Bearing tolerances

I am in the process of attempting to repair an article, initially by creating a drawing both of the broken part. In doing so, I have come across a 6000RS bearing, nominally 26 x 10 x 8, whch is a very sloppy fit in it's housing. I suspect the sloppiness is because the housing itself, made of cast iron, appears to have been worn away, measuring 26.5mm, but on measuring the bearing I find that the measured OD is 25.98/25.985 and the measured thickness is 7.96/7.97mm. I have used two micrometers and two dial calipers, thus taking four measurements for the diameter and the thickness, and all four devices show similar readings.

It seems to me that this bearing is undersize in both dimensions, but having attempted, and failed, to find specified tolerances, I wonder what other people think. Is it possible that the bearing has itself been worn down, but I can't see any obvious wear marks?

I will be replacing the bearing anyway since the ID does show some damage, and I may have to replace the axle as well but that has yet to be determined, so really this is of academic interest only.

Cheers,

Peter G. Shaw

I think for a bearing that size it can be 0 to -10microns so 26.000 to 25.990

Have a look at table 8 on this page

Edited By JasonB on 13/04/2021 15:24:13

Or the second table on this page, where ABEC 1 ( 'Normal' ) allows -13um i.e 25.987

Further down the page a bunch of other bearing tolerance standards are listed. Possibly one of them allows slacker fits?

Dave

Edit: pesky smilies, '

Edited By SillyOldDuffer on 13/04/2021 16:13:18

If the bearing has been spinning in the cast iron housing enough to wear half a millimetre of metal away, it's more than likely there has been at least some wear on the bearing in the areas that rubbed, resulting in the very slightly reduced dimensions. May well have a nice polished finish on the races from being "lapped" by the cast iron.

Depends too on the brand of bearing too. Cheapo Chinese bearings could be all over the place. Your best shot might be to buy new SKF etc bearings, take your measurements off them and size your new hole accordingly.

With half a milimetre of wear on the old hole, Loctiting would be a bodge at best. Better to sleeve and rebore or make new.

.

Depending, of course, on the ‘duty’ of the bearing ... If it can be centred in the housing, then Loctite 641 would probably do the job.

The Technical Data Sheet has a cure-time curve for 0.2mm bond-gap, and I doubt the extra 50microns would cause too much trouble.

MichaelG.

.

Ref. http://tds.henkel.com/tds5/Studio/ShowPDF/243%20NEW-EN?pid=641&format=MTR&subformat=REAC&language=EN&plant=WERCS

Edited By Michael Gilligan on 14/04/2021 10:18:28

Do you have the original bearing, or is this another bearing you have purchased?

Bearings can be made to order, and quite often for various reasons are made to a non standard outside diameter. It is often seen in model engines where the bearing may have a 23.03mm diameter outer race, or a 10.5mm inner race or the inner being 13mm and the outer being 23.5mm.

0.5mm wear on a cast iron housing is a lot and I would be looking for the real reason for the failure instead of the quick fix. Loctite does make a bearing repair material for taking up the gap of a bearing in a worn housing. But relies on the rest of parts to centralise the bearing as it sets.

As for being tolerances it is a mine field to look through. You have the concentricity of the inner race to inner bore, the outer race to the outer bore, the centre line of the race to the bearing width (can be ordered offset by specified amounts) the roundness of the bearing balls, the cage type etc. Then the internal clearance as well. WIB bearings make a lot of special order bearings for very reasonable prices. They have a lot of info about bearing tolerances etc

Is there enough metal outside the bearing housing to allow it to be bored out and bushed back to correct size?

If so, buy a new bearing, (Carriage will cost more than the bearing! ) and bore the bush to suit, before using Loctite to secure the bush into the housing.

Everything should then be back to standard without fear of wear in either part causing problems.

Howard

If boring the existing hole, for a bush which you intend to bore to fit the OD of a new bearing, if the existing bore is worn you need to establish where the wear has taken place to centre the new bore correctly so alignment of the shafi is correctly positioned. If the existing worn bore still measures circular it is most likely worn evenly around the original centre but that should be checked.

Emgee

Edited By Emgee on 14/04/2021 13:36:49

.

A common reason is that the bearing has failed, locked-up, and spun in the housing.

