Trying to build a anemometer to measure wind speeds. Been reading a bit about existing weather monitoring systems using reed switches. After stumbling across this the build on reading RPM from pc fan (I was going to post a link, but its my first post..xD)
With this, is it possible that the pc fan could be used to measure wind speed?
Thanks.
I believe the problem would be the high resistance of the fan due to the magnets would nullify much of the wind power and so give you a very innacuarate account of the wind speed. If you only want to know the difference between 10 and 20 mph you might be fine, but most of the fans I have dealt with would fail to register anything from 0-10mph. if you can find a fan that has almost zero resistance and then figure out how to make sure it faces into the wind, perhaps it is possible, but not ideal. A test might just be to blow softly on the fan and see if it turns.
Most PC cooling fans are brushless motors, power goes in through a controller and is switched to drive the coils. The speed is read off a third wire that acts as a tach output. The usual method for speed measurement is to derive a pulse from one of the coils being fired and then dividing by how many times it's being fired per fan revolution.
I'd use a six inch ball bearing fan for this so that the vanes would derive enough force to overcome the attraction between the magnets and the poles. You'll have to hack it to bypass the controller and pull the pulse directly from one of the drive coils...
You're way better off just buying a reed switch (or hall effect transistor) and a magnet, and then mount that to a propeller on a very smooth bearing. A muffin fan is not the way to go as others have explained above.
What you need to do is get a nice propeller that spins very freely. Glue the magnet to it and mount the reed switch on a stationary part of the propeller housing. Make sure the magnet passes as close as possible to the reed switch with every rotation of the propeller. Each time the switch closes, it's one revolution. Sense when the switch closes with an input pin, do a little math and you have the RPM reading. You can either just look for the closed state or have it interrupt and run an ISR.
I'd advise you to check out mass air flow meters fitted to modern engines.
You can get one from eBay for cheap.
Not sure how you'd calibrate it to match wind speed - but they work perfectly for an Arduino with a 0-5v output (12v supply)
Really appreciate the replies. The point about fans and their high resistance was something that completely went by me. Didn't realise how bad the magnets were until I was out of breath. xD
Placement of a magnet onto propeller fan and then reed switch near by is brilliant. How did I not think of that....
I'll be looking into those mass air flow meters, thanks for the suggestion.
Thanks everyone.
arduinow:
There's an article about this subject in the current (October 2009) issue of Circuit Cellar magazine.
About Circuit Cellar Magazine - Circuit Cellar
Airflow Analysis
by Jeff Bachiochi
Do you need a way to measure airflow in an air duct? Try deconstructing a small CPU fan, calculating the RPMs, and putting together a circuit to output data to an LCD. Airflow analysis made simple. p. 60
Don
I was playing with the MAF sensor this afternoon - after writing above. Some things going for it - it has no moving parts so doesn't rely on a "force". It can detect if I gently blow on it... right up to a vacuum cleaner.