From: "Sir Charles W. Shults III"
References: <email@example.com> <6389039.2W1945d7O5@mixdown.org>
Subject: Re: I need a simple design of ultraviolet sensor
X-Newsreader: Microsoft Outlook Express 6.00.2600.0000
Date: Fri, 18 Oct 2002 07:44:10 GMT
NNTP-Posting-Date: Fri, 18 Oct 2002 03:44:10 EDT
Organization: RoadRunner - Central Florida
Oh, just about any sort of metal coating can do it- like silvered Mylar
plastic. Glass is pretty good at it too! The advantage to glass is that it is
transparent to everything else, pretty much. Look at how thin the envelope from
a mercury vapor lamp is, and it blocks virtually all the UV they produce.
Now, for what you want, a piece of microscope slide or window glass should
do it. And it is easy to check, too. Position your light sensor somewhere
secure, read the output with a meter. Now put the glass in front of it and see
how much the signal changes.
If the result is too small to measure, place the two sensors in parallel,
one forward and one backward. This makes the output voltages cancel each other.
Now put your window glass in front of one of the cells and see what the
remaining signal is.
This should be the difference voltage from one sensor seeing the UV and the
other sensor not seeing it. They need to be close together for minimal error.
In the case that you get a light sensor with a lens, chances are that the
lens itself will block UV. In that case, things are a little tougher. Some
phototransistors have the lens stuck right in the top of a metal transistor can-
usually a TO-92 style.
It is possible to remove the lenses and then put quartz glass over one (from
an old EPROM) and regular glass over the other one. Epoxy on a toothpick can be
used to stick the new covers in place. Keep in mind that a good UV sensor seems
to run about $400 for some reason.
My robotics, space and CGI web page - http://home.cfl.rr.com/aichip