I'm not aware of the particular application, but ifyou have changed bearing type, you may well need to shim for proper clearance (well, plenty to allow for adjustment by the spindle nut). That will be particularly obvious if your nut is currently threaded on to the end of the threads on the spindle.

The bearing races should be seated, preferably by pressing rather than thumping, so the adjuster nut should not be pulling bearing races into position.

I completed a rebuild of the spindle of a Sieg C3 using taper roller bearings. This necessitated the making of new spacers from aluminium…the plastic ones are not suitable for preload applications. Yes the inner races should be a sliding fit on the spindle, so the correct preload can be achieved. Same would apply to AC bearings, perhaps our bearing guru,  Ketan would care to comment?

Edited By John Rudd on 21/09/2017 12:01:32

Edited By John Rudd on 21/09/2017 12:01:45

Firstly, I'm no expert when it comes to bearings…..but as my Sieg came with standard ball raced bearings which dont rely on preload, it made sense to replace the spacers with metal ones as the plastic would tend to crush slightly….

As for a sliding fit on the shaft, they need to be a tight fit, but not that tight they cannot move….Think of the front wheel bearings on a car ( that is rear wheel drive…) that are tr bearings……

As for fitting new bearings, you could use thread bar nuts,washers and spacers to pull the new ones in to place if you dont have access to a press…..

This doesn't seem right at all.

The bearing arrangement is essentially that drawn on page 2 of the document you can download here:

http://www.model-engineer.co.uk/news/article/lathe-bearings/19136

If you have longitudinal movement, then you should be able to push the spindle to the left and take up the slack with the c-spanner.

If you can't take up play by adjusting the bearing then either the bearing is stuck OR have assembled it incorrectly so the c-nuts are not contacting the bearing.

Leather mallets should not be needed.

If the bearings are not seated properly, they will move as soon as you use the lathe in anger., Stop, readjust the preload and continue…

The hard plastic spacer is more than man enough to apply preload to the roller bearings, I think they actually need rather less than the original ball races. Shortening it is not a big issue, it needs to be about 1cm shorter to make sure the bull wheel is correctly aligned. At a push you could cut it with a junior hacksaw and finish it by hand.

Neil

How to press the bearing into place without a press. Run some allthread through the headstock with a bar right across thee headstock at one end and a shorter bar across the bearing. Tighten carefully making sure you keep the bearing dead square. This is a roller bearing outer but it works with ball races as well. The bar needs to be long enough to put the load on the outer race.

Neil,

Colin has angular contact ball bearings. they are same size a the radial ball bearings he replaced, so he does not need to shorten any spacer.

Ketan at ARC.

Edit: it looks like our messages crossed.

Edited By Ketan Swali on 21/09/2017 14:40:38

FWIIW, and based on my recall of SKF design advice:-

For a bearing fitted onto a rotating shaft, the inner ring should be a tight, interference, fit on the shaft, otherwise the ring will creep on the shaft. This would apply to a lathe mandrel.

The stationary outer ring needs only to be a push fit in its housing. It can then move to apply pre-load if required.

For a rotating load on the outer ring, the reverse is the case.

Clive Brown

.

We appear to be of opposing view on this

… I suspect that SKF's advice may refer to a rather different assembly.

Would it not be exceedingly difficult to assemble and dismantle a typical [hobby] lathe headstock, if the bearings were configured like that ?

MichaelG.

[ready and willing to be educated on the matter]

Edited By Michael Gilligan on 21/09/2017 16:42:58

Point taken, not that I make a habit of dismantling my Boxford.

My offering was based on SKF general advice, but you are correct, the typical lathe headstock would not allow much of an interference fit.

Having said that, I think it would be a good thing to have a reasonably firm fit on the spindle, whereas the fit of the outer races isn't perhaps so demanding.

Not a good analogy, the axle on the front wheels of a rear wheel drive car doesn't rotate, the outer race does. Clive is correct, the race that revolves relative to the load (usually but not always the inner race) should be a tight fit on its axle/housing. The fact that this makes it difficult to adjust preload on a hobby lathe doesn't change this. As others have said it's all a matter of what you mean by tight. Clearly they have to be tight enough to avoid creeping on the spindle, but loose enough that you can adjust them without bashing them with a hammer. Not easy. Having the outer race adjustable would be a much better design. If in doubt consult the manufacturers book of words

**LINK**

Or see the link I posted earlier which has plain advice from Timken aimed at machine tools with beautiful drawings.