At 06:36 PM 5/9/2007, Rick Karlquist wrote:
>I would like to monitor wind speed using a simple
>cup anemometer to determine when it is time to
>crank down the tower. In looking for anemometers,
>I have only found extremely expensive solutions that
>indicate wind direction as well as speed, and require
>a complete "weather station" in the shack, and even then, no
>analog wind speed output voltage. All I want is a cup
>assembly connected to a velocity to voltage transducer
Almost any cup assembly will work and you just have to couple it to
any old DC PM motor. The 3 and 4 cup rotors are popular because the
calibration is easy and fairly independent of construction details.
However, you will inevitably have "sealing" issues. Most of the
commercial units these days use a magnet and some sort of sensor
(hall or reed switch).
One clever approach is a something like a Savonious rotor (two offset
half cylinders on a common axis) and a couple of magnets and a reed
switch. Very durable. Just highly nonlinear with wind speed, but
that's something you can calibrate with driving down the road in your car.
If you get a pulse train, you'll need some sort of thresholder that's
appropriate. A couple 555s will do, or you can use one of the
tachometer circuits in the old National Semi apnotes (back in the
days when 55 mi/hr was the limit, there was a raft of road speed
alarm circuits using a magnetic pickup on the drive shaft)
Another "no moving parts" approach is a hotwire anemometer. The
usual strategy actually uses a pair of solid state temperature
sensors, one of which set up to dissipate some heat. You can also
use diodes. one senses ambient, the other senses the change. This
can also be set up with an opamp to drive one sensor to always be
some constant temperature differential from the other (e.g. 5 degrees
hotter), and then the output signal is the heating current.
Again, there's a fair number of apnotes around on this approach. The
oldest one I have is for a NatSemi LMsomethingorother, but Analog
Devices had some too.
I believe there's one from Dallas Semi/Maxim in the One-Wire series.
>I found a source of anemometer cup assemblies at a
>place called Forcefield. I am about ready to build
>my own anemometer out of an old hard drive motor,
>but I thought I would ask what other people with tower
>anemometers are using (I know you are out there...)
>before embarking on this science project.
You can also just buy the sensor unit from someone like oregon
scientific or Davis instruments as a replacement part. The cup
assembly only (with magnet) is $15 from Davis.
>TowerTalk mailing list
TowerTalk mailing list