Subject: Re: Rain Detector
From: Henry Spencer <henry@zoo.toronto.edu>
Date: Aug 12 1996
Newsgroups: sci.electronics.design

In article <4ug368$q3v@titan.saturn.net> Russ Martin <viscount@saturn.net> writes:
>...I've also tried the same PCB that you have and found that
>the first night it got corroded and wouldn't work the next day...

Uh, guys, you *do* know that you should never drive such a sensor with DC,
don't you?  That's essentially a guarantee of electrolytic corrosion,
almost regardless of what games you play with protective platings.  Try
driving it with an AC signal (preferably with severe current limiting too)
instead. 

If you're trying to sense actual flowing water, like rain, a more robust
method is a glass or plastic block with light bouncing off the *inside* of
one surface (at a 45 degree angle) en route from a light source to a
detector.  The amount of light getting from source to detector will drop
dramatically when the outside of the reflecting surface gets wet. 
-- 
 ...the truly fundamental discoveries seldom       |       Henry Spencer
occur where we have decided to look.  --B. Forman  |   henry@zoo.toronto.edu