Is unstable frequency reference bad for the VFD

[Isakovic]

Dear all,

I have one installation with VFD running a PID loop and keeping a constant pressure. At some lower set point values it starts to oscillate, not much, at maximum output fluctuates +/-1Hz per second. It doesn't affect the process, but I have to leave it that way for about a week before I get to stabilize it and see what is causing it. Drive is 11kW, Yaskawa. Up/Down ramps are about 30 seconds if that is of any significance.

My question is: Is this kind of fluctuating output frequency bad for the drive itself? Is it putting extra strain on the DC circuit? It's been bugging me.

Thanks

 

[itsmoked]

Not as bad for the drive as it likely is for the motor and possibly the pump. It doesn't sound like a fast oscillation but still this could be reversing the torque sign depending on the magnitudes.

Sounds very much like the PID is not adjusted correctly. If there is a valve anywhere in the pressure side of the system you may be able to stop the oscillation with a very slight adjustment of the valve. Only you can make that call though. You could slightly close the valve and note if the oscillation slows further or stops. This would only be a stop-gap and the PID should be better tuned as you're planning.

Note that some processes may make it very hard to tune the PID when not running the actual process of interest. In that case on-the-fly tuning may be needed. Many VFDs allow this with "bumpless" tuning. If you understand PID tuning you would make your informed tiny adjustments to the coefficients, one-at-a-time, and watch the results.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Forbidden frequency?
If you have programmed a forbidden frequency and the commanded speed is in the forbidden zone the drive may be going above and below the forbidden zone.
On an 3600 RPM pump 1 Hz = 60 RPM forbidden zone.


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

 

[Isakovic]

Thanks for replies fellas.

There are no forbidden frequencies, it is just an unstable PID. I was worried about VFD, if frequent run/brake/run/brake... cycle could harm it.

 

[waross]

Over to you, Jeff.

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

 

[jraef]

Shouldn’t harm the VFD so long as the oscillation is not on the DC bus and CAUSING it, ie signs of a bus capacitor starting to fail. But generally that won’t happen at a particular speed.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[rotw]

Is this driving a pump or a compressor?

I think VFD's are notorious for inducing torsional ripples that may lead to damage/failure on the shaft line. It of course depends on the VFD topology, power magnitude involved, etc. In some case a torsional study is done to understand the risk and take possible mitigations, if any (operate with blocked/forbidden speed ranges, torsional measurements during string test, etc.).

Back to your question, I do not know if the oscillations you are mentioning have been accounted for in the machinery design or not and what could be the impact. These are certainly a source of electrical excitations which translate as mechanical excitations on the machinery downstream.
Torsional vibrations are called the silent killer, because there is no way you can know something is going wrong until you have a (catastrophic) failure. Except for the fact that if you have a gearbox between the driver and the driven equipment, the torsional vibration can generate some lateral vibrations that can have the advantage of prompting further investigations.

 

[Isakovic]

It's actually a fan. It keeps constant pressure (vacuum, underpressure) in a big filtering silo. It works better now, PID parameters were too aggressive.

It didn't occur to me that this type of behavior could create mechanical problems, I was only worrying about VFD.

 

[rotw]

It didn't occur to me that this type of behavior could create mechanical problems, I was only worrying about VFD.


Certainly it can:

http://www.engdyn.com/images/uploads/96-vfd_torsional_-_turbo_2008.pdf

http://www.betamachinery.com/assets...s_Equipped_with_Variable_Frequency_Drives.pdf

Electrical and mechanical need to talk together and to talk with VFD vendor.