Running a 7.5 MVA synchronous motor from VFD

[Skogsgurra]

Hi, just returned from a (hitherto) not so successful attempt to run an externally excited 10 kV synchronous motor from a 500 V + transformer frequency inverter. The set up is used only to accelerate the motor (instead of a pony motor) with no load coupled to the output shaft. We had our doubts as to the possibility to have a stable operation all the way from 0 Hz to the 53 Hz needed. And our anticipations were met - and more. Right now, we have been able to run up to about 12 Hz, then there is a violent oscillation and the drive shuts down because of high DC link voltage (comes from acc/ret action caused by oscillations).

We are running a pure scalar inverter with no frequency controller. Everything the inverter is supposed to do is output a frequency and an almost proportional voltage (we use a 2 % boost decreasing to 0 % at 10 Hz).

Any experience out there? Tips? (If I might say so; spare the P and I tips for the non-existing controller). It is a pole angle oscillation we are getting and not a controller instability. There are no amortisseur windings and no other damping in the motor.


Gunnar Englund

http://www.gke.org

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

 

[Laplacian]

When is the excitation field applied?

 

[Skogsgurra]

We have a three-phase field applied "counter rotation" so we have a fully developed exitation from start (slip 100 percent at standstill). It gets higher as we get up in speed (slip >100 percent).

Gunnar Englund

http://www.gke.org

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

 

[itsmoked]

What is the period over which this freq ramp(0Hz-53Hz) is occurring? (Wondered where you got off too!)

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Laplacian]

Is it possible for the excitation field to be causing stability issues with the VFD output? Can you monitor the external field voltage/current when the VFD trips? Is it behaving the same way? Isn't the field trying to bring the rotor into synchronism, while the VFD is trying to maintain some slip?

 

[DaveScott]

Forgive my ignorance, but I thought that sync. machines generally use a dc field?
For the pole angle oscillation affect to the main windings (not the oscillation itself), I wonder if reducing the field current would help? For the oscillation itself, some small mechanical load?

Is the syncronous machine designed to operate from a cyclonvertor (low frequency - 12Hz is about right), or is it used for power factor correction (50/60Hz)?

 

[Skogsgurra]

Smoked: We have an extremely slow ramp. Five minutes. The oscillations start after about 90 seconds into the ramp.

Laplacian: Yes, there is some interaction. We have a (slight) change in excitation current when the oscillations set on. We have a manually adjusted, i.e. not automatically controlled, excitation. The variations we see are a result of the oscillations - not the cause for them.

DaveScott: Yes, of course, the field is DC. But we have to feed the excitation machine a rotating field (hence three-phase) to get excitation at speed zero. We have applied a small mechanical load (a 2x4" fir beam pressed against coupling by lifting one end in an overhead crane, doesn't change the behaviour). We have tried excitation down to 1/4 of rated. The machine is designed for use at 50 Hz.

Next moves: Running a faster ramp so we don't dwell in the 12 Hz region as long as we do now. Running torque control instead of scalar U/f and frequency ramp. We will also find out what the transformer does. Suspect that the primary 500 V do not produce desired 10 kV.

Any more comments/tips? Anyone done this before?

Gunnar Englund

http://www.gke.org

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

 

[DaveScott]

Can you check the resonant frequency of the electrical system? Since you have a sync. machine on light load, it can be inductive / capacitive depending on the excitation, and depending on the pole angle lead/lag.
It might be resonating with the transformer and rest of the supply system.

 

[waross]

Hi gunnar;
Two problems/questions for us;
1> What is causing the problem?
2> How can we get the motor started, despite the problem.

It may be important to know if the oscillations peak at around 12 or 13 hz. and then subside, or if they get progressively worse as the frequency increases.
If the oscillations peak at 12-13 hz and then subside, we may be able to treat it as a forbiden or critical frequency (similar to the critical frequencies sometimes encountered when running centrifugal pumps from VFDs).
If it is a critical frequency, we can probably jump from 11.5 hz to 13.5 hz.
If, however, the oscillations are progressive, a jump to 13.5 hz. may be disasterous.
Have you considered changing the exiter frequency? If you jump the exciter frequency up to 60 hz. with a VFD, a small generator, or some other means, and then observe whether there is a change in the frequency at which the oscillations start, it may give an insight as to whether or not the problem is related to excitation.
Respectfully

 

[Skogsgurra]

Thanks Bill,

That's a thought worth thinking. the 12 Hz are very close to 1/4:th of the mains frequency.

We get different resonant frequencies at different excitations. Higher the more excitation we have. I think that points more to a "mechanical" resonance than an electrical (between machine and transforme) one. Will try and record motor voltage and current to see phase relationsships.

Gunnar Englund

http://www.gke.org

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

 

[itsmoked]

[!]"Next moves: Running a faster ramp"[/!] Is where I was going. I'd try ~30-45s.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

Hi all,

We finally got the beast dancing to our tune! Scalar (V/f) did not work. Nor did a special "textile" mode used for PM motors. Activating a "damping parameter" seemed to make things even worse. It was only when we switched to "current source mode" that we got a firm grip on the rotor. After having seen the machine oscillating at 11 - 12 Hz and motor cables carrying so much current that they wiggled on the floor, we finally had a clean, nice start. From zero Hz to 50 Hz.

I am certain there is a lesson to be learned from this. But too tired to even think about it now.

Gunnar Englund

http://www.gke.org

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