From: Steve Turner
Subject: Re: Humidity sensor - peculiar behavior
References: <firstname.lastname@example.org> <email@example.com>
X-Newsreader: Forte Agent 1.92/32.572
Date: Thu, 09 Jan 2003 02:47:48 GMT
NNTP-Posting-Date: Thu, 09 Jan 2003 02:47:48 GMT
Organization: AT&T Worldnet
Spehro Pefhany wrote:
>>I should mention one other thing -- this kind of calculation is quite
>>beyond me. I know the power dissipation in the circuit -- about 30
>>mW. If I knew the volume of air that this was heating then I could
>>figure the heat rise, but I haven't the faintest idea what the
>>"effective" volume of air being heated is. It is certainly not the
>>entire volume of the chamber. This is very much a non-equilibrium
>>situation. I would guess that there is convective streaming upwards
>>from the voltage regulator, and the effect on output will be greatest
>>if the humidity sensor sits in this stream. But what its temperature
>>is -- sheesh, I don't know.
>Right, but my inclination, as an engineer, is to try to find some
>way to ball-park it, perhaps by working backwards and finding out
>what the temperature delta would have to be to make that much change,
>and seeing if that appears to be reasonable.
I understand, and I agree it would be advisable to do this.
This "small" addition to my weather station has soaked up far more
time than originally anticipated. I enjoy learning and experimenting,
and it's unfortunate that I also have to earn a living -!
> There does not seem
>to be much left that it could be- the RH in your chamber, in the
>vicinity of the sensor, simply is not what it should be, probably due
>to the heating, would be my conclusion at this point, but I'd
>look for a "sanity check" on the conclusion before accepting it.
Agreed. And I am still NOT 100% convinced that thermal effects are
wholly responsible for what I'm seeing. There is STILL some drift
even when the sensor is completely isolated thermally from other
circuitry (self heating from 200 uA of power consumption!!???). The
explanation that I posted is just the one which seems to me to fit
[most of] the data. That doesn't make it correct.
>Liptak ( _Process Control_ and _Process Measurement_ ) is one
>of the bibles. A good bunch of info on humidity can be sourced
>from ASHRAE and in the commercial vendors' info such as General
>Eastern and Vaisala.
Okay, thanks for the tip. It would seem to me, though I have yet to
do the calculations, that the only empirical data needed would be a
table [or graph] of moisture content of water-saturated air (i.e.
100% RH) as a function of temperature.
Real address contains worldnet instead of spamnet