[aprssig] Chemical sensors

[scott at opentrac.org]

>If such modules were available, I'd sure have a bunch measuring all sorts
of
>strange things.
>1-5v, 4-20ma, 10-50ma inputs, adjustable values for zero and span, output
>(modulate) an APRS signal and ptt and squirt it out on a transmitter.
>The packet data fields would have to be open to accept values based on
>hundreds of different types of sensors so that
>any one of them that anybody could think of could be used.....like your gas
>sensor....which I am sure are out there.

Been working on just this sort of thing with the OpenTracker.  For
1-5v/4-20ma use, it's a little limited by the fact that it's only got an
8-bit A/D converter.  However, I've been working on 1-wire sensors, and I've
already written the code to read from a 1-wire A/D.

What I'd really like is a comprehensive description from someone who
actually uses such things of what they want it to do and how.  Then there's
the problem of putting it in an APRS message.   I've been trying to address
this in OpenTRAC by providing simple 'unit' elements that'll let you report
things like temperature, angular rate, density, or whatever in SI units,
with 8, 16, 32, or 64 bit precision.  The idea is that you attach these
elements to an entity describing a particular sensor, and you can include a
text description of the sensor in another element.  It's not as efficient as
it could be, though, and I really don't have much time to work on it.

I was going to put in some more work on the OpenTracker telemetry last
weekend, but wound up spending the majority of the weekend bushwhacking on a
SAR call, as evidenced by the fact that I'm now totally covered in poison
oak rash.

Let me know how you want those calibration and setup factors to work, and
I'll see what I can do.  The SMT tracker modules are especially nice for
applications like this.  Now if only we could get one of the radio
manufacturers to repackage a small HT as a headless, MCU-controlled embedded
transceiver module...

Scott
N1VG