It’s fairly straightforward to use the Modern Device Wind Sensors with either a Mini Pearl Logger or a Nano Data Logger. I had great success with the Rev. P sensor.
- The wind sensors have two data pins which must be connected to analog pins on the Arduino--very easy to hook up.
- A sketch can read the analog output and use a calibration equation to convert the voltage to miles per hour.
For the Rev. P sensor these two lines are all you need:
int windADunits = analogRead(OutPin);
float windMPH = pow((((float)windADunits - 264.0) / 85.6814), 3.36814);
Powering the Rev. P wind sensor is a little unusual. It requires 8 to 12 volts (not sure of the range). So the typical batteries I use for data loggers (3 or 4 AA or AAA) will not work. I powered the Arduino with a 9 volt battery and powered the wind sensor directly from the battery (or from the power input pin on the Arduino). Both the Arduino Nano and Pro Mini can be powered with 9 volts (they have voltage regulators).
- Both the Nano Data Logger and the Pro Mini Logger can read the wind sensor, real-time clock, and other sensors (temp, humidity, barometric pressure) and save the data to microSD card. I saved the data every five seconds.
- The Nano Data Logger does not require soldering and should work great for a three-hour kite flight powered by a 9 volt battery. I used an Arduino Nano, but I soldered together an SD module and RTC module (and BME280 sensor). Starting with a Nano Data Logger will reduce the soldering to the two minutes needed to add a header to the Rev. P.
- The Nano Data Logger has a USB port, so no UART adapter cable is needed.
A Mini Pearl Snap Logger will require a little extra soldering to get power from a 9 volt battery and also power the Rev. P. It will also be a bit more work to build compared to plugging the Nano into the Nano data logging shield.
I found the wind speed data from the Rev. P to be a little noisy, so my next attempts will probably incorporate some averaging. I will read the wind sensor output maybe 20 times in 5 seconds and take the average.
- I found that the Rev. P wind sensor reported lower wind speeds that the Kestrel anemometer. When I moved the Rev. P wind sensor so the KAP rig offered less obstruction to the wind, the wind speed data was less different from the Kestrel data. So next time I will move the wind sensor so it is even less obstructed by the KAP rig. This means it will be sticking out and be vulnerable to being broken. This is the trickiest part of this mission: mount the Rev. P wind sensor on a stick so it is out in the clean wind but develop a protocol to prevent it from getting broken.
- You should ask the AS220 folks how the Rev. P sensor should be oriented toward the wind.
The first two times I tried the Rev. P wind sensor, I also tried the Rev. C sensor. The Rev. C sensor did not work as well as the Rev. P, so I focused on the Rev. P afterwards. Although I never got the Rev. C sensor to work well, that was probably user error. There are some good reasons to consider the Rev. C sensor, especially for student projects.
- The Rev. C wind sensor is cheaper ($17 vs $24).
- The Rev. C sensor is less vulnerable to breakage.
- The Rev. C sensor can be powered with three or four AA or AAA batteries instead of requiring a 9v battery (in fact, the specs say the Rev. P needs 10-12 volts, which is downright inconvenient).
Another consideration for student projects is to include other sensors in the payload. A record of wind speed for a kite flight becomes more interesting and educational if there are simultaneous records of other variables. A BME280 sensor provides very good information about air temperature, humidity, and barometric pressure. Barometric pressure is an excellent proxy for altitude, and temperature and humidity vary with altitude in predictable ways because science.