From: John Larkin
Subject: Re: Quickie Thermal Probe
Date: Thu, 12 Dec 2002 14:20:20 -0800
Organization: Posted via Supernews, http://www.supernews.com
X-Newsreader: Forte Agent 1.91/32.564
On Thu, 12 Dec 2002 13:21:02 -0800, Mike Monett wrote:
>I'd like to make a sensitive thermal probe to check for very small air
>leaks in an appartment. I envision two 2N2222 transistors in a
>differential amplifier with one exposed and the other enclosed in a
>hollow plastic tube. The voltage on the collectors will drive a vom, and
>I can adjust the base voltage on one side to zero the output. I can run
>several milliwatts in the transistors to heat them, and an air leak
>should provide cooling to the probe transistor and unbalance the
>I'm not interested in accuracy, all I want is a sensitive indication of
>airflow so I can investigate visually. The air will be approximately room
>temperature, not cold outside air.
>My question is what is the best transistor type to use? A 2N2222 has a
>fairly large metal case and should respond to the ambient temperature. On
>the other hand, a small plastic SOT-23 has less thermal mass and might be
>Any words of wisdom from someone who has done this before?
transistors should work fine, but personally I wouldn't literally use
a diffamp; I'd rather each transistor be privately biased at a
constant current, which would avoid thermal runaway issues. You could
use, essentially, two emitter followers, bases connected, and use an
opamp to nab the differential emitter voltages.
Sure, a SOT-23 has a high theta j-a (like, 350 k/w) and a shorter
thermal time constant. I'd go for 100C or so idle temperature and get
lots of signal.
I did a thing like this once. We took it to the park and hung it way
out a car window to calibrate it.