Low temperature = short bearing life?

[Skogsgurra]

Hello,

I have been involved in the investigation of bearing problems on 22 small (0.37 kW) asynchronous three-phase motors. They have been running for about a year and the bearing life has been between two months and a year. All have been changed - some of them twice. The number of motors installed is 18.

The motors run in a freezing room and the temperature is minus 25 centigrades. They are fed from frequency inverters with a pure sine output (no PWM residues left on the voltage). The higher speed limit seems to be 1500 RPM. It not known what the lower speed limit is.

The bearings are 6202 and sealed. When the customer took one bearing apart, he found that there was no grease at all in it. "Completely dry" as he put it.

Now, there is a discussion about what causes this problem. I know very little about bearings, but I have checked that the frequency inverter does not produce any shaft voltage and there is no EDM in the bearing.

The mechanical guys do not believe me (even when I explain why and how I do the measurements and let them watch me doing it). They still say that this must be an electrical problem. I say that it is not.

I need some input from someone with good knowledge about bearings and low temperatures in combination with variable speed motors. We do not know if all bearings were without grease. Only one was checked. Can grease disappear from a sealed bearing? There is no evidence of EDM. No microscopic photograph has been taken.

 

[Carburize]

First reaction would be that the absence of lubricant is the problem and I can't think of a reason why exposure to minus 25C would cause grease to disappear leading to an initial conclusion that the bearing had no grease to begin with!
Any elecrical discharge damage due to stray current should be fairly easy to see.
What does the damage that causes the bearing to be replaced look like?

 

[Skogsgurra]

Carburize,

I haven't seen the bearing. So I cannot say what it looks like. No microscope pictures available. I do not think that all bearings were "dry". Just the one that the customer looked at. All other bearings seem to have been thrown away by the Spanish motor manufacturer.

We will have a closer look at the next one that fails.

 

[electricpete]

Since you ask about temperature, one question we might ask is whether the originally supplied grease viscosity is low enough at the low temperature seen during initial motor starting.

You have already told us the bearing number. Any info on the grease used? (perhaps full bearing p/n and manufacturer can help us to track down grease).

If someone is asking for explanation why grease was apparently found missing one bearing, I can only think of two possibilities:
1 - It was not present initially.
2 - The bearing underwent a temperature excursion which allowed the grease to liquify or vaporize and somehow get through the seals. Although surely this would leave brown staining on the bearing surfaces.

With the magnitude of the problem, whoever is the owner of these machines should demand a careful examination of failed bearings including photographs and grease samples.

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

 

[Tmoose]

What does the motor do? Direct mounted fan, belt loads, ??

As others said, No grease suggests it may not have been there intially.

Regardless, the unusual temperature raises questions about things like bearing housing and shaft fits when cold, the bearings' internal clearance when cold, and the suitability of the lube when starting from such cold temps.

I think you (or better yet, the manufacturer) need to grab a new motor from stock. Then evaluate the fits and probably correct them, and sneak out the seals and clean and re-grease the bearings with the synthetic lube recommended by Mobil or Kluber. That motor would be installed in place of the next failure, and become your baseline.

 

[Skogsgurra]

Thanks all,

I have had a discussion with SKF HQ in Gothenburg about this and this is the theory so far: The motors probably (no one seems to know) had normal grease in the bearings. They are speed controlled and there is no lower speed limit, so they can actually stop if the controller thinks they should. The grease gets rather hard and when the controller wants the motors to rev up again, they are more or less stuck (0.37 kW, not much punch) and the rotor heats. The heat makes shaft and inner race expand while end bell and outer race still are cold. When the motor finally starts rotating, the balls are pre-stressed and it is likely that something happens to the bearings under such circumstances.

SKF recommends WT grease or LGLT grease. I am not directly involved in this, so I will let the machine manufacturer do the digging.

Does the reasoning sound too outlandish? Or have anyone had this before?

 

[electricpete]

That's interesting.

I have heard of similar scenario for rapid starting under cold conditions. The startup period is critical because of potential for temperature increasing at different rates. With a very cold mass of end-bell housing surrounding the outer ring easy to imagine it would be slower to heat up.

One would think increased internal clearance (C4) might help that scenario.

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

 

[Tmoose]

Some folks used to provide a motor heater function using a real low output from the controller. Forever live sounds like the kind of thing that might be illegal nowadays. Still, I'd think any modern controller could warm them up a bit, then keep 'em rolling.