So now i’ve been bitten by the heli bug it seems that every time i look outside and there is no wind, by the time i get outside with the heli there is crappy gusty wind.
So, being a bit of a DIY person with some skills in fabricating, electronics and programming i decided to have a shot at making an Anemometer!
The process began by looking at some other DIY designs for the actual wind vane but that’s pretty much where the similarities ended.
Nothing was purchased for this project other than the PIC chip (AUD$10). – it was all stuff i had laying around home.
Mechanical:
I started with a PVC ‘T’ joint and worked from there. I turned down a chunk of aluminium to plug one end and bored a hole through it to support a bearing. The bearing i sourced from an old hard drive. It was the bearing that the head mechanism pivoted on.
I then turned down the end of some delrin rod to use as a shaft for the unit. This slides through the bearing and is drilled and tapped on the end to support the vane assembly.
The vane assembly is made up of a disk of aluminium, carbon fibre rods and 1/2 ping-pong balls.
Mounted on the shaft is a delrin disk with the magnet from a PC case fan glued around it (more on this later)
Electrical:
The actual sensing of the revolutions is performed by a hall sensor salvaged from a PC case fan. Being brushless motors, they rely on a hall sensor to take care of energizing the coils to spin the motor. I pulled the hall sensor and the circular magnet from the fan. The magnet was glued to a disk mounted on the main shaft. The hall sensor was then mounted up near the magnet.
With the sensor taken care of, I needed to be able to time the duration of the pulses to give me some feedback. Ths is where the pic chip comes in. Now, i am fairly new to PIC’s but with datasheet in hand and using the Proton Development Suite Lite edition i was underway. The code was the hardest part. This is the first time i have used an LCD on a PIC and it took a while to get the code right but i got there in the end.
Basically, the PIC waits for a pin to go high and then low again and times the duration. It then calculates the speed based on the diameter of the vane. These vanes generally spin at about 1/3 of the actual wind speed, that is also accounted for. The hall sensor inverts signal every 90 deg so the duration of the high signal is only 1/4 of a rotation. This means we can get 2 readings per rev. If i really wanted to, i could make this 4 but it gives some time for writing to the LCD and calculations while the signal is low.
The hall sensor runs from 3-30v or something like that - a really wide range. It’s a 4 pin device, Vcc, Vdd and 2 ’signal’ pins. The signal pins simply toggle to ground depending on the magnet orientation. I am only using one of the signal pins. It is connected to a pin on the PIC which is pulled high via pullup resistor and then the hall sensor pulls it low. Other than this, the LCD is basically directly connected to 6 pins of the pic as well as power, ground and LCD contrast. I am currently powering the setup from USB so it has a stable 5v supply without a voltage regulator. The final revision will likely have a VREG.
The prgram in the pic simply initialises the LCD and waits for the pin to go high, it then starts a timer and waits for the pin to go low. It then stops the timer and does the calculations for speed and displays them on the LCD. It contains a ‘timeout’ of 10s. i.e if the signal does not invert for more than 10 seconds, it assumes wind speed is 0 and resets the timer.
the electrics are just on a proto board at the moment and there is still more work to be done, this is just the start. I hope to have a programmable ‘diameter’ for the vane and also RS232 output of the speed so it can be logged to a PC.
Pics:
gud day, it is a very interesting yet simple anemometer output. could you please email me detailed construction and parts list of the electrical/schematic of this unit. i find it very useful to monitor wind speed in my locality (Northern Philippine) as we are bombarded now with strong typhoons. PLEASE.
Unfortunately I don’t have any other info other than the source code for the chip (i had a hard drive die on me). It is very simple though. All you need is a voltage regulator to supply 5v to the chip, hall sensor and LCD, then just match wires to pins – easily done from source code. As you can see, i made the initial one on proto board which proves how simple it is.
thanks andrew!