Overtorque at short circuit conditions?

[thinker]

The test system configuration includes a VFD driven induction motor with stand alone gearbox to simulate a prime mover for the test specimen (MW range). Customer specification requires that gearbox shall withstand a three phase short circuit of the prime mover at 100% voltage. Can anybody advise how to estimate the expected (?) overtorque at short circuit conditions? May be this spec does not make much sense for VFD driven motor?

 

[TurbineGen]

It might make sense if there is a fault on the motor. The braking torque would be quite high for several cycles. Perhaps long enough to damage the gearbox.

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.

 

[electricpete]

Krause' "Analysis of Electric Machinery and Drive Systems" section 4.13 provides analysis of 3-phase short circuit at the motor terminals with example analysis of 3hp motor and 2250hp motor.

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

 

[elmotor]

I would say that the spec does not make sense in VFD, but for the net supply as the spec seems to be written for that. If the short circuit happens on the motor terminals, the drive will try protect itself (IGBT's) with fast overcurrent protection and will not create high torque.

 

[thinker]

Pete, thanks for the reference, I've already ordered this book.
Elmotor, what you say is exactly I was thinking about reading the spec.

 

[waross]

Once saw a 400HP motor do a lot of damage. The contactor was undersized and when it opened under load, it arced over phase to phase. This sudden short on the line terminals of a running motor developed so much decelerating torque that
A> The direct drive coupling exploded.
B> The motor shaft keyway was badly damaged.
C> The motor shaft was bent and had to be replaced.
and
D> The fan shaft was bent and had to be replaced.

The motor in effect becomes an induction generator with a short on the terminals. The inertia of the load supplies the energy which is mostly dissipated in the motor windings and the rotor. As the motor terminals are shorted, the VFD, while it may or may not contribute to the fault current, does not contribute to the torque. This happens so seldom that it is usually not considered and any resulting damage accepted as part of the overall breakdown.
If it is a concern, I would consider using a coupling somewhere in the drive train designed to limit the torque by slipping or failing. Even though a gearbox may be able to withstand the direct torque developed if the motor is "Crow barred" the shaft may still be damaged or bent by the forces of a key rolling out of a keyway as a coupling explodes.

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

 

[Masbibe]

Can somebody compare torque in this situation with nominal direct online starting torque. Wich one is bigger. According to damage that Warros describe torque with shorted motor terminals should be much higher than direct online starting torque.
But I believe that if you have strong network direct on line starting current should be nearly same as motor current when you have shorted terminals (same voltage, same impedance). OK I now that current is not linear with torque, but relative slip (in rpm) is also same in both cases.

Milovan Milosevic

 

[Masbibe]

I done one computer simulation with 400V, 160kW ,50 Hz , 1487 rpm. Current in moment of short circuiting is nearly same as in starting but torque is about 8*Tn in breaking direction. I attach graphs for those who are interesting this. Short circuit moment is 1.5 sec.

 

[Masbibe]

And graphs for Torque and speed.

Milovan Milosevic

 

[Skogsgurra]

Elmotor says "If the short circuit happens on the motor terminals, the drive will try protect itself (IGBT's) with fast overcurrent protection and will not create high torque". I need to correct that.

It is obviously not an external current (from the VFD) that causes the damaging torque. Instead, it is the internal EMF in the motor that does so. The EMF is a result of the DC flux that exists in the rotor and is locked in by the rotor bars. When there is a short across the motor terminals, a strong torque - initially equal to, or higher than, the starting torque - is developped. The torque is quickly reduced as the rotor flux is reduced and will exist for tens of milliseconds for rotors with a high R/L ratio and hundreds of milliseconds for rotors with a low R/L ratio.

So, the phenomenon exists. Regardless of how the motor is supplied - genset, DOL or VFD - no difference if the short is close to the motor terminals.


For Milovan: Waross' last paragraphs are worth reading once more. If the supply is weak, you can start the motor without any torque problems. But a bolted short across the motor terminals will cause problems.



Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Masbibe]

Gunnar,

OK for weak network I agree that motor should start without problems, but I think that motor should be design to hold out direct online start on strong network.

But as I see in my simulation torque with short circuit terminals seems to be much more higher than direct online starting torque on strong network.

 

[Skogsgurra]

Sorry, I was writing my post when you posted your simulations.

Your current and torque traces are very interesting. The fact that the torque dies after one or two cycles is clearly seen. The extremly high current (one half cycle, negative) is also very clear. I do not know what R and L your line has. But if you reduce them to zero (probably not possible), you should get a similarly high current peak during first half cycle of start. The reason, as I see it, is that even a very strong supply line does have some impedance while a bolted short on the motor terminals doesn't.

And, yes, of course. The motor, coupling and everything shall be designed so it doesn't break during start. If there is a diesel generator, the start usually isn't very violent because of the generators impedance. But, a short across the motor teminals can still prduce severe damage.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Masbibe]

Gunnar,

I completly agree with you about reduction R and L of network.
Current if there is no network impedance should be almost same during direct on line start as with shorted motor terminals, but torque is much higher with shorted terminals. Probably because of rotor speed (only diference between direct online start and shorted motor terminals during run). Probably because of this angle between stator and rotor current is bigger and you get bigger torque.
All of this is in relation with stator and rotor electrical parametars and I assume that there is difference in behavior when you have wound rotors and squirrel cage motors.
In simulation I used some Matlab preset model of squirrel cage motor and for network short circuit level I put 15MVA and X/R ratio 2.5.