Wind Speed Meter

Idea: Elektor Lab

The principle behind this wind speed meter without moving parts is based on the measurement of a temperature difference. We use two transistors as NTC resistors; one is inside the housing and the other is outside. When the wind cools the latter, its internal resistance (Rbe) changes - that resistance increases. When both transistors ("NTC resistors") are included in a (Wheatstone) bridge circuit, we can derive a voltage that is a measure of the wind speed. The electronics are not terribly complicated, as you can see in Figure 1.

The two transistors used as temperature sensors are T2 and T3. To use the (differential) temperature meter as a wind speed meter, a small trick is needed: transistor T2 (the one outside the housing) must first be heated up before it can be cooled by the wind. Transistor T1 functions as a "heater." Needless to say, T1 and T2 must be mounted close to each other!

The rest of the circuit serves to control the moving coil meter M1 (a 100-µA unit).

We can be brief about the construction. It is fine on a breadboard. Make sure that T1 and T2 make good thermal contact, and that the connecting wires from these transistors to the rest of the electronics are kept as short as possible (to prevent oscillation).

The circuit is adjusted as follows. First, R5 is short-circuited temporarily with a wire bridge and point b is connected to ground. Then P1 is set so that at the output of the opamp (pin 6 of IC1) just 0 V is measured (multimeter). Then remove the wire bridge and the connection from point b to ground. The power supply is then switched off to allow T1 and T2 to cool down. Temporarily shield T1/T2 with a box or something similar so that it cannot sense air flow.

After 5 minutes (or longer), turn the power back on and immediately adjust (!) M1 to full scale with P2. The sensor will now slowly warm up, causing the meter to drop back. As soon as the return stops, adjust the readout with P3 to 0.

The last two settings influence each other slightly, so they must be repeated several times. After that, the meter is ready for use. To calibrate the meter, you must of course compare it with a "real" anemometer. Perhaps you know an amateur meteorologist in your area.