Re: Belt vs. Direct Drive

Alex:

Certainly. This is somewhat akin to the basis for the hexapod and
other unconventional geometry machines that are out there - e.g:
stacking two rotary tables and using some special pre-processing to
get XY motion.

Things like backlash, lead errors, and out-of-squareness are not
hard to deal with in software. But the dynamic deflections that come
with low rigidity are. Now software can't turn a drill press and an
XY table into a Mitsubishi VMC, but generally you'll be ok if you
stay within reasonable limits.

For example, my mini-mill has about .007" backlash on XY (varies),
but it can make a bearing bore good enough for a press-fit by milling
with backlash comp on and not being in too much of a hurry about it.

Good luck.

Dave Kowalczyk
Everett WA
http://www.dakeng.com

>----- Original Message -----
>In this respect I wonder if it is possible to build a very crude
>machine, with a worst possible screws and backlash everywhere and
>try to compensate for it in a software to the point of an
>accurate machine?
> Alex