My suggestion is to determine switch closure by a severe change in impedance.
Purchase a small toroidal choke for each input channel. Then wind another
isolated10 turns of #26 magnet wire on each coil, and send both wire ends
out to be connected remotely by each switch. Toroids are a necessity,
because they don't interact with each other very much (flux is
self-contained). These are available all over the place both new and
surplus. Dip each toroid to keep the new winding from moving or unraveling.
I think the coil should be about 1mH for good LF operation.
On the PCB, the PIC will control the sweep of an analog multiplexer, which
will rotate the connection of a simple oscillator to each coil. If you
design it right, you will use GND as a common pin to each toroid, so each
one can be selected on the "hot" side only.
The oscillator can be a simple LC oscillator, but try to keep the freq very
low, like 100hz-1Khz, in order to reduce capacitive effects and radiation.
The idea is to measure the current required to MAINTAIN the oscillation;
when the remote switch is closed, the shorted turns will dramatically
reduce the Q of the toroid, possibly even stopping it. But in any case, the
OSC current will INCREASE when the remote switch is closed, and you can
detect that current with the PIC A/D converter .
You will need to breadboard the oscillator circuit and the number of turns
ahead of time.
The cable to the remote switch will need to be twisted pair to reduce
radiation. The length can be considerable (100 m or more) if the osc freq
is low. The switch will attach across the pair, normally open. When it is
closed, it will shortout the flux of the the toroid and change the osc
current, which can be detected.
BTW: Be careful if you PCB-mount the toroids; the copper around / under
them will affect the oscillator, too. Try to mount them upright, above the
PCB, if possible.
Sounds like fun!
--Bob
At 10:14 PM 8/7/2003 +0200, you wrote:
{Quote hidden}>I want so sense the state (open/closed) of a number of switches. The
>switches must be isolated from my circuit and from each other. Thers is
>as yet no power available with each switch. One way would be to include
>a small DC/DC converter and optocoupler for each switch, but this feels
>like overkill. Another way might be to use a small transformer (as used
>in a modem) with the switch at one side and measuring the inductance on
>the other side. Any experience with this method? Other suggestions?
>
>Wouter van Ooijen
>
>-- -------------------------------------------
>Van Ooijen Technische Informatica:
http://www.voti.nl
>consultancy, development, PICmicro products
>
>--
>
http://www.piclist.com hint: The PICList is archived three different
>ways. See
http://www.piclist.com/#archives for details.
---------------
NOTICE
1. This account can accept email & attachments up to 10M in size.
2. Federal Monitors: At request of client, some attachments are encrypted.
Please DO NOT delay traffic; please reply with credentials for password.
--------------
--
http://www.piclist.com hint: The PICList is archived three different
ways. See http://www.piclist.com/#archives for details.
