Seating bearing into Ebay ballscrews & other issues

[misgis]

I got those cheaper ebay ballscrews. It came with BK BF bearing blocks.

One end of the shaft fits 10mm bore bearings and the other end near the shaft coupler to the stepper, fits 12mm bore bearings.

In the default state and at room temperature, it is not possible to fit the bearing into the shaft without hammering or other forms of impact. So how do you seat the bearings?

One solution is thermal expansion I suppose. It wouldnt sound too nice to put a flame torch or heat the bearing assembly. So maybe cool the shaft? Maybe heat the bearing block to a certain acceptable temperature? How did you do it? What temperatures?

Another thing is the axial loose fitting nature of one of the locating bearing blocks, I think the one with the angular contact bearings. This degree of play may become a problem presented in the form of 'backlash' even though the ballscrew is probably close to zero backlash on its own. How do you fix it so there is no axial play at all due to the axial play of the bearings?

I guess preloading? so on the normally non-locating bearing end, I need to apply axial pre-load from that end to engage the shaft to eliminate the small degree of play at the angular contact bearing end? I will be able to do this since the bearing on the other end also has a tight tolerance fit with the bearing block, albeit that side technically non-locating.

Finally, the issue of dust. It seems the angular contact bearings bearing block doesnt have seals. Do I need to put one in and also how and what grease do you use to lubricate the bearings? Similarly, what lubrication for the ballscrews?

How do you keep dust out of the ballscrew and does it matter? Looking at most DIY gigs, it seems like people's answers are generally no, but I put leadscrew covers on my standard mill even while most dont do it so I might do it here too.

[Walky]

I just used very fine grit sandpaper (using oil to make it smoother), testing the fit regularly. Now it only needs hand pressure to fit.

[misgis]

Is it meant to be such a tight fit normally?

The problem with sanding is there is a possibility you cause eccentricity but I guess its very minor anyway. I have to remeasure my shafts later, but I think the shaft is a little over sized?

Also, did you sand the shaft or the bearings? If one will sand, presumably the cheaper one should be the one to take the toll? So sand the ID of the bearings instead of the shaft?

Did you angular contact bearing block have a little axial play in it when you press on the inner race?

Yeah Im afraid that after I fit the thing I cant take it out. Im on edge about whether to use their bearing blocks or make my own since it limits my design layout a bit too. Plus, their angular contact bearing doesnt seem to be too great given the little play it has, I might buy a new one too. THe ballscrew themselves seem to work great though.

[louieatienza]

I'd have to say, get some very fine wet-or-dry sandpaper, at least 1000 and may be 1500 and go at it. They make special bearing heaters, but that might be overkill for something like this...

I have some THK precision ground ballscrews and THK blocks, and with a little force they'll slip right on the bearing journal. I also had some Kuroda ballscrews machined for me (with 3/8" drive journal instead of 10mm) and they were a little tighter than the THK, but I was able to do it by hand. Just judging by these two examples I'd have to say I'm either lucky, or extreme force should be avoided.

I'm not sure if it's possible to deform the inner ring (and therefore deform the inner bearing races) by using force, but if I had to pound it in, I'd get a tube with an ID the same as the inner ring, so you're not banging the rest of the bearing around.

[misgis]

Ok I guess judging from your higher cost ballscrews and bearings, then basically there shouldnt be such a tight press fit.

I have some 800 on hand, and also some diamond paste metal polish so hopefully the combination will do.

I guess I will sand it down a bit. Im deciding whether that be the shaft or the bearings. Im leaning towards bearings which are cheap and replaceable up to a certain point.

Cause even if I manage to fit it by heating and cooling parts, when they reach back to room temperature there may be extra stresses on the different parts too so that may not be good.

[louieatienza]

Originally Posted by misgis Ok I guess judging from your higher cost ballscrews and bearings, then basically there shouldnt be such a tight press fit. I have some 800 on hand, and also some diamond paste metal polish so hopefully the combination will do. I guess I will sand it down a bit. Im deciding whether that be the shaft or the bearings. Im leaning towards bearings which are cheap and replaceable up to a certain point. Cause even if I manage to fit it by heating and cooling parts, when they reach back to room temperature there may be extra stresses on the different parts too so that may not be good.

I'd place the screw in such a way that you can roll it, to keep it as close to round as possible. I'd also back the sandpaper with brass shim stock or whatever so that you don't unintentionally create a divot. You also don't want to "reverse cone" it where it's a tight squeeze then all the sudden it's loose at the shoulder of the journal. It's only a few thousandths so do a little at a time...

[Walky]

Yeah, just wrap the shaft (don't sand the bearings) with the wet/oily sandpaper and then spin the screw. Eccentricity would be so low it will hardly cause any trouble. If you measure the shaft diameter before and after you get it to fit, I bet the difference will be very, very small.

I'm actually glad my chinese screws' end machinings are usually slightly oversized, since it is way easier to fix than to have an undersized shaft which would slip on the ID of the bearing instead of turning with it.

BTW, I'd first sand the last mm (the tip) of the screw's borders just to be sure is isn't somehow wider for some reason (because of a small dent, deformation from an impact or something like that).

[misgis]

I am actually ok if it is a press fit, except when I need to remove the bearings it may very well be near impossible without damage to the shaft and certainly the bearings.

I am also glad they are oversized if they were to not be a sliding fit. But I dont know, wouldnt the bearing be a better sacrifice to make than the shaft? Later I will go measure which one is actually the accurate reading to design parameter and sand the one whos off.

Another question, what is the type of thread on the shafts? Is it M10 or M12? standard course pitch or some special pitch?

[louieatienza]

Originally Posted by misgis I am actually ok if it is a press fit, except when I need to remove the bearings it may very well be near impossible without damage to the shaft and certainly the bearings. I am also glad they are oversized if they were to not be a sliding fit. But I dont know, wouldnt the bearing be a better sacrifice to make than the shaft? Later I will go measure which one is actually the accurate reading to design parameter and sand the one whos off. Another question, what is the type of thread on the shafts? Is it M10 or M12? standard course pitch or some special pitch?

If it's a 16mm ballscrew, it's probably M12, don't remember the pitch (I think 1mm)

BTW, move the ballnut to the other side and wrap the screw with plastic and tape. The last thing you want is any of that silicon carbide dust getting into the nut!

[misgis]

thanks for tips.

sanded it down some since not going to break any records for machine precision anyway, and now the bearings fit nicely.