Nslu2 wind sensor

The story behind this experiment was to investigate the average wind speed to motivate a future wind turbine project. A commercial loging weathers station cost much. So this is an attempt to build a good alternative -cheap- solution.


The plan was to use a small linux box and some sensor device. We used the NSLU2 because of the price and the magnitude of applications available for it.



An old logitec ps-2 wheel mouse, and a ps-2 to usb adapter. Cost $3



..and inside.



Some aerodynamic suitable kitchen utils. Cost $1



The sensor wheel removed and put on a pice of surgical steel pipe.



The Nslu2 was flashed with the alternative firmware (OpenSlug 3.10) usb-hid and perl was installed.

A custom script was written to format the raw data from the mouse device. (/dev/input/event2) An good fellow named Kirtap made it work. Thanks!

Download script and all files.

To start script, type "cat /dev/input/event2 | perl /etc/vindlogg.pl &" or whathever you put them.



Sensor wheel and ir-diode moved from mouse. In this image the "north" reset sensor is clearly visible. Every time the small plastic part to the right passes between, the rotation angle is set to zero. Every time the script is started, or restarted wind direction sensor must be turned one complete revolution in order to be calibrated.



Four 100cc baking cups works as rotor for wind speed sensor.



Rotor and sensor done.



Inside the box.



To calibrate the wind speed sensor we just tejped it to the car and drove around. We used a referents wind speed sensor (black thing with small fan)



My father in law helped me allot with the lacing of the plastic parts. Here he is wondering what is going on...



DONE. Next mouthing on roof.




Wind sensor but no wind. Power connected to 1500VA 220V UPS. This was a fun experiment. Next I will build a self contained wind/solar powered weather station on top on a near by mountain. To measure more accurate data for the future wind turbine project.





Download prescript and files used.

More recent files are here, and all the logs from the last 200 days. Notice that the files for the 15th to the 19th dec are zero in size. This is because the temperature was -30C below and the rotor was frozen solid, but also notice that the NSLU2 was still working!!
All new -in use- files here

Wind direction have failed because directing wing fell off during a snow storm. I will put it back some day.... :)
/janne a--t sfe.se
Update: The logitec mouse finnaly failed after some 250 days uptime. After displaying readings at 1000% off scale for 2 days. Well good job anyway for this little device :)