Synchronous motor machine angle( transient stability) 1

[gdeep]

Hi,

I have a question about the machine(torque)angles in synchronous motors with exciters.The sync motor is a 6500hp 36 pole motor @ 4.16kV, which is started as an induction machine.The load is a high inertia load which gets engaged when the moor reaches sync speed.

When the motor is running at synchronous speed and load is applied to it, would the rotor deaccelerate first, the machine angle would increase, the exciter would then try to increase the excitation field to accelerate the rotor to try to get it in sync speed with the stator fields, and thus reducing the machine angle. But the rotor would overshoot the sync speed( assuming it a cyclic load changing every 5 s).

So would the torque angle be consistently changing between a range of values and would the speed of the rotor be also changing slightly around its rated rpm? Is there a way to calculate the max/min range of values of the angle using the swing equation?
Do you guys know of any good reference where I can learn about it more?
Thanks

 

[juleselec]

The rotor angle will oscillate when you engage the load, but you'd hope that the oscillations are damped and will settle to a steady state at a fixed rotor angle (proportional to the amount of mechanical load) and synchronous speed. This will depend on the inertia of the machine, the damping characteristics of your amortisseur windings, the time constants in your exciter and also the control structure / type of exciter (e.g. brushless, static, etc).

In theory, you could use the swing equation to work out the oscillations, but you'd also need to calculate the damping coefficients, which I imagine would not be particularly straightforward. The standard way to investigate this would be to model the machine in software (e.g. PSS/e, ETAP, PowerFactory, etc) and run a time-domain simulation, i.e. using numerical integration to solve the differential-algebraic equations. To really get it right though, you'd need an exciter step response plot (ideally measured from the actual machine) so that you can tune the exciter parameters. Otherwise, you'll just be plucking numbers out of the air, which to be fair is what happens in most preliminary stage studies!

A good reference book is Kundur's Power System Stability and Control. You'll find it on the desk of pretty much anyone who does stability studies.

 

[waross]

The average RPM will not change short of pole slip. As the speed oscillates the rotor angle will be the significant variable. As the angle increases the restoring torque will increase, even without AVR action. The AVR may or may not contribute.

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

 

[DHambley]

Theoretically, with no damping the error angle will continue to go back and forth like a pendulum (assuming it doesn't slip so much that it pole hops). You should make a model which includes every damping parameter in your system which you know about. For example, this would include items like wire resistance and source impedance and the effect on reducing the delta between back EMF and the control waveform, etc.
Have you got a fixed line frequency or a variable speed controller?

 

[gdeep]

Thank you all for the response.

The sync motor is reactor started and drives a high inertia load such as a ball mill. Since the load is cascading inside the mill, the mechanical torque experienced by the motor is consistently changing and hence the machine angle. Damping in the system and motor would try to smooth the machine angle but since the acceleration torques are changing, I am not sure if the pendulum effect would still exist and how much would the AVR affect.

What i am trying to accomplish is to get max torque out of the motor when the load is engaged.

 

[waross]

The amortisseur winding will provide damping. Is this a real problem or a hypothetical case?
If this is a real problem I would suspect either an under sized motor or an overloaded mill or both.

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

 

[electricpete]

You mentioned it's a high inertia load - that inertia tends to reduce effects of load torque variation as seen at motor.

Pendulum effect? I think maybe you're referring to a concern about resonance - if the impacts occur periodically at a rate that matches the natural oscllating frequency of the electromechanical system. As mentioned damping does help the situation. Large periodic torque impacts seems like a possibility for a hammer mill, I wouldn't think it would be as likely for ball mill (I'd think the impacts are smaller and more random rather than periodic... knowing very little about mills)

One thing you might look at is the variability in the motor current. Does it oscillate on the panel current meter? If the suspect frequency is fast enough to be filtered by meter response then perhaps get a current trace and look at the envelope.


=====================================
(2B)+(2B)' ?

 

[electricpete]

What i am trying to accomplish is to get max torque out of the motor when the load is engaged.

Boost the field voltage as high as allowed?

Are you suspecting that you've been having pole slips?


=====================================
(2B)+(2B)' ?

 

[7anoter4]

See:

http://bauhaus.ece.curtin.edu.au/~em302/Lecture Notes/Lec 8-10 SM Dunamics.PDF

 

[LionelHutz]

What i am trying to accomplish is to get max torque out of the motor when the load is engaged.

Well then look at the motor data sheet and read the pull-out torque. You'll get that much torque if you have rated line voltage and rated field current.

You could temporarily field force the motor to a higher current as you engage the clutch to get a little more torque.

In my experience, if you're having trouble engaging the clutch without slipping a motor pole it's likely because the mechanicals have cranked up the air pressure so the clutch doesn't slip and therefore wear as much on engagement.

 

[gdeep]

Thanks for the pdf 7anoter4

Yes, this is real case scenario where the motor is slipping poles when the clutch is engaged and trips the motor. The clutching time is about 5-6 sec and they do allow the clutch to slip @ first and hopefully they are maintained properly.

Electricalpete: I am trying to get some power data from the plant but what I have seen is that the voltage on that bus does swing +-5%

I have two thoughts about this

1) I was thinking about raising the bus voltage to 4.5kV(ensuring that the insulation can withstand that voltage) during a clutch in and then bringing it down to 4.16kv level when the motor is running. The only concern i have is if the motor core can get heavily saturated and what effects it may have for 5-6 sec

2) Over exciting the machine before a clutch in - I have to check what are the exciter ceiling voltages and what effect it may h

 

[LionelHutz]

Is this an older mill that did work at one time? If yes, then try to find something that has been changed from the original design. Typically, the older mills have motors that were over-engineered and over-built so they have no issues operating the mill.

 

[electricpete]

36-pole... must be salient pole.

Is it common for poles to slip without damaging the machine?

=====================================
(2B)+(2B)' ?