Renishaw scales giving 5.0 um to 0.2 um

[Doug Fortune <pentam-at-home.com>]

I am looking at a big old mill with average-worn
acme screws (ballscrews maybe someday).

I am assuming with high precision linear scales on the
tables, then whether its acme or ballscrews shouldn't
matter that much to the final achievable precision (if
I gear down the  motors to be compatable with the
encoder resolution).

I'm looking at http://www.renishaw.com/encoder/encoder.html
which has a gold plated "sticky" tape  available in
lengths up to 70 meters, and various digital quadrature encoders
(which all read the same tape, and therefore are
field upgradable for increasing resolution just by
swapping the reader module) variously giving approx:
 -  5 um  .000 196"
 -  1 um  .000 039"
 - .5 um  .000 019 6"
 - .2 um  .000 007 8"

(see Modern Machine Shop  May 2000 page 282).

I guess I'll have to wait until Monday to find out
I can't afford their stuff!

The Question:

Stepper Motors:
If I install steppers, then I can buy Dan Mauch's
4 axis board, which feed backs to EMC  (but how,
is it an ISA board like Kaluga's http://www.mcs.net/~kulaga/dro.html
and does EMC recognize it?).

Servo Motors:
My choice today is really only the Servo to Go board.
The above Renishaw encoder should feed back directly
to the S2G board (and hence EMC).   Is this right?

Are these presumptions correct, and can I expect
around 1 um <table> accurracy using the 1 um head
and sufficiently low geared motors to drive the acme
screws in those increments?

Doug Fortune