Motor Bearing Spun

I am repairing a 3 phase geared motor that mounts with the shaft vertical, and the gearbox is on top. The lower bearing was making a chirping/squealing noise. Took the motor apart and found the lower bearing was stiff and had been spinning in the aluminum housing. There is a thin spring washer in the bottom of the housing as well that pushes up on the bearing. The bearing is 40mm OD 17mm ID, and 12mm thick.

My question is, what are my options regarding the housing being worn. I assume because of the spring washer the bearing should kind of float in the housing, but it should not spin. lock tight, epoxy, bore out housing for larger bearing, bore out and install a shim, other ideas?

Thanks for your help!
Brian

Bore it and sleeve it and then bore it to a close slide fit for the bearing race. An aluminum housing needs a steel liner for the bearing seat anyways, so boring for the next size bearing is not really a good option to take (if there were even to be sufficient meat in the endbell to do this).

Agreed. I have repaired lots of these failures. There usually is some of the original surface left to indicate on for centering. Bore it to clean up and for a wall on the sleeve more than a few thousandths. Make the bore of the sleeve slightly undersized and the OD just fit the housing. Loctite is fine here. After it sets, bore the ID for a close slide fit as Hu says.

Bill

So if i were to make a steel sleeve, what would you suggest for an aloy. Any issues with 4140 as I have plenty of drops. And what do you think a minimum wall thickness on the sleeve?

kb1cjp said:

So if i were to make a steel sleeve, what would you suggest for an aloy. Any issues with 4140 as I have plenty of drops. And what do you think a minimum wall thickness on the sleeve?

Click to expand...

D-2 comes to mind.....

PLain ole mild steel will suffice just fine.

Also that skim cut after you loctite it in place, if you use 4140, you
might work harden it, and run into problems.

Not going to work harden 4140 doing this...

1018 or any mild steel would work well. If you had it on hand bronze will also work. I wouldn’t use 4140 purely due to galling when driving the hardened brg housing into it. I try to make sleeves at least 0.060” wall.

RC Mech said:

Not going to work harden 4140 doing this...

Click to expand...

I disagree, OP may try using negative rake carbide, to take out .003....BTDT
But not with a bushing loctited into the motor housing.

Green locktite. 626 iirc.

RC Mech said:

. . . wouldn’t use 4140 purely due to galling when driving the hardened brg housing into it. I try to make sleeves at least 0.060” wall.

Click to expand...

That's typically an easy slip-fit. If it needs driving, it needs re-doing.

Gordon Heaton said:

That's typically an easy slip-fit. If it needs driving, it needs re-doing.

Click to expand...

Obviously not “driving” as in with a 10 lb sledge. “Slip fit” is the exact problem right now. If we want to get technical the brg manufacturer will publish limits on fit. My WAG is 0.0005-0.0008” larger than nominal.

Been-everywhere Done-everything Doug, next time you turn three thou off 4140HT anything, be sure to take a video.

I've literally done hundreds of these, fractional to 5000hp. Same step for most

Visual inspection of endbell for cracks.

Look at bearing boss wall thickness. There should be plenty of meat to bore and put in a sleeve.
A steel, aluminum, cast iron, brass sleeve works well. Don't over think it. It's not a space pen. I almost exclusively used mild steel seemless tubing out of convenience.

Set up in lathe 4 jaw chuck. Dial
I'm using any remaining reference
Surfaces for the original bore.
The outside stator alignment spigot and face should also be used to dial
In.
Grip very lightly as you will distort the crap out of it!
The stator spigot, bearing housing, motor shaft opening and motor mounting flange/spigot should all
Run concentric/parallel.
Bore out and toss in a sleeve. 0.05" thick or more wall.
Be cautious when boring, often aluminum endbells have a steel insert cast in place for the bearing fit. They can come out funny sometimes.

I always preferred to make my sleeve +0.0005" plus 0.005" per inch diameter interference.
Light heat on bore and chill sleeve.
Drop in place.
Let cool.
Verify alignment again with dials.
Bore 0.0002"-0.0008" clearance to the bearing od.
EASA publishes recommended bearing fits for most sizes and applications.

I cannot caution enough to not distort the endbell with your chuck.
Super light grip. It's really a fine balance between safety and not distorting it.

Considering it's a vertical
Motor to a gearbox your dailing will have to be excellent or it will add stress/preloading and fail premature.
After re assembly I would place a mag base on the motor shaft and run dial along the motor mounting face and diameter to verify concentricity.
You should see no more than 0.001-0.002" of deviation.

The spring washer is to preload the bearing slightly and the sliding fit is to accomodate thermal expansion of the rotor. A press fit will shorten bearing life considerably.

Stirling said:

I always preferred to make my sleeve +0.0005" plus 0.005" per inch diameter interference.

Click to expand...

Can I guess you meant "plus 0.0005"? I'm sure you're not doing .005" per inch presses...

