Possible causes of TEFC NDE brg spinning on shaft 5

[electricpete]

This is a 100hp, 3600rpm TEFC motor, with 6313 bearings.

In 2003, the motor bearings were replaced. Motor operated satisfactorily from 2003 until January 2011, when it started developing noise and increasing vibration (mostly broadband… like a “raised floor”).

We removed the motor and sent it to a repair shop.

The shop reports that for the outboard (NDE) bearing, the bearing inner ring is spinning on the shaft (a very undesirable condition, should be interference fit). Additionally, the outer ring is “locked in the housing”.

Additionally, their preliminary view of the cause of the event is that excessive greasing caused overheating which caused the bearing to grow enough to spin on the shaft.

I don’t know how much grease is present or what degree of overheating evidence is present.

Since it is a somewhat unusual failure, I am going to the shop tomorrow to take a look for myself. Before I do that, I want to think through what the possible causes are in general, and what I should be looking for during inspection of the disassembled motor.

My initial thought is that any indication of overheating doesn’t prove anything about the cause of the bearing spinning on the shaft… after all it will likely overheat once it starts spinning regardless of why it started spinning. Maybe the location of overheating will provide some clues, along with quantity of grease.

I also think that in general the most likely cause of any bearing spinning on shaft would generally be insufficient interference fit when bearing was last replaced. Shaft dimension and bearing housing dimension were not checked when bearings were replaced in 2003 (not by this motor repair shop).

Are there any other thoughts on what the possible causes of a bearing inner ring spinning and the shaft are..., and what I should be looking for during inspection?

How would outer ring locked in the housing relate to any of these scenarios?


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

 

[ccjersey]

With the shaft getting slightly longer when the rotor temperature comes up, something has to allow axial movement. Usually the bearing is allowed to slide in the end bell and perhaps a spring washer used to ensure preload on the bearing.

If the bearing seizes in the end bell, then every thermal cycle would degrade the interferance fit on the shaft.

 

[electricpete]

Thanks. What would cause bearing to seize in housing?

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

 

[MikeHalloran]

Being installed dry and clean, without the slightest trace of grease, would certainly allow, if not cause, a bearing to freeze/rust into a housing.



Mike Halloran
Pembroke Pines, FL, USA

 

[electricpete]

Mike
I'm not sure I understand. Are you suggesting to apply grease to the seating surface of the housing (where outer ring od sits)? I haven't heard of that.

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

 

[flandrax]

As far as grease goes you might look for mixing of incompatible grease... I have seen it cause bearing failures bad enough that it caused the inner race to spin on the motor shaft and was so bad the it bent the shaft. It too was on a 3600 Rpm motor. 300 Hp.

 

[MiketheEngineer]

I think he means no grease on the races or the balls themselves??

 

[MikeHalloran]

Yes. Apply grease, or at least leave the preservative grease on when you remove the wrapper from the new bearing.

I'm guessing nothing like that is mentioned in the maintenance manual, and probably not in the process sheets of the original manufacturer, either.

Without meaning any insult to sparkys, it's just silly to expect a clean steel bearing to slide, even slowly, even just a fraction of a mm, inside a clean ferrous housing absent some kind of lubrication.

Old-time motor shops probably didn't worry about it because all available hands and work surfaces were covered with grease anyway. Nowadays, the lawyers who write standards expect workshops to be clean enough to not blemish their silk ties.


Mike Halloran
Pembroke Pines, FL, USA

 

[starkopete]

This is from our repair spec.
"To mount the endbell, it must be clean and should have a coat of light oil (711 or WD40) applied to its bearing surface, also, a coat of light oil may be applied to the cooled bearing’s outer race."

Thanks

 

[electricpete]

I can see light oil would be helpful. I would be a little cautious in applying grease… might help during installation, but over the years as the oil bleeds out and the thickener is left… seems like it might actually make the situation worse.

Bearing OEM’s provide very specific bearing installation instructions regarding bearing housing fits, deviation from roundness of the housing, I think I have even seen roughness specification for the housing. Personally, I can’t recall seeing bearing OEM guidance regarding this practice of lubricating the housing, which I would expect if it were critical. I would be appreciative if someone can provide further references regarding this practice.

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

 

[electricpete]

As far as grease goes you might look for mixing of incompatible grease... I have seen it cause bearing failures bad enough that it caused the inner race to spin on the motor shaft and was so bad the it bent the shaft. It too was on a 3600 Rpm motor. 300 Hp.

Are you talking about a situation where the bearing binds internally such that inner ring can't move relative to outer ring?


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

 

[electricpete]

I did find the SKF Bearing Installation and Maintenance Handbook states "The housing bore and bearing outside diameter should be coated with a thin coating of lightweight oil."



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

 

[electricpete]

I did find the SKF Bearing Installation and Maintenance Handbook states "The housing bore and bearing outside diameter should be coated with a thin coating of lightweight oil."

