Re: WiFi range ...

On 8/2/26 11:59, Robert Komar wrote:
> c186282 <[email protected]> wrote:
>>
>>    Now you CAN make a wet/dry-bulb unit using two DS
>>    18b20 temperature sensors - one kinda wrapped in a
>>    light wet cloth. Suggest dipping both sensors in epoxy
>>    to enhance water resistance. DO adjust the reported
>>    temperatures, dry, to be the same. The Dallas 1-wire
>>    devices are GOOD, but no two of them will read
>>    EXACTLY the same. Calibrate at the bench, using a
>>    scientific conventional thermometer as reference.
>>
>>    Now the wet bulb, how the hell do you KEEP it wet ?
>>    This can be a PROBLEM depending on environment. Where
>>    I was, there was surely a little rain every 7-10 days.
>>    A cheepo reservoir was easy to rig. It's also possible
>>    to lace the wet cloth with a chemical, ammonium nitrate
>>    for example, which is extremely hydroscopic - but the
>>    chosen salt WILL affect the reading and compensation
>>    will be needed.
>>
>>    Anyway, wet+dry will give you a very real figure, but
>>    at added complexity.
>>
>>    Do you really NEED super-accurate humidity readings ?
>>    For like a big commercial greenhouse, YES. For yer
>>    back-yard, close-enough may be close-enough.
>>
> Back when I was an student in the 80's, one mechanical engineering
> professor described doing wet/dry bulb measurements.  The thermometers
> were tied to lanyards, and you would whirl them around in the air at
> a certain rate to get more air flow over the bulbs during the
> measurement.  It increased the difference between the wet and dry
> readings for better accuracy, and the tables were adjusted for that
> affect.  My old thermodynamics textbook has a Psychrometric Chart in
> the back that you could use for calculating the absolute humidity from
> the temperatures.  It doesn't say whether the bulbs should be whirled
> around or not, maybe that was common knowledge back then.

   "Whirl-around" IS best.

   However, for a fixed unit these days, maybe the
   best is a tiny PI fan that sucks air over the
   thing. Start the fan like 60 seconds before you
   make the reading. Doesn't need MUCH power.

   Anyway, kind-of EXACT humidity readings are a sort
   of black art. A number of cheepo approaches, some
   small and low $$$, but expect at least a 15% variance
   from reality.

   I did lots of environmental field units, tried a
   bunch of 'solutions'. None of the 'cheap' ones were
   very good - though kinda sorta "good enough". The
   one diff was for a commercial greenhouse, good yield
   required precise monitoring. Just for humidity I had
   to use a $250 unit to get 'precise'. They DID pay,
   but didn't LIKE it. Faked a wet/dry unit on some
   field units I did later using dual 1-wire DS
   temperature sensors. Cheaper but still Good Enough.

   Greenhouse, if humidity too low, you sent pulses to
   a 'mist' unit - sprayed water mist into the air.
   Wait 30 minutes, test again.