I find that about .002" interference is max for the sleeve insert fit into an aluminum endbell. Also use some Loctite sleeve retainer. Aluminum will want to gall if you put more interference on it. With a cast iron endbell, you can use more interference.

The aluminum endbells with the cast in steel liner I bore until the liner is completely removed. This usually requires a bit heavier sleeve because of extra stock removal to get the bore round after the liner has dislodged and come out.

Always bore the sleeve to near finish ID (leave maybe .030") before finishing the OD. You don't want to force a heavy wall sleeve into the endbell (might crack it). The sleeve may also end up loose if you remove a lot of material from the ID when doing the final boring operations.

I usually use about a .050" wall sleeve for bearing seats up to 150mm, then get a little heavier.

Note that getting a .005" interference sleeve to start into the housing might be difficult. Bore a little step (or turn a smaller diameter step on the OD of the sleeve) for about 1/8" of length, with no interference. This will let you hammer the sleeve in easily to begin with, as it can be started 'square' to the bore.

As was noted above, the endbell can be distorted easily by chucking. I still chuck 'em tight (unless it is fragile). The lobing introduced by chucking force still needs to be balanced. That is to say, the low readings at every jaw should be close to identical, and the humps in between should be close to identical. Endbells often have some degree of warpage that you have to cope with anyways. This can be frustrating for someone to deal with who barely does any lathe work, and who can barely chuck up a piece of solid round bar on center.

You can cheat by indicating on the shaft hole bore (if it has not been rubbed out of round). And if the face of the bearing boss has been machined, you can indicate against that to get the endbell running true in the other plane.

I prefer to do the job on a Bridgeport. Clamp it down any way that works, is solid, and doesn't distort the bell. You can put the indicator on the part and watch for movement as you tighten the clamps. Then tram the head to the spigot on the end bell. Position the knee so you can lift the indicator over your clamps.

gbent said:

I prefer to do the job on a Bridgeport. Clamp it down any way that works, is solid, and doesn't distort the bell. You can put the indicator on the part and watch for movement as you tighten the clamps. Then tram the head to the spigot on the end bell. Position the knee so you can lift the indicator over your clamps.

Click to expand...

Can you charge for all the time it takes to do all that? Or make any money at it?

I've used the mill a time or two for a very awkward to hold endbell. But the customer paid a premium, because of the abnormal setup.

Milland said:

Can I guess you meant "plus 0.0005"? I'm sure you're not doing .005" per inch presses...

Click to expand...

Good catch. I missed a zero!
Yes 0.0005" press fit per inch
0.005" would likely crack a smaller bore

Thanks for the catch

I've used a mill for lots of odd endbells.

Make a set of L shaped brackets.
Bolt through upside down to mill table. Use endbell stator bolt holes to bolt endbell to the underside of the L.
This allows you to bolt up against the stator face register guaranteeing zero distortion.
Honestly I preferred a lathe in most cases, but when I would get multiple of the same endbell a day I would use this as the setup would not change.
Also helps with SUPER flimsy aluminum endbells. Like chucking a tin foil hat!

Again I cannot stress the distortion issue. It is so important to have a round bearing bore when you remove the part from the chuck.
Not to mention the more off centre the bearing is. The more off centre the rotor is from the stator, this causes loss of hp, increases spot heating on the winding and creates a magnetic vibration during rotation.

Not to mention vibration damage to the driven equipment.

Also noted above. Make sure your bearing fit in the enbdbell is clearence!!!!
If not thermal growth will take it out very fast!!

1.5750-1.5753 on the finished bearing bore, you want the bearing to just be snug. Line up the endbell using the step where it locks into the motor housing to dial in the part. I also prefer a mill for this work, but I use a big Cincy vertical, most of the ones I am doing are 4-8" diam. Unless the bell will fit squarely in the chuck jaws, you often spend more time trying to get it lined up in the lathe because it rocks and crawls with the chuck jaws trying to grab the radiused profile. Throw some bolts with nuts under the bolt holes in the outer edges of the endbell and use those for jacks to level it. Bore it about 1/4" over your bearing size, so you have some meat to work with (1 3/4" will be fine for your application).

Make your bushing on the lathe you don't have tied up boring out the endbell. Modern chemical engineering is your friend. Bore the ID on size as specified above. Turn the OD to .001-.002 under the housing bore diameter. Green loctite and bump it home with a block of wood. Wait 20 mins for he loctite to set up... done.

I used to play the shrink fit and leave enough for to bore it again game, but the green loctite eliminates that. If you have the bore within .002, the most it can be out of line is .001. If you can give the bushing a spin with your fingers in the bore before the loctite sets, it will self center to withing tenths. If you are really worried about the loctite failing, put two opposed 6x32 grub screws in the intersection of the bushing and housing.

Thanks everyone for all the input. This project has got bumped do to some higher priorities, but will re-visit shortly. Will end up doing this on the lathe because it is small and will be easy to support, and plenty to indicate off of. Most of the steel that I have this size is of unknown alloy other than the 4140.