Re: PIC Based DRO

>> It seems like the problem will be how to couple to everyone's shafts.
>> What would you use for a shaft encoder that would be somewhat
>> universal? There may be no way to fit full size machines as well as
>> tabletop machines.

>> Also, a shaft encoder cannot manage the backlash,
>> so your resolution will be no better than the current lash in the
>> lead screw, right?

Exactly, a shaft mounted encoder cannot measure backlash - or wear - or
misadjusted gibs - or lack or parallelism in the ways - or racking of
the table - or any of a number of other sources of positional error.

An encoder placed on the leadscrew IS a DRO, but what it is Digitally
Reading Out is NOT the POSITION of the table. What IS being read out,
even with the best of leadscrews, is the number of turns of the crank
and an APPROXIMATE table position. I guess that's better than just
counting the turns manually, but it hardly seems worth the trouble.

Capturing the actual motion of the table is a whole lot more difficult,
but also, a whole lot more useful.

A DRO is a pretty simple thing to code, for the basic functions; that
is, everything except bolt holes ) Been there, done that, works fine.

( My code, with error checking, LCD display, lathe mode, and
simultaneous absolute and incremental fits in a 1k PIC. Metric
functions run JUST over 1k, although it could probably be squeezed
into 1K too with a bit of work.)

BUT, getting ACCURATE table position information at reasonably high
resolutions is still proving troublesome.

Alan

--

Alan Rothenbush | The Spartans do not ask the number of the
Academic Computing Services | enemy, only where they are.
Simon Fraser University |
Burnaby, B.C., Canada | Agix of Sparta