault and reduce the risk of fire.
Yet, Brushless excitation has been the system of choice for operational/maintenance and cost benefit. Some utilities provide a breaker in the generator neutral circuit to cut-off fault currents but this can help only in case of SLG faults.

 

[waross]

With good differential protection there is a good chance to shut down on turn to turn shorts before an arc develops.
Arcing will often damage the core.
Brushless excitation is the method of choice for small machines.
As the size of the machines increases, so do the arguments in favour of brushed excitation.
On a major utility generator the consequences of a load dump induced transient overvoltage are much greater than the possible consequences of a load dump on a 1000 KVA standby generator.
A trained maintenance staff is available full time on a utility class generator.
Not so much with standby generators.
Many standby machines get nothing but fuel from the day that they were installed until the day they quit.
I was the one who got the call when they quit too many times.
(But the work helped to keep two kids in private schools.)

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[HenryOhm]

Thanks for everyone's helpful discussion! I should clarify, I am only concerned about arc flash energies inside the switchboards that these generators are feeding into. There is a pair of switchboards with bus-tie interconnection and six generators each about 1MW in size. At least one would always be held in reserve. I'm just trying to understand how effective a trip of generator brushless excitation would be versus a generator circuit breaker trip.

 

[waross]

Someone should suggest to the system integrator that it is not appropriate to cheap out on the generator breakers.
Using switches or under rated breakers for generator disconnection rather than breakers capable of interrupting the generator fault current is not a good practice.
There will be significantly more arc damage to the switch boards with an AVR trip than with a breaker trip.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!