Bearing Failure 1

petronila · Aug 1, 2011

Dear All

The attached pictures ilustrates a wind generator N.D.E bearing failure.

I am interested in to know the possible causes for this damage.

Thanks and regards

Carlos

waross · Aug 1, 2011

Nice pictures. From the bottom left picture it may be that the damage to the inner race is much heavier on one side. That may indicate a combination of very little grease and moisture contamination during an extended period of inactivity. Would this fit with the history of the machine?

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

electricpete · Aug 1, 2011

I’d say your challenge is big. The damage to inner ring is pretty severe and most likely covered up whatever started it. But there might be clues in the appearance of the bearing or the facts surrounding the failure that might steer you toward some conclusion. I doubt we can solve anything from here (it’s tough enough when you have the bearing right in your hands), but I’ll just mention some thoughts/questions for you to consider. (no need for response).

Basically I see heavy damage on inner race. Light damage on 1 or 2 balls. Some fretting on outside of outer race that doesn’t seem unusual.

Can’t see much of the outer race. What does the ball wear path look like on the outer race? Is it offset as if the bearing has been carrying thrust load? Is it tilted as if there was misalignment? Is it heavy as if there was excessive interference/lack of internal clearance ?

How do the shaft and housing fits compare to specification?

Looking at the inner race damage, it spans an arc of maybe 90 degrees. But the weird thing is that it goes almost all the way accross the width of the inner race. That seems unusual to me, usually spalls stay more toward the center where the balls run. Was there a lot of looseness present in the bearing before you removed it?

I see a few stripes in the inner ring pattern. Is it possible those are spaced at exactly ball spacing? (that would likely indicate damage that initially occurred while stationary, such as false brinnelling or pressing the bearing onto the shaft by pressing on the outer (vs inner) ring).

=====================================
(2B)+(2B)' ?

electricpete · Aug 1, 2011

is it grease or oil lubricated, and what was the condition of the lubricant?

=====================================
(2B)+(2B)' ?

petronila · Aug 1, 2011

Thanks Waross and Pete,

The lubricant is grease, the grease was very dark but we found loose fits in the bearing housings(shields), I think thats was the problem , attached you will find more pictures

Thanks again

Carlos

electricpete · Aug 1, 2011

Bill may be on to something. It may explain why the damage spreads over a wider area of the inner race than expected. I have a very detailed case study which points to similar cause (I'll post it as soon as I can organize it). There are some similarities in appearances and some differences. Most notably for differences, the area of corrosion is not just on races, but on other parts of the bearing. I don't see anything resembling rust on anything other than inner race of the bearing.

=====================================
(2B)+(2B)' ?

rasevskii · Aug 1, 2011

Was there any possibility of bearing current? Is the NDE bearing housing insulated from the generator frame? Are other such machines that are in service also affected, or is this the only one out of several others? It is usual for NDE bearing pedestals/frames to be insulated in the case of sleeve bearings on, for example, hydro units, but is this also the practice on wind turbine generators with ball/roller bearings?

Just an idea.

rasevskii

electricpete · Aug 1, 2011

Attached is the case study I mentioned before. I think it is a very interesting case study due to the number of different things that could be observed looking at the bearing. In retrospect, moisture as a contributor is during idle period is very likely based on bearing inspection and history review. Our bearing shows lots of moisture and it is pretty easy to imagine the corrrosion occurred only where grease did not protect the race. That included most of the lower half of the bearing face, as well as a tiny portion of the race. The tiny part of the race where the corrosion reaches the ball track is where the spall begun.

=====================================
(2B)+(2B)' ?

electricpete · Aug 1, 2011

There is discussion of vibration findings and troubleshooting activities in-place Spring 2011 (prior to above bearing inspection) here:

http://maintenanceforums.com/eve/forums/a/tpc/f/3751089011/m/75220887663/p/1

=====================================
(2B)+(2B)' ?

electricpete · Aug 1, 2011

Regarding our case study, the motor space heaters are energized during storage. 2 possible reasons we still got moisture:
1 – the space heaters are are above the lower bearing, so may not be as effective for moisture coming past the shaft to bottom of the lower bearing.
2 – perhaps space heaters were secured for some period.
I think I will at some point combine it with our vibration data to produce a little whitepaper for another thread (don’t want to side-track yours).

but we found loose fits in the bearing housings(shields), I think thats was the problem

I’m wondering how this affects this particular failure.
The bearing does not appear to have been spinning in the housing (would have given distinct circumferential marks), so shouldn’t generate heat from that interface.
I’ve heard fretting can cause crack that reduces fatigue strenght of ring leading to ring crack, but that did not occur.
Are you thinking that the debris from the fretting got into the bearing?

=====================================
(2B)+(2B)' ?

waross · Aug 1, 2011

Pete: Was your motor stored horizontally or vertically?
Thanks
Bill

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

gattie · Aug 2, 2011

Was any Arc Welding done on or near the motor?
Any possibility of welding machine ground return finding its way thru the bearing?

Also outer race fit tolerance appears to have been excessive with consequent rotation in bearing landing.

ScottyUK · Aug 2, 2011

Attached is a useful reference document published by FAG. It may have been posted here some time ago - I can't remember where I got a copy from.

----------------------------------