The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: firstname.lastname@example.org (Bill Sloman)
Subject: Re: linear optical encoder - reflective
Date: 19 Sep 2002 16:17:50 -0700
NNTP-Posting-Date: 19 Sep 2002 23:17:50 GMT
email@example.com wrote in message news:...
> >> > hi all,
> >> >
> >> > i'm trying to build an optical encoder. i've done it many times before,
> >> > this one is giving me trouble....
> >> >
> >> > it needs to be reflective, and linear.
> >> >
> >> > that is, it's a straight steel rod with a series of black stripes on it
> >> > (about 0.5mm spacing) that i want to use to detect linear displacement.
> >> >
> >> > i've been playing around with lenses and photodetectors and the guts of
> >> > old optical mice, but it's just so fiddly and I cant make it work.
> >> >
> >> > I have some small cylindrical pieces of some clear acrylic stuff (about
> >> > 3mm diameter) to use as cylindrical lenses, and some HLC2701 optical
> >> > detector modules, but I cant make it work.
> >> >
> >> > does anyone have any tips on this sort of thing? or some off the shelf
> >> > parts that can do it directly?
> In aus.electronics news-server wrote:
> > I've seen a fellow do this with a laser printer and transparency - to
> > produce a fine linear graticule.
> > This was then 'scanned' with an IR pair to act s a linear transducer.
> That's kind of what I'm trying. I've laser printed a transparency with the
> 0.5mm stripes, and glued it to a shiny steel rod with a flat along one side.
> I made one work reliably with a spherical lens, and that's fine. but i havent
> got any more lenses like that, and i need to make three more. I can always
> get it to sort of work, and sometimes work well, but it's just too unstable.
> thanks everyone for your pointers, though,
The usual way to go - build into the Heidenhain sensors and very
nearly everything else I've seen - is the Moire approach.
You have a single moving grating, and close up against it you have two
stationary gratings, printed on the same transparent substrate (or cut
through the same opaque substrate). the Two stationary grating have
the same period as the moving grating, but differ in phase by 90
You have to mount the stationary grating so that its stripes are
parallel to the stripes on the moving moving grating to a better
accuracy than the length to width ratio of the stripes, but once you
have done that you don't need focussed optics. Shine a single LED onto
the stripes, and measure the light reflected through the in-phase and
quadrature stationary stripes with two separate photodiodes.
Dead crude, very effective and any patents must have expired by now.
We slapped one together at Fisons Applied Sensor Technology back in
1992 and it worked like a charm - somewhat to my surprise. A few years
earlier I'd been exposed to an idiotic system where the in-phase and
quadrature detectors had to be set up independently, and that had
proved to be excessively vulnerable to being knocked out of alignment
- the system was exceedingly cheap (until you figured in the costs of
alignment and frequent realignment) and the management remained very
attached to it right up to the point where the company went bankrupt.
We took it over and the guys working on the system managed to lose the
interfacing data in the transfer (probably by design), giving us a
great excuse to dump the system and replace with the more or less
indestructible Sony system based on reading magnetic markings in a
Hope this helps.
Bill Sloman, Nijmegen
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup