Newsgroups: comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!yale!zip.eecs.umich.edu!yeshua.marcam.com!charnel.ecst.csuchico.edu!olivea!sgigate.sgi.com!enews.sgi.com!wdl1!mail!cps201!dombrows
From: dombrows@lds.loral.com (Brian Dombrowski, 5424)
Subject: Re: ELECTRONIC WINDSENSOR
Message-ID: <1994Sep19.134626.19986@lds.loral.com>
Sender: news@lds.loral.com
Reply-To: dombrows@lds.loral.com
Organization: LORAL Data Systems
References: <35cnoc$pt2@handler.Eng.Sun.COM>
Date: Mon, 19 Sep 1994 13:46:26 GMT
Lines: 72

In article pt2@handler.Eng.Sun.COM, cmcmanis@Sun.COM (Chuck McManis) writes:
> 
> Sounds like a propeller on a motor and an A/D converter is all you need. 
> 
> --

Well, I guess that might work but it won't be very sensitive.  Here is
the way I've seen it done most of the time:

You get an anemometer assembly which is three or four cups mounted
around a rotating shaft.

The shaft goes into a transducer body and is attached to a disk
with four holes punched in it at 90 degree offsets.  On the top
side of the disk is a IR LED.  On the bottom is a IR phototransistor.
As the disk rotates, the holes in the disk pass between the LED
and phototransistor allowing the LED's light to reach the phototransitor
and drive it into saturation.  You can tie the collector of the phototransitor
to Vcc with a 10K resistor and then run a lead from the collector to the
input of 7414 Schmitt trigger inverter.  At the ouput of the inverter
you'll get a nice pulse train of square waves when the anemometer turns.
The frequency of the pulse train is directly proportional to wind speed.


                     Vcc
		     |
                     /
		10K  \
		     /
                     |               7414
                     +---------------|>--------> to pulse accumulator on uP
                   C |
                    /
phototransistor B -|
                    \
                   E |
                     |
                     V
                     GND

If you decide to use a 68HC11 for the uP then your lucky because it
has a pulse accumulator input (PAI) pin that causes a hardware
register to increment at every pulse edge.  By counting the
number of pulses per a unit of time you can compute wind speed.
I've done this.  It's simple, it's reliable, and it works.

Heath Kit in Benton Harbor MI used to sell weather station
kits.  I bought one about 12 years ago.  The kits came
with a manual that had an order form for replacement parts.
They would also sell the manual by itself.  You could order
the manual then use it to order just the parts you needed
for the anemometer assembly.

Alternatively you could make your own.  The toughest part
is making the cup assembly but that could be done by
cutting ping pong balls into half spheres and mounting
them on ABS plastic tubes ( you know the kind used for
building models.)  The rest is a piece of cake.

You'll still need to calibrate the thing yourself.
Some weenies out there will try to tell you that the
anemomter will turn faster than the wind speed it is
measuring which is BS.  The Heath Kit anemomter I have 
( which has teflon bearings and gives for pulses per
revolution ) has the following slope:

	wind speed in MPH = pulse frequency * 1.17

Hope this helps,

Brian

