I came in this morning to some comments regarding my lineage and abilities due to a "bearing failure" yesterday. Apparently I didn't "pick it up?". Well, first of all, I don't collect data on it because I would have to cross the plane of a guard to reach it which is a no no here. Secondly, upon inspection, I see that the tab is broken off the lock washer, the nut has backed off, the roll has shifted and the journal has turned in the bearing. The bearing itself is OK. I have not been able to pick up this kind of thing in the past nor do I think I would have picked up this time except if I could hear the lock nut or washer ringing around inside the housing. Any comments?

Ron,

I have also been "fooled" by a bad locknut. (Fooled to the extent that I knew something was wrong but couldn't put my finger on the exact cause until the lock nut backed off and welded itseld to the motor end bell.) Lost a vacuum pump motor at about $20000 and several hours of production at $9600/hr. It turned out that the motor shop had reused the locknut and the tab engaging the shaft had broken off. It wasn't all bad though. They saved $12 on the lockwasher.

If they are serious about catching this the next time they won't mind a few hundred dollars to mount an accelerometer so that you can safely collect data. If not, then they are probably looking for someone to blame for everything anyway and vibration guys make excellent scapegoats. Even if you have an accel. mounted this will be difficult to catch. All I saw that would have helped was an elevated floor level in PeakVue. I think that listening to it with the yellow handled analyser would have probably been better. The best method employed was to change motor shops. Han't happened since.

Good Luck,

I've picked up a couple in the past. First indication I found was looseness on a pretty new bearing. I opened it up myself, and found that locknut had backed off, but the tab was still bent a little. I reset the clearance, and installed a new locknut with a chisel I'd ground down flat to fit the tab.

Thanks for the replies, I feel better already. This has happened before so as you say Danny, they wouldn't mind spending a few bucks on remote accels. It will be done............

It's pretty easy for a locking screw or collar to come loose, and the shaft just creeps around the race at about 99.9 of shaft speed. No impacts and not much screeching either. Someday the shaft will have a groove worn about 1/8 inch deep.

I liked to peek at the shaft/bearing interface for 30 seconds with a stobe at 1X to see if everything is staying in synch. (unintentional Timberlake reference there)

Good pun........
If the bearing race is visible, unfortunately this one is inside a housing

Those I have found has been when trying to balance them, it makes a complete different creature to balance when the bearing is stuck on properly, when it´s loose, balancing is like dart. Olov

There is a hefty element of luck in finding loose sleeves.
Felt roll bearings having withdrawal sleeves that come loose (for whatever reason) can go from being OK to being very un-OK in a matter of hours. I've seen rolls come out where the damage has been substantial: lost 1/4 inch or more off the shaft diameter, sleeve all but disappeared, inner ring so hot that the rollers have "sunk" into it.
Vibration data taken a few days before showed no problems.
If you are lucky enough to capture data while it is failing, you tend to see looseness.

Ian

Back in April, I, too, was charged with a 'miss' for a lock nut coming off despite a note in my predecessor's report in January (and every month thereafter in my reports) that there was a pile of metal dust under the OB fan bearing. Absolutely nothing in the data. When they finally got around to changing it, there was an 1/8" groove in the shaft where the bearing was sliding. When I got a little defensive about how it could be called a miss, they said, "Well, you never said the bearing was loose."

I have found that ultrasound measurements (airborne and contact probe) are a great means of detecting loose components (on shaft and fixed locations). I know that this is primarily a "vibration board", but you can't use an adjustible wrench for everything!

Exactly Walt, your input is welcome. One of the clues to this problem was that the roll itself had shifted about 1" to the right. There was also considerable black debris on the bearing housings (both ends) and on the floor. So, in this case, a good eye was all that was required to catch it.

How did it shift 1"?
I can't picture that.

Ron, how are those lock nuts tightened? A rod and a hammer? A key with a hook and a hammer? A hook key and an extension? A hydraulic nut torque machine? I had once a loosening nut plague on several drier cans tending side. Not until all was tightened with the hydraulic nut machine, all problems just vanished. But it was not the machine that did it, it was their crew training man:

Detailed inspection showed that the shafts (20+ yrs old) were so worned from all cleaning and polishing that the sleeve did not have a good cylindrical surface to clamp to anymore. The nut machine´s vendor trained the mechanics and when he showed the machine he was insisting in manually doing an extensive work blueing/polishing the shafts almost perfectly cylindrical. So the terrain for the sleeve was much better, allowing a real tight fit, and still give the correct bearing play.
Thinking of your machine, could it be that the nut gives a good brg play much too early before the sleeve does not sit tight on the shaft? Regards Arne

Pete; the other bearing is a floater (NDE) and was pushed so far that it's lock washer was squished out.

Arne; always a pleasure to get your input the roll is very old, the bearings are straight bore to a shoulder and the lock nut holds it in place. You're probably right about the wear. I didn't do the last install but I can imagine the guys saying "boy, those went on real slick! we'll be done before morning coffee!"

Oooh, so there is no grip at all, just a hope that the inner ring won´t slip. A bit optimistical. Why not a K - cone? Do you need that shoulder due to a critical axial position? Couldmaybe still let it stay against the shoulder if you shorten the taper sleeve a bit before installing so it does not stick out before final tightening.
Whatever, next time could you include strong permanent Loctite for threads next time? I cannot reach the number now, but the good one that really grasps steel. Take care Anre

Thanks Arne, when the roll is ready for re-install, I'll hover around and make the suggestion.