Screw pump bearing failure 3

[Evlaar]

Hi,

I have had a bearing failure on the NDE of a motor (fixed speed) - after the motor was extremely noisy and vibrating. The bearing was well lubricated, with the correct lubricant and the alignment checked. There were no signs of water ingress, corrosion or additional signs of failure mechanisms i.e. excessive temp.
The bearing has large ridges on the inside of the outer race that span half way round as seen in the photos. There are 5 screw pumps in the station - and this is the only one with repeated bearing failures. Any suggestions for the cause of failure?

Thanks in advance

 

[Evlaar]

Photo attached

 

[JJPellin]

This looks like classic ball bearing fluting damage. This would likely be a result of stray current that can have a number of different causes. Search for ball bearing fluting and there is a lot of literature on-line that can help you understand and address this failure mode.

https://www.motioncontroltips.com/e...n of bearing,enough that failure is imminent.

Johnny Pellin

 

[Evlaar]

My confusion is it only being part way around the bearing? I am struggling to visualise this - thanks

 

[Strong]

"My confusion is it only being part way around the bearing?"

I agree that is sure looks like electrical discharge machining (EDM) from stray current passing through bearing. I have seen this several times and it does not necessarily track the entire race. Possible reasons for partial tracking are bearing load zone location and dominant electrical path. Replace the bearing and be sure to eliminate cause of electrical current.

Walt

 

[quadtracker]

It may be false brinelling caused by vibration caused by equipment nearby sending vibration across the floor to a poorly supported pump foundation. This vibration passes to the pump and when it is shut down and others are running the bottom half of bearing gets brinelled from the weight of the pump resting on those bottom rolling elements. Pump foundations need to have a concrete and grout base to reduce vibration from all sources. Steel bases without concrete will conduct vibration easily. Example: hammer a 30 foot steel beam you can feel the vibration 30 feet away, hammer a block of concrete you can't feel vibration 2 inches away.

 

[3DDave]

I do like transmitted vibration as a possible cause - it would explain why only one pump in one location is susceptible.

Possible diagnostics:
vibration study to evaluate the levels at each pump, operating and resting.
swap the entire pump/motor with one of the others to see if the problem follows the pump/motor or stays at the location.
swap just the motor to see if it's the motor or the motor/fan combination.

Once the race is even slightly damaged it is likely to accelerate - as one bearing hits the defect the neighboring elements take a load that causes damage in alignment with them. With gravity the load is higher on the bottom, so if it starts just a little bit when resting, it will get worse aligned with the major force vector. If this is a vertical shaft, then that's no it.

I've read about back-up generators that fail in a few minutes of operation because vibration from neighboring equipment hammered the bearings over years on standby storage. Fuel is checked and batteries are checked, but not running the unit prevents moving the bearing elements, and they are destroyed while not moving.

 

[electricpete]

False brinneling will typically include marks which have irregular instensity (owing to different durations of stopped condition) and irregular spacing other than ball-interval spacing between marks from a single stopped period (irregular spacing owing to random stopping places).

The marks in the photo have fairly regular intensity and regular spacing.... much moreso than any false brinneling I have seen. So I vote with JJPellin on this one, fluting from electrical causes, not false brinneling from vibration of a secured machine.

I don't have a good explanation for the pattern existing only half of the race for fluting, but fluting is quite mysterious to begin with and certainly the difference in contact pressure at the bottom of the bearing could plausibly play a role for all I know (ask someone to explain the spacing between marks in fluting, it is quite bizarre to begin with).

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