Odd Torsional Excitation - VSD Motor

[Franko]

A machinery train has vibration at the first torsional critical, but excited only near a certain speed. It is a "direct torque control" variable speed, 4-pole induction motor with 50 Hz utility grid frequency. The speed for high vibration events appears to be where:
Difference between 2 times line frequency and 2 times motor inverter output control frequency is equal to the first torsional mode frequency.
The 2 times in the equation is not normal - there could be typical excitation at 6 times and/or 12 times difference frequency for VSD's but these are far from operating speed.

Possible reasons anyone?

 

[GregLocock]

Sounds electrical to me, the doubling could be some rectification issue.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[electricpete]

I have some questions (which is not to say I'll have answers):

What is the machine speed where the vibration occurs?
What is the frequency of the vibration?
What is the frequency you have identified as torsional frequency? And how did you identify it (calculation?)
What makes you suspect the vibration is torsional in origin?
Is the vibration directional?
What is the load? Is it directly driven by the motor or thru gearbox?

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

 

[MotorVib]

Two time line speed I have only seen from electrical, and if it is changing with the freq. from the VFD then electrical is the most likly cause of the problem.

Chris

"In this house, we obey the laws of thermodynamics." Homer Simpson

 

[Franko]

To Greg Locock's comment: What are some possible causes of doubling for rectification in a 4-pole motor that should be more symmetric than a 2-pole - I'm not a EE.

Answers are below questions from Elecricpete:
I have some questions (which is not to say I'll have answers):

What is the machine speed where the vibration occurs?
Speed where motor speed equals 2 times inverter frequency as it’s a 4-pole motor - there may be some slip involved - not sure. Close to 1300 rpm with max. speed rated near 1600 rpm

What is the frequency of the vibration?
Close to 13 Hz - close to predicted first mode with high torsional amplitude at motor.

What is the frequency you have identified as torsional frequency?
Close to 13 Hz

And how did you identify it (calculation?)
Agrees with calculation within 10%

What makes you suspect the vibration is torsional in origin?
Gear has radial vibration as it should for torsional/lateral coupling. Motor and compressor are OK and both gear rotors have it, so it shouldn't be a lateral mode of either of the rotors - although one having a lateral critical near 13 Hz could give additional coupling. Gear is still to be reviewed.

Is the vibration directional?
Radial shaft vibration is high for all probes on the pinion and the bull gears. Don't know if there is also axial vibration on the gear at 13 Hz - dry diaphraghm couplings have axial modes far from 13 Hz.

What is the load?
Over 50% - may be closer to 80% of full load.

Is it directly driven by the motor or thru gearbox?
Helical gear

 

[MotorVib]

A quick check is to run motor on rubber pads and without the gear box. If the vibrasion is still there and changes with the freq. then it would know that it is electrical.

Now for what causes the electrical vibration it is related to how the motor is wired (inside) and the rotor are gap.

Have fun with this, when I worked on motor and generators I lived for problems like this.

Chris

"In this house, we obey the laws of thermodynamics." Homer Simpson

 

[electricpete]

Just out of curiosity - what is the machine on the other side of the gearbox, and what speed is it?

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

 

[Franko]

Centrifugal compressor runs about 11000 rpm with motor near 1500 rpm.

These large train components can't be mounted on rubber pads.

I'm thinking they still may have missed some grounding issue to give the odd difference equation.

 

[GregLocock]

Sorry, I'm a mechie, so my comments are purely based on experience as an observer who has learned the basic rules of HV electrics.

Just look at the output voltage from the VFD on a scope. Ideally use a signal analyser.

If the power supply and/or the waveform synthesis stage in the VFD have problems you'll probably see intermodulation products at that frequency.

Dry soldered or corroded electrical connections can form diodes which are a good source of second order.







Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[MachineryWatch]

Did you get this figured out?

http://www.machinerywatch.com

 

[Franko]

Still under investigation - not proven but most likely due to electrical grid intermittent faults increasing component of 50 Hz grid frequency. Transient monitoring may still be done in the future.

 

[Strong]

Franco,

I do not understand this comment:
"What is the machine speed where the vibration occurs?
Speed where motor speed equals 2 times inverter frequency as it's a 4-pole motor - there may be some slip involved - not sure. Close to 1300 rpm with max. speed rated near 1600 rpm"
How can the motor speed be 2x the inverter frequency? A basic 4-pole induction motor synchronous speed for 50 Hz power is 1500 rpm (30 Hz), and this is well be 2xLF of 100 Hz. I could be mistaken, but I thought motor speed is below inverter frequency and therefore well below 2x inverter frequency.

Walt

 

[Franko]

Sorry - should have said motor speed is 1/2 of inverter frequency. It is a 4-pole induction motor.