Strane phenomenon with VFD's

[micjk]

This may or may not be an issue. I happened upon a strange phenomenon with the supply cables for our variable frequency spin pump motors.While collecting vibe data on these units I noticed a strange "noise" in my headphones. It was almost like an arching sound when my metal encased accelerometer cable rubbed across a stainless steel drip pan hanging from these motors (we have a problem with gearbox oil getting into the motors that we are working to rectify).I had a PCS tech check the pan for voltage from motor to pan using a Fluke meter, and found the pan to be "charged" with @ 20 volts. I can hold my accelerometer in my hand, touch the pan and hear a distinct "buzzing" sound. I can also wrap my hand around the supply cable and it gets louder. Not only can I detect it in this manner, but if I get my analyzer close to the supply cable, in a location 20' from the motors, I also pick it up. I have generated a spectral and waveform plot of this and have a peak at 7875 Hz. Any idea as to what is causing this spike in the spectrum? Is this "voltage" leakage or magnetic field? If I can pick this up with the analyzer could this "leakage" also be affecting the inverter components? We have had problems with these drives. My thought is that we may have the wrong type of cable (not sure what they used) on these units (280 motors). Curiosity is getting the best of me and I am not real familiar with VFD's. I'd like to think that I may have stumbled onto something here, but need "educational clarification" from the experts in this forum.
Thanks in advance for your valued input,
Kind regards,

Roy Gariepy
Maintenance and Reliability Dept.
Dorlastan Fibers LLC.
Goose Creek, South Carolina USA

 

[jbartos]

Comment: VFDs normally produce higher frequency harmonics that can induce voltages and produce common-mode voltages and currents in equipment they control as well as inside their enclosures. The internal switching effects, mostly due to harmonics, are normally solved by the VFD manufacturer to a great extent; however, the external effects are up to VFD application engineers and electrical contractors to minimize such that they are harmless.
There are many links and web sites on the internet dealing with the VFD applications and associated side effects.
e.g. Visit

http://www.ecmweb.com/ar/electric_understanding_variable_speed_3/

etc. for more info

 

[micjk]

Thanks jbartos
I'll put that with the paper I have written by Howard W. Penrose from the University of Illinois Energy Resourses Center.


Roy Gariepy
Maintenance and Reliability Dept.
Dorlastan Fibers LLC.
Goose Creek, South Carolina USA

 

[Skogsgurra]

micjk,

You have experienced one of the many phenomena that variable frequency drives generate. This particular one is probably the "Frame Voltage" phenomenon.

It is best explained by considering the stator winding and its capacitance to the frame. The PWM (Pulse Width Modulated) voltage from the drive contains the carrier frequency and lots of higher frequencies. The close to 8 kHz peak in your spectrum indicates that the carrier frequency (or switching frequency) is 8 kHz, which is a little higher than usual.

The voltage that can be measured between frame (pan) and ground is not a "charge" but the voltage drop in the PE conductor caused by the high-frequency capacitively coupled current from the windings. This current sometimes finds its way from frame through the bearings and to ground if the shaft is grounded. It sometimes destructs the bearings (EDM) if preventive measures are not taken.

Read more about it in the ABB Technical Guides (search for +ABB +"technical guide" in Google).

If you are using "HF Mode" or envelop mode in your vibration analyser, you will probably have noticed excessive, but atypical, noise from the bearings. This noise could actually be an attempt to interpret the electrical disturbance as a BPF. I am working with these problems, so some feed-back in this forum would be much appreciated.

 

[micjk]

Thanks skogsgurra,
Very informative info. Your correct about the carrier frequency being 8000 Hz. In my search for more info on this I have found several papers explaining VFD's. I'm not sure why the carrier is set so high but am going to see if it is possible to lower it, which if I've understood thus far what I've read, will reduce leaked voltage. As for the bearing damage due to EDM, we are faced with the same dilema. We were going to put ceramic bearings in these motors but unfortunately when we went to rebuild the first of 280 motors we found that someone had ordered the wrong size for the OB bearing ($18,000 shot to Heck). I am now looking into some sort of grounding brush for the motor shaft.
kind regards,

Roy Gariepy
Maintenance and Reliability Dept.
Dorlastan Fibers LLC.
Goose Creek, South Carolina USA

 

[jraef]

micjk,
Addressing your query as to why the carrier freq. is so high, it is because the PWM output to the motor causes some audible noise in the motor itself. Higher carrier frequencies move that noise into ranges too high for us to hear, making the problem "go away" as far as most are concerned. Lower carrier freqs are better for the motor in some ways, so it is not a bad idea, but when you lower it you will probably notice an increase in motor noise.

"Venditori de oleum-vipera non vigere excordis populi"

 

[Skogsgurra]

micjk,

I think that you should stop and think for a while. First a few words about terminology; There are no ceramic bearings, but there are hybrid bearings and insulated bearings with a ceramic coating. Hybrid bearings (steel races and ceramic balls) are very expensive and you are likely to move the problem elsewhere - it depends on what the driven machinery looks like. Insulated bearings (with a ceramic - usually Al2O3 - coating are not very effective since the capacitance of the insulation is rather high and leaks a lot of HF current.

There are better solutions, like common mode filters and shaft grounding systems. Sometimes a correct installation (correct cabling and some care regarding grounding) can solve the problem.

 

[jbartos]

Suggestion: Why not to review excellent publications by specialists in this area? Visit

http://www.ab.com/drives/techpapers/ieee/ieee.html

for a list of relevant papers
In the past, the papers were viewable; however, as the internet matures, many viewable informative resources become not viewable for some reasons or others.