Build/Makin a Wind Sensors

[Kevin]

Hi all,

I'm having some difficulty working out the best approach to creating a
wind sensor, and hoped someone might be able to point me in the right
direction.

I want to use some sort of propeller to detect the wind speed, but I
haven't been able to find what I should connect it to.

I've seen some rotary switches in my local electronics shop, but it's
very hard to rotate. This project involves someone blowing, so the force
will be very light.

Can anyone recommend a component that detects rotation that's
exceptionally easy to move? Alternatively, I can build a switch like
this myself too, if necessary.

I'll eventually be creating 6 sensors that will be hooked up to a
microcontroller to interpret the input.

thanks,

- Kevin
kevin!multiblah!com

 

[Tim Auton]

I want to use some sort of propeller to detect the wind speed, but I
haven't been able to find what I should connect it to.

I've seen some rotary switches in my local electronics shop, but it's
very hard to rotate. This project involves someone blowing, so the force
will be very light.

An optointerrupter similar to those you get in ball mice adds zero
friction and only a little inertia. Basically you want a disc with n
holes that rotates between a light source and a light sensor. Every
time you detect the light that's 1/n rotations. The ones in mice
actually have two detectors per axis (though sometimes in a single
unit) and use quadrature encoding to detect both speed and direction.

Searching for "optointerrupter" and "quadrature encoding" should find
you a load of information.


Tim

 

[CFoley1064]

Subject: Re: Build/Makin a Wind Sensors

From: Tim Auton tim.auton@uton.[groupSexWithoutTheY]
Date: 4/8/2004 6:54 AM Central Standard Time
Message-id: <[email protected]>

I want to use some sort of propeller to detect the wind speed, but I
haven't been able to find what I should connect it to.

I've seen some rotary switches in my local electronics shop, but it's
very hard to rotate. This project involves someone blowing, so the force
will be very light.

An optointerrupter similar to those you get in ball mice adds zero
friction and only a little inertia. Basically you want a disc with n
holes that rotates between a light source and a light sensor. Every
time you detect the light that's 1/n rotations. The ones in mice
actually have two detectors per axis (though sometimes in a single
unit) and use quadrature encoding to detect both speed and direction.

Searching for "optointerrupter" and "quadrature encoding" should find
you a load of information.


Tim

If you want to make an anemometer with an optocoupler, you can just use an
opto-interrupter like the Fairchild H21A2, and make a small slotted disk out of
cardboard. You will probably want to do some signal conditioning of the output
to clean up the signal and prevent multiple clocks per transistion. You could
look at something like this with a 74HC14 to clean up the output (view in fixed
font or M$ Notepad):

VCC VCC
+ +
| |
.-. .-.
180| | | |10K
| | | | 2/6 'HC14
'-' '-'
| 1 H21A2 | |\ ___ |\
'-----. o------|H>O--|___|-o----|H>O----o
| 3| |/ 22K | |/
V ~ |/ ---
- ~ | .01uF ---
2 | |> |
.-----' 4| |
| | ===
=== === GND
GND GND

created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

http://www.fairchildsemi.com/ds/H2/H21A2.pdf

I believe most anemometers actually use very small magnets on their rotating
shaft and reed switches to do the job.

Good luck
Chris

 

[Mikal Hodvik]

Kevin,

It's possible to make a wind sensor without any moving parts, by measuring
thermal loss due to air movement around a heated transistor or similar
sensing element. Its generally necessary to use a matched transistor/sensor
that's *not* in the air flow to null out ambient temperature changes, but
the circuitry can be remarkably simple. Here's a link to some relevant
information: http://www.elecdesign.com/Articles/Index.cfm?ArticleID=3990

Mikal Hodvik
Decade Engineering
www.decadenet.com