----- Original Message -----
From: <batwings@i-plus.net>
To: <porkproducts@home.com>
Sent: Monday, April 24, 2000 8:31 PM
Subject: Re: Mounting ball screws

At 10:30 PM 4/19/00 -0700, you wrote:
>Hi Hoyt,

Hi Peter, nice to hear from you.

>you are discribing. The key feature is that the shim is larger than the
>inner races but smaller than the outer races of the two identical
>ballbearings, correct?

That's right. The outers are directly clamped and that shim inside inners
preloads the balls in an angular direction. You can also use spacers of
slightly different length for inner and outer, as I did on drive end to get
extra overhang moment to carry the pulley. In that mount, the pair of
bearings has a preload clamped by a nut and one bearing is also against
housing shoulder. The other bearing block on other end of screw has one
bearing against a shoulder and the lead screw has an adjustment and locknut
there for slight bearing preload too. I actually used a spacer there
between bearing and shoulder to make the lead screw pre-finished length fit
the pre-made housings, extra take-up being in the bore of housing that was
three-four times as long as the bearing width.

>Does the attached drawing show what you were discribing?

Exactly.
>
>Could the shim be a regular washer?

Too thick, have to use very small increments because the clearance in even
a low-precision bearing is small and too much load kills them in minutes.

>What type of bearings do you suggest? The more specific you can be the
>better for when I go bearing shopping.

I just pick things out of my rat-boxes, make sleeves and spacers as needed
on the lathe. Certainly any bearing with a 10 mm or bigger bore will serve.
If you decide to buy, pick them based on how you finish the ends of your
ballscrews.

>How much axial and longitudinal play is there in the final set up?

You can adjust the bearings to zero. The ballscrew final play depends on
your adjustment to a degree, but the mfgr sez you have to make final
setting by sweeping the ballscrew and picking your radial fit on the
tightest spot, that seems to leave me with about .0002" but I think that
can be made zero because these will tolerate a bit of radial preload by
being slightly flexible in that direction. As delivered and in a proper
housing but not adjusted they have about .002".

>Does your system have any advantages over thrust bearings? There is quite a
>bit of axial play in my setup. This may be the answer or mine might not be
>preloaded correctly.

Your preloaded bearing pair is standard for precision mounting of
ballscrews. You can achieve that by using angular contact, radial or thrust
bearings, if properly mounted. In use, all of these will have ball thrust
axis in approximately the same orientation.

>Why not use angular thrust bearings like the expensive pre-assembled NSK
>mounting plates use? I assume NSK uses them for good reason. I don't know
>how much single angular thrust bearings cost, do you?

This is most likely dedicated to precision and smooth operating, AC
bearings being first choice in the application when high-tech is required.
You have to have first-class components all around to take advantage of that.

>How do you generate the preload forces? Something must be squeezing the
>ballscrew threaded part towards the mounting plate, correct?

Yes but I don't know how they do it.

>Do you have any more information about CNCv239 that you can send me?

The manual is attached. I'll send a demo when I get past a couple few
revisions I haven't quite finished.
Attachment Converted: "C:\INCOMING\CNCV239.TXT"

 

Best wishes,

Hoyt McKagen

 

Belfab CNC - http://www.freeyellow.com/members/belfab/belfab.html
Best MC Repair - http://www.freeyellow.com/members/batwings/best.html
Camping/Caving - http://www.freeyellow.com/members/batwings/caving.html
Please don't panic, I'm only here to help!!