Re: [CAD_CAM_EDM_DRO] Re: Ballscrew Slop Question

Hugh Currin wrote:

> John:
>
> At 04:29 PM 2/13/01 +0000, you wrote:
> >Hello All, I just installed ballscrews on my Bridgeport mill, in
> >preparation for a CNC conversion. They were purchased from Hi-Win
> >Industries. After installation, I noticed that there is about .002-
> >.003 slop in either direction on the X axis. (the Y seems OK) I have
> >been told that my bearings that are in the left side end cap might be
> >at fault.
>
> I recently installed Rockford ball screws on my Bridgeport size mill. I
> have some .002 backlash in the Y. I had about this in the X but worked
> with the alignment which eliminated it. I convinced myself that if the
> nuts and bearings are in very good alignment the backlash was much better
> or was eliminated. I still think this but don't really know why it would be.

If the nut and end bearings are not aligned, it will force the screw to bow.
The bow will change with load, and cause all sorts of strange results.
Two other possible causes of apparent backlash are a loose yoke (the
block that mounts the two X and Y nuts in the center of the saddle) and
loose ways, allowing the table to rock as well as slide along the axis.
A loose yoke can be detected by putting your hands on both the
X and Y handles at the same time, and making small motions of
one while feeling for a response in the other. If one handle definitely
delivers a 'bump' to your hand while the other handle is rocked from
one end of the backlash to the other, it indicates something loose in
the yoke. That area is usually so hideously dirty, it is hard to not get
swarf under the yoke when reassembling. A loose attachment from the
nut to the yoke will not show up this way, but it will still cause backlash.

Rocking of the table is much more subtle. One test is to use a dial indicator
like when checking backlash, but apply force to one end of the table,
at right angles to the normal direction of travel. The loosest spot is usually
near the center of travel. Press frontwards and backwards on each end
of the table and look for dial indicator deflections. Also, press up and
down on the table ends. A worn table way can allow the table to rock
in either of these directions, and it can appear to be a linear backlash,
but in fact is a rocking of the table in one of these 'cross' axis directions.
The dial indicator reads it because the indicator is not in the same plane
as the screw.

> Please let me know if alignment gives you any gains in backlash.

A misaligned screw may bind up (my experience), will not be accurate, and
may have a variety of odd behaviors, as well as causing excessive wear.

Jon