Re: [CAD_CAM_EDM_DRO] Re: Backlash effect on Part accuracy

Michael Larson wrote:

>That's good to know. As long as we're on this topic, what can I do to
>reduce backash on the mill I just got? It already has ballscrews but my
>guess is that they're pretty worn out, 10 thousandths in Y and 3 in X.
>Can I buy oversize balls, adjust something, get new screws, or what?
>
>

Well, it depends on the design of the anti-backlash nuts. Very old ones had
two complete nuts with either an adjustable space between the nuts or a
system
of very stiff springs, often Belleville-type that spread the nuts
apart. Newer ones
generally are not adjustable, and the size of the balls allows the free
play to
be adjusted. The balls get a lot more wear than the nut or screw, so
generally
you can replace the balls when they are worn. But, .010" backlash is very
unlikely to be worn balls! I would first inspect the bearings that
constrain the
screw at one end. Something may be wrong with the setup there.
Generally, two
angular contact bearings are used back to back, with some spacers to
apply preload
to them. Some are factory set so that you just clamp them together and
the preload
is correct. It is again hard to believe that .010" of wear has
developed, so I would
check to see if the bearings were installed wrong, the clamping
mechanism is not tight,
etc. Also, check the end brackets, the yoke that holds the nuts to the
saddle and all
other mechanical parts in the chain that transmits force from screw to
table for any
sign of looseness.

Once, a long time ago, on an Acme-screw Bridgeport, I noticed something
really odd.
I had my hands on both the X and Y cranks at the same time, and I felt a
little twitch
in the X handle when I reversed direction on the Y handle. I thought
about that some,
and finally realized that the yoke had to have worked loose for me to
feel this. I ran the
table off to one end, and sure enough, the 4 bolts holding the yoke to
the saddle had
worked loose. I'm sure I would have noticed some really funny stuff
during machining
in a few days, if it continued to loosen. So, I would check the yoke
and end brackets
for looseness first, then the end bearings. A .010" axial shift in the
leadscrew should be
visible by eye. Also, the ballnuts are probably secured to the yoke by
some bolts, check
those, too.

Jon