Re: gEDA-user: Universal Turin^H^H^H^H^Hmilling machine

[Charles Lepple <clepple@gmail.com>]

On Mon, 8 Nov 2004 02:55:28 +0000, Karel Kulhavy <clock@twibright.com> wrote:
> What abut exposing the codes on a sheet of document film, epoxying to a glass
> plate and using linear CCD for readout?

It's a gray area (no pun intended)-- the results will depend on
construction details, and the output characteristics of your CCD.

At the very least, you might want to consider a magnifying lens in
front of the CCD or photodiode. There are also some signal processing
tricks you can do if the CCD produces a quality analog signal, and you
use diffraction techniques to vary the brightness (instead of having
sharp on-off transitions). It's like the difference between
sigma-delta and Flash ADCs-- for a given sampling speed, you're
getting closer to the analog speed of the shaft via better sampling
resolution.

Check out some of the home pages of high-precision encoder
manufacturers-- they may have some useful hints. Renishaw and
Heidenhain come to mind, but there are others as well. One thing you
might want to search for is sinusoidal outputs.

-- 
- Charles Lepple