Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!hao!gatech!udel!rochester!rutgers!super.upenn.edu!eecae!crlt!russ
From: russ@crlt.UUCP (Russ Cage)
Newsgroups: sci.electronics,sci.physics
Subject: Re: barometric pressure transducer ideas needed
Message-ID: <772@crlt.UUCP>
Date: Thu, 17-Sep-87 06:35:00 EDT
Article-I.D.: crlt.772
Posted: Thu Sep 17 06:35:00 1987
Date-Received: Sat, 19-Sep-87 15:12:51 EDT
References: <1790@sputnik.COM>
Distribution: na
Organization: CRLT , Ann Arbor, MI
Lines: 49
Summary: Ideas for transducers
Xref: mnetor sci.electronics:1332 sci.physics:2194

In article <1790@sputnik.COM>, chuckb@tc.fluke.COM (Chuck Bowden) writes:
>
>Can anyone suggest a way to sense barometric pressure electronically? I have
>a 0-5V a/d converter for my computer, and would like to write a barometer
>chart recorder program. The output level of the transducer is unimportant,
>since it can be amplified so that it falls appropriately into the 0-5V range.

Here's one idea, and it doesn't even need a A/D:

Take or build a mercury barometer.  Put a contact in the bottom of the
"pot" of mercury, and put a plate above the mercury such that as the
pressure changes and the mercury level in the "pot" goes up and down,
the spacing between the plate and the top of the mercury changes as
well.  Use the resulting variable capacitor as part of the tank circuit
of an oscillator, and connect a frequency counter to the output; read
the count into your computer, and convert to barometric pressure by
interpolating from a table of known values.

Temperature control of the barometer is important for this one.

Here's a second idea, which would use your A/D:

Build the same mercury barometer, but fuse a fine nichrome wire into
the top of the vertical tube and run it down to the bottom end.  Connect
a wire to the nichrome when it comes out of the glass, and another to
the mercury pool at the bottom.  The resistance between the two contacts
is the resistance of the nichrome above the level of the mercury in the
tube, plus the (very small) resistance of the mercury column below that.
Thus, the net resistance is something like C - x*P, where C is the
resistance of the entire nichrome wire, and x is the resistance of the
wire per millimeter.  Feed a current through it, get a voltage out, put
it through your A/D.  Or, use it as the resistance part of a 555 timer
circuit, clock the oscillations, and interpolate again (getting rid of
the need for the A/D).

If you don't like the idea of mercury in the house, you can probably
do the variable capacitor trick using an aneroid barometer, with
slightly more difficulty.

>Thanks!

You're welcome.  (Hey, I enjoy brainstorming, it's fun.)

>Chuck Bowden, WB7R, chuckb@tc.fluke.COM

-- 
  The above are the official opinions and figures of Robust Software, Inc.
HASA, "A" division.                      Go ahead, flame.  I bought Dow stock!
Russ Cage, Robust Software Inc.		    ihnp4!itivax![m-net!rsi,crlt!russ]