Talking of off-the-wall ideas; this one worked experimentally though it
is quite weird.

I love those rotary encoders that have a detent action, you know, you
turn them in small clicks. Usually these consist of a cogged wheel with
a small steel strip to provide the detent action, and an optical encoder
to generate quadrature phase signals.

The other day I picked up a small permanent-magnet stepper motor (about
1 inch diameter) and twiddled the shaft idly, when suddenly the thought
occurred, what sort of transducer does a stepper motor make?.

Hooking it up to the scope, the answer was, a very good one. The motor I
used had four coils, sharing a single common ground line. Picking two
coil phases, I placed a 100 ohm resistor from each coil output to
common, then connected the outputs via two 4.7K resistors and a silicon
diode (1N914), anode to the coil,  to the bases of two NPN transistors
which had 2.2K collector resistors to +5V.

This produced an excellent quadrature square wave.

You could build, using this, some interesting gadgets. Being virtually
indestructible - after all, its a rotating armature and a set of coils -
no brushes etc, it would be quite reliable in a harsh environment e.g
dust and dirt, where optical sensors tend to fail. Fed into a PIC, which
generated a multi-phase output, you could connect an identical motor at
the output and you have a rotary positioner where the output tracks the
input (antenna rotator, for instance). Because the two motors are
identical, rotating one by x degrees will cause the other one to track
exactly. Of course, on power up the position is unknown, since it is an
open-loop system, but an index sensor would allow you to calibrate at
power up, much as a floppy disk drive does a track zero seek, looking
for the track zero photosensor to go high.

> ----------
> From:         Rick Dickinson[SMTP:rtd@NOTESGUY.COM]
> Reply To:     pic microcontroller discussion list
> Sent:         Thursday, January 08, 1998 4:50 PM
> To:   PICLIST@MITVMA.MIT.EDU
> Subject:      Re: Best way to read a POT without the A/D?
>
> At 02:37 PM 1/7/98 -0600, Norm Cramer wrote:
> >Thought of this but I need to know the position of the pot.  It needs
> to
> >physically stop at the high limit and physically stop at the low
> limit.  I
> >couldn't find any encoders that would do this and give me 128 steps.
> >
> >At 11:25 AM 1/7/98 -0600, you wrote:
> >>maybe you should look at an encoder. they require two inputs but are
> >>digital.
> >>www.digikey.com lists them starting at about $2.50USD.
>
> Here's an off-the-wall idea:
>
> Use an encoder, such as the Digi-key #3315Y-1-006-ND, which gives 6
> pulses
> per revolution on each of the two outputs.  Thus, each revolution
> gives 24
> detectable states.  Use two gears, one on the shaft of your knob, and
> one
> on the shaft of the encoder, to multiply this 24 out as far as you
> need it.
>  A six-to-one ratio will give you 144 states per revolution, which
> will
> allow you to have 128 states with only eight ninths of one knob
> revolution.
>  Physical stops at each end of the range can be achieved by adding a
> pin
> sticking out of the knob's shaft or the side of the gear, and two
> appropriately placed stops for it to bang up against.
>
> What is the requirement that the end stops be physical?  Appropriate
> visual
> or audible feedback of the current setting might be simple to achieve,
> even
> with an encoder that requires multiple turns to cover the full range
> of
> settings, and might make an appropriate substitute for "physical"
> feedback
> (e.g.: end stops).
>
> Hope these ideas help you think in directions that help you solve your
> problem!
>
>  - Rick
>
> +---------------------------------+---------------------------+
> | Enterprise ArchiTechs Company   |                           |
> |     Lotus Certified Notes       |  Never underestimate the  |
> |  Appl. Design & Administration  |  bandwidth of a station-  |
> |(818)563-1061  rtd@notesguy.com  | wagon full of tape reels. |
> |   http://www.eArchiTechs.com    |                           |
> +---------------------------------+---------------------------+
>