It is usually a very simple matter to equalise the film of Loctite 641, by spinning the outer race a little ... it’s just the right viscosity to spread nicely [anyone would think the product had been designed!]

... but if you’re not confident/convinced, just add three little pieces of 0.2mm shim, equispaced around the circumference. Not only will this centralise the bearing, but it introduces some copper ions to catalyse the curing.

MichaelG.

.

Again, I have assumed that the housing recess is still circular, not worn pear-shaped.

A lot of questions, so here are the answers:

JasonB,

I looked at that website, but didn't find that particular table. I suspect probably because I was somewhat bamboozled by all the different tables, tolerances, etc and I ended up going round in circles, Thanks anyway - that certainly would seem to suggest that this particular bearing might just be within tolerance.

SOD/Dave,

Interesting reference which I didn't find before. or maybe I did, but ignored it. As above to JasonB, I do have a tendency to go round in circles, especially when I'm trying to discover information about which I have never had any training.

To everyone else.

What has happened is that a cast iron hub has broken right at the point where the inner edge of the bearing housing occurs. Which also happens to coincide with the narrowest diameter of a V on a PolyV drive. The bearing make appears to be XMCXV and the type is 6000RS. As mentioned in my original post, there is definite wear on the inner diameter so the bearing will have to be replaced. I haven't yet checked, but I do wonder if there may be matching wear on the axle on which the bearing sits. When all said and done, for wear to occur on the bearing ID, that suggests that the ID has been moving on it's shaft.

Now, the shaft, or axle, is actually inside, along the centre line, of a hollow hub. Think, if you like, of a bicycle wheel where the shaft/axle is stationary, but the hub of the wheel rotates around, but is clear of, the axle. This diameter is of the order of 22 mm whilst the axle/shaft is obviously 10mm to suit the bearing ID. My thoughts are to make a replacement, or perhaps I should say addon, hub which will carry grooves for the Poly V drive, carry the bearing housing for a new bearing, and which has a longish hollow spigot to fit inside what is left of the original hub, and glued in place possibly using JB Weld, that is once the broken bits have been removed. I'm also thinking of possibly slightly increasing the diameter of the Poly V bearings in order to slightly increase the thickness of material at the break point. Ok, this will result in a slight loss of gearing, but that will not matter in this situation.

However, the first thing I have to do is to draw the existing, hence measuring the bearing and what's left of the broken hub. From that, I can then move on to modifying the drawing to show the result of any proposed increase.

It may, of course, be a total failure, but it is already that, ie I have nothing to lose in attempting a repair, and even if it is a failure, I will have learned something, because as I keep saying, I'm into model engineering for self education by experimentation. I've also been somewhat circumspect in that I haven't said what it is that I am attempting to repair - the reason being that I have thoughts about submitting it as a project to MEW if the repair proves successful.

Regards, and thanks for all the thoughts.

Peter G. Shaw

My observations seemed quite reasonable ... from where I was, [i.e. up the garden path] ... but now appear to be irrelevant.

MichaelG.

Edited By Michael Gilligan on 14/04/2021 15:01:13

Michael,

Your observations were indeed quite reasonable seeing as how I did not mention the breakage in the first message. Mea culpa, and all that. This has been something of a failing of mine in not explaining things properly over the last half century or more. That, and assuming that people know what I'm talking about. So sorry, and all that.

I think that probably what happened was that I was so "hung up" on the bearing OD tolerances that I didn't relate the complete story.

Regards,

Peter G. Shaw

No problem, Peter

It just turns out that your situation is more like being in heavy traffic on the M25

MichaelG.

So call the bearing a nominal 26mm and carry on. You can size the hole exactly to fit your new bearing when you get that far.

Hopper,

That's very true, and is more or less what I've done. But I do like to check, check & check again whilst drawing it, and it was whilst checking to see if the bearing OD was similar to its housing ID (I expected it to be a push fit) when I found the discrepancies, and then one thing led to another, and another, and another... etc.

And of course, I have now learned something, ie that the bearing can be +0 to - (a few microns) in size. Plus the suggestion that if the housing has worn to the extent that it appears to have, then there is the possibility (probability?) that the bearing OD will have worn, but worn smoothly such that there are no signs of wear.

Regards,

Peter G. shaw