BUT thinking some more, it should be noted that the same reference also states: "To press bearing on shaft, first apply a thin coating of lightweight oil to the bearing seat and bore."
Clearly the reason we coat the bearing bore is NOT to faciliate movement of the bore on the shaft during subsequent operation (which would be undesirable), but simply to facilitate installation of the bearing. So we have to consider the possibility that the lubrication of the outer ring /housing is stated for the same purpose (installation, not to facilitate movement during operation). Along the same lines, we have to question whether the lubricated/non-lubricated condition of the bearing seat after 8 years of operation has any relationship to it's initial state.... that original light coating can certainly disappear... the vapors from the bearing grease might also replenish some lubrication.

In summary, on the subject of importance of lubricating the housing, I'd like to say it's a good practice, and I think most would do it simply for ease of installation, but I'm not convinced that failure to do it would be a fatal flaw. I don't have a strong opinion one way or the other yet. I will say that the whole idea of bearings moving within the housing to faciliate thermal growth is fraught with many unknowns and uncertainties ... for example it can be faciliated by equipment vibration so a smooth well-aligned machine could be a liability. So again it certainly doesn't hurt to do what you can in terms of initial lubrication.

I think I will have a pretty good chance tomorrow to test the hypothesis of whether binding of the outer ring in the housing caused subsequent movement of the shaft..... I will clearly mark the orientation of both bearings before dismounting them, then cut them apart and check the direction of thrust loading on each bearing evidenced by the ball track. If there has been enough thrust to force that bearing out, there should be clear evidence of that thrust loading on each bearing in the direction associated with restraining an expanding shaft

===========

Just thinking about one other thing... ccjersey already came up with a good hypothesis of how a stuck outer ring within housing could cause sliding inner ring on shaft. Is there any way that sliding inner ring, resulting in a lot of heat, could end up binding the outer ring in the housing? (I can't think of any way, but maybe you guys can).

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

 

[MikeHalloran]

>>>BUT thinking some more,<<<

You gotta stop doing that; you're going to hurt yourself. ;-)



Mike Halloran
Pembroke Pines, FL, USA

 

[MikeHalloran]

>>>the whole idea of bearings moving within the housing to faciliate thermal growth is fraught <<<

Agreed. I prefer mounting the bearings in flexible diaphragms, or almost anything except expecting the bearing itself to move in the housing.

I'm sure it looked okay on the drawing board. ... and I'm pretty sure that no M.E. was looking at it at the time.



Mike Halloran
Pembroke Pines, FL, USA

 

[electricpete]

That drawing board has knuckle drag marks all over it ;-)

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

 

[rhatcher]

Pete, if the previous bearing replacement did not include a recording of the bearing journal and housing fits then you probably cannot determine the cause of failure with any certainty without the assistance of a bearing expert (ie. bearing failure analysis by the manufacturer)

My experience suggests that an improper bearing fit due to age related wear is more a more likely cause of failure than a perfectly fitted bearing with too much grease. Another likely cause is improper installation practice as discussed in previous posts.

I would suggest that your motor repair specification for any motor, regardless of size, must include checks of the bearing journal and housing fits at a minimum of 120 degree intervals. Any fit that does not meet specification must be repaired by plating, flame spray (if allowed), or sleeving.

PS: for a grease related failure, I would expect the stator to be literally filled with grease.

 

[edison123]

I am with rhatcher. Probably the original shaft interference fit wasn't enough (we are talking less than a thou here, so the measurement and the measurer have to be accurate).

In my shop, we apply a light coat of oil on the housing bore and the bearing OD to ensure an easy mallet fit. The inner races are always induction heated to about 100 deg C to 110 deg C before fixing on the shaft, regardless of the size of the shaft.

And yes, we check & record all the fits on the shafts and the housings, before we fix the bearings. Sleeving is done only for the loose housings. For loose shafts, our SOP is to change the shaft.

Muthu

http://www.edison.co.in

 

[rhatcher]

Pete, I will add that the ODE bearing on a motor like you describe will move in the housing to allow for thermal growth of the shaft length dimensions for the driven equipment and the motor.

Some ball bearing motors use a 'wavy washer' to presumably (?) resist this growth and to move the shaft back towards the 'cold' dimension during the motor cooling after the motor is stopped.

MikeHalloran referred to the 'wavy washer' as a 'flexible diaphram'. I believe these terms are synonymous.

 

[MikeHalloran]

Uh, NO NO NO NO.

A wavy washer is not a flexible diaphragm.

A wavy washer comes from a place like Smalley Steel Ring, and in this instance would be used to force the bearing's outer race to slide to one end or the other of its travel, or maybe to center the bearing within its travel range, sliding within the housing.

My reference to a flexible diaphragm means pretty much that; a motor NDE bell that is radially rigid, but axially flexible, like a flexure, so that if/when the bearing outer ring seizes in the housing bore, it doesn't matter, because the housing will flex to allow axial displacement of the bearing.

Of course nobody actually builds motors that way.



Mike Halloran
Pembroke Pines, FL, USA