Re: EMC and lead screw compensation

Smoke wrote:

> Maybe we should be installing laser positioning devices on the

machines now

> to indicate actual backlash "on the fly"? Would that work well

with ACME

> screws? That way we could eliminate the use of ball screws

altogether.

> This would be ideal for designing new machines....especially

lathes...where

> some people want both CNC and Manual capabilities..
>
> Smoke
>
>
> >-----release notes-----
> >7. Added bidirectional lead screw compensation. Files of

compensation

> >points look like this (these sample comp values are exaggerated):
> >
> >0.000000 0.000000 -0.001279
> >0.100000 0.098742 0.051632
> >0.200000 0.171529 0.194216
> >0.300000 0.292485 0.291030
> >0.400000 0.439511 0.367981
> >0.500000 0.451323 0.493962
> >0.600000 0.618011 0.618044
> >0.700000 0.669925 0.735979
> >0.800000 0.798637 0.787879
> >0.900000 0.889286 0.927685
> >1.000000 1.007024 0.959853
> >...
> >
> >where the first column is the commanded axis position, the second
> >column is the measured position (like with a laser) in the forward
> >direction, and the last column is the measured position in the

reverse

> >direction.

Wally replys:

Smokie, i could not tell if you were kidding or serious in your
reply. If you have acme screws and try to circle interpolate a
circle you get a axis jump at 4 points of the circle from the
backlash. It is very doubtfull this big jump could be compinsated
for as the stepper would have to reverse direction faster than the
table would jump. You Can't make a silk purse out of a sows ear.
Bridgeports typically have about .03 backlash.


Also to put a laser and associated measuring equipment on each axis
would probably cost much more than the ballscrews would.