Installing a ball bearing

"Jim Wilkins" fired this volley in news:mk89h1$mft $ snipped-for-privacy@dont-email.me:

Not TOO much of a problem, so long as you have a 'keeper' to prevent it's migrating out of its cavity. That, and so long as the speed is kept low, it won't damage the bearing to 'roll' on its outer periphery.

But (cough!) why would you be pounding on the INNER race to remove a bearing? The access hole at the bottom of the cavity (through which the shaft will penetrate) can be only a few-thou smaller than the outer race, and it will be retained on that end just fine. Then you'd be able to use a proper outer race-sized drift to remove it, were the fit too snug.

Lloyd

Or he can make the bearing a slip fit and use some bearing retaining loctite to keep the outer race stationary .

The reason is because I'm copying the existing wheels closely in hopefully better metal instead of re-engineering a heavily loaded design I don't fully understand. The HF wheels have an 18mm center hole that catches a snap ring on the 17mm axle. Presumably the solid metal from 18mm out to 40mm stiffens the wheel rim against distortion.

I could drill three holes for push-out pins and use one wheel to hold and guide them while pressing the bearing out of another.

-jsw

If the concern is that you don't trust your mics, remember that you have the bearing that's going in the bored hole. Treat the mic as a comparator and the bearing as your standard. Ideally you want a line to line fit to at most a couple tenths interference. You can interpolate the graduations closely enough for that. I'd be more concerned about the machine used for boring than the mics.

On the other hand, if you just want to be sure the bearing goes in OK the first time you can always heat the wheels if the fit after boring is too tight. Assuming the wheels are steel, a 300F temp rise will open the bore .003. If you can get within .001, you should be good. You do need to plan and work carefully when it's time to drop the bearing into the hot wheel - the last thing you want is to get the bearing cocked and jammed in the bore. It doesn't take long for the bearing to warm up and expand.

Excellent idea.

Because you won't have any side loading (to speak of) acting on them, it shouldn't be a problem. The taper should keep things together just fine.

I forgot to say that you can firm up a race in a seat by tapping a center punch several times (equidistant) around the seat ID. The dimples will help hold the race but not be high enough to deform it. I've done this on disc and drum brake hubs for decades, all without a single failure or return.

Obviously, there are no keepers or access holes. Chiwanese Tech, y'know? ;)

Larry Jaques fired this volley in news: snipped-for-privacy@4ax.com:

??? He's MAKING THE WHEELS HIMSELF! Such features don't just 'come' like gravy in the pan. One MAKES them.

Lloyd

I practiced measuring the setting ring. This Enco 508-00150730 inside mike doesn't have a good enough feel to repeat to better than half a thousandth.

The machine is my 1965 South Bend 10L, with a 3/4" boring bar that I cut and ground a 5/16" square HHS bit for. It's more rigid than any other boring bar I've used so this is my first go at boring a smooth hole to a more precise diameter than Oilite requires.

On outside diameters that I can measure to 0.0001" the lathe can shave a few tenths with a freshly honed bit.

-jsw

"Jim Wilkins" fired this volley in news:mk9oh7$54k$1 @dont-email.me:

So, turn a "setting standard" that IS the couple of tenths smaller than the bearing, and use that to feel out your hole as you enlarge it a half-thou at a time (after boring to a nominal under-sized dimension).

Lloyd

My usual method is to bore with a coarse enough feed to leave ridges that the pressed-in part deforms, so there's more compliance than if the bore were smooth. In this case the steady (

I suspect that the HF trolley is a "value-engineered" copy of a well-designed US original, perhaps this one:

-jsw

I missed the fact that you're using an inside mic. I have a couple of that style, but rarely use them. For this sort of thing I'm more comfortable with a telescoping gauge and outside mic, thus my suggestion to use the bearing as a standard. It seems to me the telescoping gauge gives a better feel than the inside mics and makes a direct comparison between the bearing and bore easier.

My question really resolves to what that feel should be, if I can't measure the ID to 0.0001" to match the interference fit from a table. Your previous response suggests that a drop-in fit would be fine as long as the axle shoulder and snap ring keep the wheel and bearing together. Subsequent machining on the other side will be much easier if I can remove the bearing and chuck within its recess rather than by the rim.

Maybe while doing it I'll learn how to bore and measure accurately enough for a future job where the bearing can't be allowed to slip.

-jsw

Thanks. I've had trouble with telescoping gauges previously, probably because I didn't have a rigid enough boring bar to get a smooth finish. I just measured the inside mike's setting ring with one and hit 1.0000" on the first try.

-jsw

I've had the 3/4" boring bar for a long time, but was stuck on how to cut a 5/16" lathe bit for it to length at the correct angle until I tried the 7" angle grinder with a cutoff disk.

-jsw

No, this is: Final velocity = (exhaust velocity) * ln ((mass at liftoff) / (mass at burnout ) )

Essentially, you do not want the bearing outer race to rotate in the hole and you do not want the shaft to rotate in the bearing inner race. You also do not want to have too much of a "press fit" on either diameter as it is very possible to cause the bearing outer race or inner race to contract or swell sufficiently to cause it to "bind" to at least some extent.

If I were doing it I would aim for a hole diameter of the bearing O.D. to perhaps 0.001 larger and a shaft diameter of from bearing inner diameter to 0.0005 smaller. If there is concern about either the outer race rotating in the hole of the shaft rotating in the bearing inner race than just assemble it with some of that bearing mounting goo.

It is not "rocket science".

But if he's following the old pattern exactly...