One of the best ways to measure wind speed is to use an anemometer. You could go out and buy one, but I find it much more fun to build my own.
There are several types of anemometer to choose from, but I am going to build one that uses electromagnetic induction. Normally, we think of electric potential as something you can get from a battery. But you also can get it from induction. It turns out that changing magnetic field also creates an electric potential. If I make the wind change a magnetic field, the electric potential this produces will be proportional to the wind speed.__ __So all we need is a changing magnetic field with a coil of wire to see this work. Try this simple demo to see it in action.
Start with a magnet---the stronger, the better. If you place the magnet inside a coil of wire, nothing happens. But if you move that magnet within the coil, you get a current. You can see the galvanometer (just a sensitive ammeter) needle move as I do this in the GIF below. The greater the current, the greater the voltage.
You don't see it in my demonstration, but the angle between the magnetic field and the coil of wire is important. This means that instead of moving the magnet, you can rotate the coil. And there lies the key to an electric anemometer---a coil of wire rotates within constant magnetic field. The faster it rotates, the greater the voltage---and, hence, the greater the wind speed.
All I need to do now is find a coil of wire that spins within a magnetic field. Oh, wait! I have one. It's called an electric motor. An electric motor is simply a rotating coil of wire and magnets. When you run current through the coil, it creates a torque that causes the whole thing to spin. I once wrote a tutorial explaining how to build a simple motor that looks like this:
Instead of running current through the wire, I can spin the coil manually and measure the voltage. And yes, an electric generator and an electric motor are essentially the same. With a motor, you supply a current and it spins. With a generator, you spin it and it creates an electric current.
So, let's start with an electric motor. You can find them in all sorts of places. The one I'm using came from tiny RC helicopter that had crashed one too many times. Be careful selecting your motor. An AC motor like you'd find in a fan won't work. Neither will a brushless motor. I made this mistake when I tried using a PC cooling fan. These motors work differently than a conventional coil and magnet motor.
After finding a motor, you need to make a fan blade. I cut one from the bottom of an aluminum soda can. It looks like this.
I mounted the motor to a Lego bracket so I could hold it out in the open to obtain better wind measurements.
It doesn't have to be pretty, it just has to work. Put it all together and it looks something like this:
You can test this easily. Just connect the two wires from the motor to a voltmeter. Spin the fan and you should get a voltage reading. The faster it spins, the greater the voltage. That's not very useful though. To make a functional anemometer, you'll need to understand the relationship between voltage and wind speed. To do this, I put my anemometer in the car and used it to measure airspeed---which should match the speed of the car, assuming I'm not driving in a gale. I followed this plan:
- Get in the car.
- Stick the anemometer out the window and hook it up to a multimeter.
- Drive at different speeds while one of my kids records the speed (in mph) and voltage. Yes, the values fluctuate. Just get the best value that you can (my 10 year old son didn't like that).
I can't say for sure how accurate my speed data is, so this might be a rough correlation. It's surprisingly difficult to drive at a constant speed of 5 mph with an analog speedometer. I guess I could have used a GPS, but I was lazy. In any event, I plotted car speed vs. motor voltage:
I chose to fit a line that goes through the origin such that the intercept is zero. This means that at zero mph, the voltage is zero. Perhaps I should have allowed the intercept to be non-zero, but I will use this for now. From the slope of this line, I get a relationship between voltage and speed. Straying from convention, I'll use w for windspeed and V for the voltage (using v for speed would exceed my quota for the letter v):
But I really want to solve for the windspeed as a function of voltage since I usually don't know the wind speed:
This means that I just take my multimeter reading in volts and multiply by 104.3 and I get my windspeed in miles per hour. But what if I want the speed in meters per second? Doing a unit conversion on that proportionality constant reveals:
Now I'm ready to rock on with some wind data. In fact, I could connect my DYI anemometer to a digital voltage recorder and find wind speed as a function of time. If I want to really use this tool, I probably ought to mount it on a pole with a wind vane so the fan always points into the wind. But I already have plans for this device.
If you decide to build your own anemometer, you must create your own calibration line to determine the connection between voltage and speed. I probably could get better data for my device, but this is mostly a proof of concept to show that it works.
