Re: [CAD_CAM_EDM_DRO] machine inaccuracies fixed by software compensation

> Doug Fortune wrote:
> >
> > Part of the motivation for this is the
> > fact that (in open loop) steppers, the
> > mechanical step acurracy varies up to
> > +/- 5% or down to +/- 2% on the better
> > motors.

Jon Elson wrote:

>I got yelled at over this calculation. Several people
>said this was 5% of ONE STEP, not per revolution!

This high quality Superior Electric Slo-Syn stepper:
http://www.slosyn.com/ssstepmotors_hi34torque.html

has a +/- 2% step accuracy. My point was actually
that "IT IS NOT PERFECT" as might be assumed by some,
and "THE SCREWS ARE NOT PERFECT" as might be assumed
by some, and further "THE ALIGNMENT MAY NOT BE PERFECT"
as might be assumed by some. Sorry for yelling.

With all this going on, (obviously some machines
are better than others), it is unlikely that a hobbiest
type machine is achieving tolerances +/- 1/2 of
his smallest (1, 1/2, or micro) step over the full
range of travel of his machine. Very very unlikely,
as with the gearing down I'm sure some machines have
a theoretical step of under ten thou.

So, I'm proposing an inexpensive way to fix that.
The US Digital encoder costs around $30. I was
thinking later, that using carbon fibre filament
thread is probably better than wire to minimize
stretching effects.

> Well, there are machines that actually do this,
> but much better, they are called ball-bar testers.

Yup. You know they cost a LOT more than $30. Many
thousands more. And there are laser interferometer
systems that are even more accurate. No surprise,
they too cost even more!

I'm proposing a $30 solution (plus a little elbow
grease) to work smarter, not spend big bucks.

I'll challange the software guys to write a version
before I get to it around Christmas......

Doug Fortune
http://www.cncKITS.com