Re: detection of rain and close window

[MFHult@xxxxxxxxxxxxxxxxxxxxx]

On Sun, 3 Jul 2005 15:31:36 -0400, "Brett Griffin"
<brett@xxxxxxxxxxxxxxxxxxxx> wrote in message
<eYmdnT7NPdGLoFXfRVn-tA@xxxxxxxxxxx>:

>I am just speculating here,

Then why bother?

>initially I thought of a rain sensor for
>irrigation, but these measure quantity of rain in inches and would not be
>effective for rain detection.

A tipping bucket meter is the only commonly used device that I know of that
"measure[s] quantity of rain in inches". A typical threshold is 0.04" or
so.

These are not to be confused with the sensors sold to homeowners for
controlling lawn sprinklers. These have higher thresholds and greater time
lags as is appropriate for the intended application.

>A humidity sensor would do the trick

Agreed -- if you mean that it would probably trick you.

>and would be more reliable, in theory.

And what theory might that be?

>Doesn't the humidity peak to 100% when it rains?

Sometimes. But often/usually not at the beginning of a rain event or during
a short one.

And the temperature-humidity conditions that create dew (a frequent
occurrence in some locations during some times of the year) would cause a
false alarm.

> So you could write a
>program that say when humidity above 95% close windows, below 90% open
>windows...

which would sometimes be wrong and sometimes be wrong.

Note that the HAI line of controllers (Omni etc) you hawk in this newsgroup
doesn't have analog inputs for analog humidity sensors nor appropriate
digital inputs for digital output humidity sensors. So you would have to
have an additional external controller or circuitry to accomplish this with
appropriate programming or analog computing. (Good reason to get a
different controller IMO).


... Marc
Marc_F_Hult