Accelerometer as weather vane?

Hi all, I was up and my room "designing" a wireless weather station that will be put outside, take measurements, and then send it to a home station in my room with some cheap rf links when I thought, how am i going to detect wind speed?

A accelerometer mounted on a plastic 2x2" plate will do.

Oddly this idea came from me blowing a piece of paper off of my prototyping area lol. Anyways the idea works like this:

I will mount a memsic dual axis accelerometer on a piece of acrylic or something similar. Then mount the hole assembly on a small, easily bendable, spring. Then mount all of that onto the main box, which is screwed onto a wooden stake. As the wind blows, it will tilt the plate which will of course tilt the accelerometer. Ill interpret the tilt as direction. Also, now that I think about it, this can be also used to sense speed at the same time. This would take a lot of tweaking and calibration but that seems like it would be sensible. All I would have to do is sense how sharply its tilting, and in which direction. This would only work for around 0-60 mph wind. Mostly because of it being broken off.

The reason I posted this is to ask if anyone thinks this is possible, if you guys any suggestions, or if anyone else needed a way to sense wind direction without a modded pot (the cheaper way I guess, but you don't get the possibility of sensing speed as well).

The only thing I think will be a problem would be the wires I have to connect...

Try it out.
The plate and spring will almost certainly vibrate, which will make interpretation of the accelerometer readings difficult.

For measuring wind speed surely a a fan weather cocking into wind would be a better solution.

I see no advantage to using an accelerometer.

A diagram would help us to understand what you mean

Good idea, I like it.
But the spring will make it tilt all over the place. Could that be dampen in a mechanical way ?
What about the sensor in a polystyrene sphere and the sphere on top of a flexible rubber/metal/spring something ?

http://saba.kntu.ac.ir/eecd/Ecourses/inst%2086/Projects/Annemometers/Anemometer_files/Page339.htm

https://www.uni-oldenburg.de/en/physics/research/twist/research/measurement/sphereanemometer/

Or upside down. Hang the sphere on a rope. Perhaps it should get some weight, or the sphere will jump wildly in the wind.

I want to know the "windiness" outside. I was thinking about a reed pointing up and the sensor to the bottom. Perhaps a strain gauge in x and y directions to measure the force and direction against the reed.
It is a different approach, but almost the same idea as yours.

I have the feeling that everything we can think of, is already in a patent.

jremington:
Try it out.
The plate and spring will almost certainly vibrate, which will make interpretation of the accelerometer readings difficult.

I see what you mean :confused: that's why I posted it here :smiley:

I did 3d print a cup anemometer and attached that to a small pager motor. Fed that through a op amp and then to a analog pin, I unfortunately forgot the gain I set. Trial and error will get those values back.

Ill most likely get direction, but your right, with the vibration it would probably be useless for an accurate speed measurement. Luckily ill only be updating every 20-30 seconds or so. Mostly because I'm going off of a solar/battery source and need to limit current usage.

Once ill get a diagram drawn up, ill post a picture here. Ive got some "spare" class time early tomorrow.

Here is a useful discussion.
There are also lots of DIY wind sensor projects on the web, e.g. http://chrismolloy.com/page.php?u=p108

Koepel:
Good idea, I like it.
But the spring will make it tilt all over the place. Could that be dampen in a mechanical way ?
What about the sensor in a polystyrene sphere and the sphere on top of a flexible rubber/metal/spring something ?

Or upside down. Hang the sphere on a rope. Perhaps it should get some weight, or the sphere will jump wildly in the wind.

I want to know the "windiness" outside. I was thinking about a reed pointing up and the sensor to the bottom. Perhaps a strain gauge in x and y directions to measure the force and direction against the reed.
It is a different approach, but almost the same idea as yours.

I like the idea of using a foam sphere, that would definitely be at least a little more aerodynamic and I wouldn't think it would flop around so much. I mess around with hanging it upside down or keeping it up in the air. Both sound pretty promising.

The only thing I see in the way of using a sphere would be how I'm going to stuff the sensor in their :smiley:

The spring's tensile strength will vary with temperature, so that adds one more layer of complexity to calibrating this.