Motor Vibration 3

[nfldengr]

I apologize in advance for the long post.

I have a 4160V, 650 HP Motor(AC,Ind.) which is experiencing some vibratory difficulties...

History:
Motor had initially wiped bearings (deep groove ball bearing which free float axially in the end housings), journals and the bearing housing. Machining was done and motor returned to site; soon found abnormally high vibration (bearings had static loaded grooves - possibly due to transport). 2 different sets of new bearings have been installed and vibration was worse than previous field operation in both cases:

Unloaded Shop tests look as follows

Non Drive End:
Horiz - Predominant 1x, as well as various other earlier orders during extended run interval. Overall ~ 0.24 in/s
Vert - Predominant 2x. Overall ~ 0.23 in/s.

Drive End:
Horiz - Predominant 1x, some 2x. Overall ~ 0.10 in/s
Vert - Predominant 2x. Overall ~ 0.35 in/s.
Axial - vibration varies across first 4 orders changes with running time.

Drive end points show significant activity at 44 x running speed with sidebands. Probably high frequency ressonance from bearing degradation but I'm not positive.

Did a coast down and everything seems to point to eccentric rotor - abrupt drop in vibration when power is cut, significant change in vibration when end housings are loosened/jacked essentially changing rotor centerline. Vertical goes down to ~ 0.12-0.15 in/s still bad but much better.

So I guess my question is given that the motor ran after the housing and journal work how now could it be eccentric? And has anybody ever seen similar behaviour?

Thanks,
Curtis

 

[rob768]

1x and 2x components indicate alignmment problems. 44x with sibebands may be a rub or a gear. Can't tell withour more information on components.

 

[nfldengr]

I'm still confused how you get misalignment when the machine is running uncoupled in a shop. Plus the housing and journal work was done in September and it passed a run test in shop so I’d be suprised if it were misalignment due to improper machining/fits. Only after are we changed out the bearings are we getting this problem with high vibration uncoupled. The 44x with sideband at 42x and 46x only seemed to come in on the axial direction when changing the end bell position so this could very well be a forced rub. However the 44x has always been prevalent over the course of testing.

What other components/info would be of value? I can send a variety of spectral and overall time domain responses to anyone who would care to see them.

Curtis

 

[MikeHalloran]

WAG:

Reduction in vibration on removal of power suggests that the rotor might not be centered axially in the stator, as if the end bells are not as symmetrical as they might appear and were swapped end for end during repair, so in the powered condition the motor is exerting a large thrust load on one bearing.

This would cause a vibration if the bearings have some axial runout. If so, more expensive bearings might help. Adjusting the bearings axially so that neither sees a thrust load would help more.

Or, it could be something else entirely...



Mike Halloran
Pembroke Pines, FL, USA

 

[spciesla]

Is this a 2 pole machine? If so, do you have enough resolution in your spectral data to determine if the 2X vibration is truly 2X or 2xLF (120 Hz / 7200 cpm).

I suspect that the 44X is the rotor bar passing frequency with 2xLF sidebands.

I am a little confused about your bearing setup as you refer to "wiped" bearings and deep groove ball bearings. "Wiped" is typically a termed used to describe journal (sleeve) bearings.

You could have a fit issue with the repeated removal/installation of bearings, perhaps the clearances have opened up. A cocked bearings is also a possibility.

 

[nfldengr]

For MikeHalloran:

I asked the same question about the end bells being swapped end for end and the shop adamantly said it was not the case.

For spciesla:

The way this motor is set-up is you actually have the ball bearing floating axially in the end-housing. There is a radial clearance between the bearing outer race and the housing bore which allows the bearing to float axially centering rotor in stator. End play in the motor is ~ 0.120 in axially (distance bearing can slide in housing while operating).

In terms of the rotor bar pass frequency: there are 40 rotor bars on a two pole motor so a frequency of 38x should be expected. When I said wiped (maybe a Canadian colloquialism) it meant they had been destroyed. And the bearing had spun on both the shaft journal and in the housing.

In terms of fits - have measured the journals, housings as well as shaft run out and all have met manufacturers specifications.

Thank you both for the input. I still feel I haven’t found the root cause yet.

 

[electricpete]

I agree with spciesla you need to check whether that "2x" is really 2x or 2*LF on a 2-pole. I don't think you've told us the speed yet? What appears to be 44x could perhaps be RBPF + 2*LF on a 4-pole motor. I tend not to place too much weight on the 44x but you certainly have indication of problems at the low frequencies.

What was the reason for the original bearing problems? I assume this is horiziontal motor?

MikeH - I didn't get what you meant by "axial runout". I can't say I've ever heard that term.

650hp is pretty big for a ball bearing motor. I would take a close look at the axial positioning aspects. Is there anything that provides a preload of the bearings?


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

 

[nfldengr]

Electricpete,

The motor in question is a Westinghouse 5010H 2 pole motor full load stampled speed is 3553 RPM. It would be valuable to take a high resolution look at the lower frequencies. I do have a question though in terms of the extent of slip speed as the motor is being shop tested unloaded how pronounced would it be. Will the line frequency be exactly 60 hz could there be any noise?

The original bearing problems where likely insufficent maintenance practices (infrequent/inadequate greasing). And yes this is a horizontal motor.

The bearing configuration is better explained in the post just previous to your own. And there is no axial preload placed on the bearings.

Thanks for the input.
Curtis

 

[electricpete]

Since there was work on shaft seat and housing bore, it would be logical to double-check bearing fits and concentricity carefully, but it sounds like you already followed that path.

I see wavey washers on most but not all horizontal motors. Clearly it is not a mandatory requirement, but it seems to me that without one, you rely on the weight of the machine which only preloads the bottom of the bearing (not as effective as axial preload). In effect you still have looseness equivalent to bearing internal clearance. Since bearing internal clearance increases with bearing size (and therefore with motor size), a preload washer is more important for larger motors. You didn't say the bearing size, but for 6320 bearing the unmounted internal clearance is in the neighborhood of 1 - 2 mils. Maybe it's possible that the lack of preload makes that clearance show up as a looseness? And when you moved the rotor you may have applied some small axial preload from magnetic centering force which reduced the vib?

That's just a thought (I might be way off base.). Off the top of my head, I'm not even sure how to prove/disprove that short of a home-made modification to temporarily install wavey washer. I'd be interested to hear what others think about this. Have you seen large horizontal ball bearing motors without preload washers?


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

 

[MikeHalloran]

Pete, 'axial runout' is a measure of how parallel the thrust faces of a ball bearing are to the bearing ball path, and hence to each other. It's normally a pretty small number, but it's vanishingly small in good bearings, less so in cheap noisy bearings.



Mike Halloran
Pembroke Pines, FL, USA