🚀 go-pugleaf

     All Electronics (818-904-0524) is selling surplus radiosondes for
$4.75 each.  They are intended to be lifted to high altitude by a balloon
and transmit back atmospheric information.  They were built by Viz Mfg.
Co.  Since I'm working on constructing various weather instruments, I
bought a couple of them and disassembled one.
     It contained two circuit boards.  One is a very simple one-transistor
radio transmitter, tuned by means of a disk of metal on a screw thread
which along with two copper areas on the PC board forms a variable
capacitor (now there's cheap :-)  The other is a data encoding and
multiplexing board (4066, etc.)
     A water-activated battery is included, sealed in a can with
activation instructions.
     There are three instruments.  An aneroid barometer bellows is
connected to a wiper arm which moves across 180 traces connected in a
complex pattern to three outputs.  I would guess it's intended to allow
the pressure to be transmitted using a small number of channels (3) by
counting pulses, while still being able to to recover from missed pulses. 
Folded up in a baggie in the radiosonde was a "Baroswitch Pressure
Calibration Chart".  It gives the pressure it what I'd guess are millibar
for each contact value from 1 to 179.  Also in the baggie was a paper
tape, perhaps giving a machine readable version of the pressure chart, or
perhaps describing one of the other instruments.  I don't have a paper
tape reader handy...
     I'd like to use the barometer, but it seems to be mainly intended as
an altimeter; the pressure on the chart goes from 0 to 1061.4, with about
10 (millibar?) between contacts at the high end.  I suppose I could put a
different sensor on the bellows and hope that a bellows intended for large
pressure changes will move a useful amount, but then I lose the convenient
machine-readable sensor it already has.  Oh well... maybe I'll send it up
in a rocket :)
     A thermistor is placed such that it will be outside of the radiosonde
when it is deployed.
     Most interesting to me, a humidity sensing element came in a sealed
can.  I opened it (yes, with a can opener) and found a rigid strip about 2
cm x 7 cm with conductive paint along the long edges and black coating
between them.  I put it in its bracket (spring wire which holds it &
provides contacts) and connected it to an ohmmeter.  It initially read
about 24K.  I breathed on it from a half meter away and it shot up to
>100K, *very* fast.  I was amazed, expecting a humidity sensor to be slow
and insensitive and possibly not have a varying resistance at all.  Since
then I've seen it vary between 18K and 26K when people aren't playing with
it.  It does seem to move with the dial humidity sensor I placed it next
to, though I haven't been keeping really close watch.
     I am curious about the sensor.  Anyone know what effect it's based
on? The fact that it was sealed in a can with a dissicant while awaiting
use implies that it has a limited life.  How long is it likely to last?
Should I expect it to exhibit hysteresis?  Would its output vary with
relative or absolute humidity, or be a function of both?
     While we're at it, the dial humidity sensor is based on a bimaterial
strip, metal and some-white-stuff which presumably expands and contracts
with humidity.  How reliable is this type?
     Has anyone made a condensing type humidity sensor?

	John
-- 
John DuBois	spcecdt@armory.com	KC6QKZ