how loose is loose?

[TonyCro]

We have a small motor that has been repaired a few times by our sister companies, and to cut a long story short; it's ended up on my lap (very uncomfortable!).

The bearings seats are undersize by 0.6mil (0.017mm) for both 6217 & NU217. Is this loose enough to cause a higher than expected vibration at load, and light.

Any comments appreciated,
Thanks
Tony.

 

[Tmoose]

Do the bearings slip on ?
If so, That's too loose for a couple of reasons.
1 - Subjected to a one directional load (gravity or belt or ??) the shaft will creep (not spin) and wear the shaft out after a while.
2 - The rotor may move off center within that clearance, and combined with the creep, produce a rotating unbalance vector that would cause a "beat" frequency in the vibration.
The right grade of Loctite would fix (1) to a certain degree, but the rotor would be off center a few 0.0001s of an inch, which WOULD >>change<< the balance enought to be significant, and might require balancing the rotor in its own bearings.
3 - for the shaft to get that loose material was worn off. The chances the material came of SO uniformly so as to maintain journal concentricity, roundness and shoulder squareness is real slim. Cocking or mis-aligning bearings is just looking for noise, vibration, heat, and durability problems

 

[electricpete]

I was about to say "no", I don't think it's a problem.

Certainly we have found quite a few bearings with housings more than 0.5 mils looser than the specified fit (usually H5). Our application is rotating machinery stationary outer race usually not much rotating load (low unbalance and misalignment). Most didn't have abnormal vib. Also not abnormal noise although in our industrial environment we aren't that keyed into noise.

I didn't think about the cocked bearing aspect. Certainly those housing fit specs are established for a reason and if you don't meet them that could be tough to defend.

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

 

[Tmoose]

Hey Pete,

The difference between a loose housing bore and a loose seat(which I interpreted as the shaft)depends a lot on the direction and nature of the load.

A loose housing with a one directional load would not be nearly as scary.

It would be nice to know if TonyCRo is asking to predict whether there WILL BE a problem, or helping to establish a suspect list for an existing problem. Of course there's the blind test aspect too.

If the speed is real high then the inner ring is even looser.

 

[electricpete]

You're right I misinterpretted it as housing.

I think the answer is easy for shaft 0.5 mil looser than spec (k-something).

Inspection of the removed bearing should show NO evidence of spinning of inner ring on shaft. If there is any evidence of spinning at that location it is a big problem with potential for catastrophic failure.

In contrast it is fairly common to see bearing outer race spinning in housing and not necessarily a reliability or vibration concern.


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

 

[TonyCro]

Sorry for any confusion, internally we refer to the shaft as having a seat and the static support is the housing.

The bearing inner ring is not a tight fit onto the rotor; it pulled off cold and checks later showed the rotor to be undersize; but no signs of the inner ring rotating.

But, I was not looking for a particular vibration cause on this motor, more just a general 'feel’ for what might be enough to cause a problem, based on experience of what people have seen!

Electricpete came up with the best response, tolerances are there for a purposes!

Thanks
Tony.

 

[Mendit]

Bearing to shaft interferance is not there just to ensure the bearing is fixed on the shaft. Bearings are manufactured with internal clearance, when the bearing is loose in the box this clearance is larger than required. When installed on the shaft, this clearance is reduced to the proper amount, more so on the roller bearing than the ball bearing. roller bearings are fitted tighter than ball bearings. Failure to reduce the clearance can cause internal bearing skiding, lubricant failure, leading to premature failure.
Bearing vibration frequencies are usually predominant.

What keeps the inner race of your roller bearing on? these are not usually secured with a nut, because the rollers will not pass over the outside of an installed nut. This could very easily be a problem.

 

[Tmoose]

http://www.mapro.skf.com/knowledge/knw_rbtech07.htm

 

[Hippo41]

Most of what needs to be said has been said, but:-

Shaft and housing tolerances are specified by the bearing manufacturers for a reason. Following their guidelines you will probably achieve optimum performance under normal conditions.

ROundness of the seats is as important - and there are simple ways of checking that too.

However, in my experience there are many "motor repair facilities" (not all!!) that pay scant regard to these recommendations and simply fit new bearings to achieve quickest turnaround/ sales. Many that I have visited over the years don't have the right calibrated measuring equipment to check anyway.

There are unfortunately as many maintenance managers that don't realise the importance of getting these things right (usually they are under pressure from "production" or "higher management" to get the plant running again).

In the real world (i.e. a production environment) it's not always easy to take the time to get everything right. But if a machine is that important that it costs x dollars downtime per hour, then it makes sense to a) have a spare and b) take the time to investigate ALL aspects of the failure to ensure that any repair will function well when required.

Different greases will also affect your vibration readings. Not only different types (lithium-calcium vs polyurea) but different manufacturers have varying particle inclusions that will be 'heard' by vibration equipment.

Lester Milton
Telford, Shropshire, UK