Potential Bearing Problems With VFD

[Hawkeye]

I was doing a bit of research on VFDs this morning when I found this site: https://www.pumpsandsystems.com/preventing-drive-induced-electrical-damage-ac-motor-bearings

They're talking about voltages being induced in the shaft of a motor being run from a VFD. I recall from my training as an electrician that different frequencies of AC require different 'amounts' of iron and copper to accomplish the same tasks. For example, AC systems on aircraft may run at 400 Hz, rather than 60 Hz, because the transformers and other inductive equipment can be made smaller and lighter.

The site was pointing out that voltages up to 70 VAC can be induced in the shafts of motors when powered by VFDs. I got curious and measured the voltage between the output shaft of the motor on my RF-30 and it's mount. When it was running, I measured up to 2.4 VAC at around 44 Hz. The readings were lower at other frequencies and lowest at 60 Hz. On my big lathe (2 HP motor) the highest reading was just over 3 VAC at 10 Hz, and decreasing as the frequency rose.

Any on this forum who have played with EDM (Electrical Discharge Machining) know that we use fairly low DC voltages to remove bits of metal. For that matter, arc welders operate at relatively low AC or DC voltages. The point of the site is that the races on the bearings can be damaged over time when running at various frequencies.

Just to be sure, and because I needed something to do to stay out of our current heat wave, I made up a couple of wipers to provide a path to ground , bypassing the bearings on my motors.

I had a bag of flat springs I got at Restore, so I used those for the contacts. The ball bearing further reduces wear. The yellow wire connects to the motor mount.


On the lathe, I had to sand off some rust from the side of the pulley, to improve the contact. You have to look closely to see the strap spring.


I don't know if this will make the slightest difference in the long run. I just thought it was interesting, and it did fill the need for something to do in the cool of the shop.

 

[kb58]

I'd be very curious what voltage is measured at 60Hz, both through the VFD, and straight into the wall socket. I suspect it may be the same.

 

[markba633csi]

My gut feeling is that around 60 Hz, on the small motors we use, the effect is so minimal that it's not worth the extra gymnastics. I have heard it can be a real problem on other installations though. It's definitely a thing to be aware of. I wonder if there exists a special graphite based grease for that. ?
I also wonder if the very cheap VFDs are worse in that regard, with very minimal (or no) output filtering to cut down those super fast risetimes?

 

[mksj]

Like you, when I first starting reading about this when I started to get involved with VFD's I was quite concerned until I read several discussions from some of the VFD designers and manufactures white sheets on the subject.

There has been some discussion of this risk in other forums that deal with industrial equipment, what has been mentioned is it is not a significant issue at 240VAC and also with smaller motors. Motors operating at higher voltages and also above 75-100 Hp it does become a significant issue and costly. In order to visualize what is happening there are two approaches that I have read about, one is using a storage scope with a probe that connects to the motor shaft, you can see the charge build up and then discharge to 0V. This happens in nanoseconds, so something you would not be able to measure or see w/o a high frequency storage scope. The other method is an RF probe which can pick up the RF generated from the discharge, it is very localized and the antenna needs to be close to the bearing. I have seen some discussion of using a high impedance AC voltmeter with a carbon brush on the shaft, but I do not follow that they would be reliable given there narrow frequency AC detection.

There are a number of methods to mitigate the charge, the first and foremost is proper grounding/shielding between the VFD and the motor chassis. Additional measures such as using a toroid on the output of the VFD cable or a dV/dT filter also helps and is commonly used. Conductive grease in the bearing hastens the lubrication breakdown and is not used, more common is some form of wiper shaft system or graphite rod against the back shaft. Insulated bearing holders and also ceramic bearings have been used. Actually a pretty complex issue when you get into it.

Measuring motor shaft voltage discharges with Fluke MDA-550 Motor Drive Analyzer

This application note explains how to use the MDA-550 Motor Drive Analyzer and a shaft voltage probe to measure motor shaft voltage discharge events.

www.fluke.com

Bearing currents

new.abb.com

A couple of VFD designers that have been involved with VFD for decades, indicated that with the smaller motors at 240VAC they almost never saw a bearing failure due to this phenomenon and felt that shaft grounding was not necessary. If it was a significant issue you would see it built into the design of inverter/vector motors in this size/class range. I have only seen one inverter rated motor from WEG that came with a carbon shaft discharge brush. Bigger motors will often have insulated bearings, as well as a number of other design elements. Interesting reading, proper design and grounding is a much more complex issue with a lot of specifics. Fortunately it would seem that this is not an issue for the weekend hobbyist, most of the factory VFD systems for smaller mills and lathes, do not seem to worry about it. Can't say I have seen a shaft grounding brush on any of the factory VFD machines I have been involved with, let alone they often do not even use shielded or twisted cables. CNC it becomes much more of an issue with regard to electrical noise, I do not know how the drive motors are designed to address this issue on larger CNC equipment.