Is it possible to build a wind direction instrument?

[Bob Poff]

You could build one with a lower resolution like this.

Have a disc. On the outer edge a ring of phototransistors or photodiodes.
Inside that ring another with LEDs.
A soda straw sitting over each device so that it can't see its companion.

Above that attached to a wind vane another disc with a small mirror.

As the vane turns the mirror reflects light down from each LEC to its companion phototransisitor.
It turns on.

And turns on a corresponding LED on a cockpit mounted display.

Wow what a bunch of wires needed. (just thought of that as I'm writing....geezz what dope)

Maybe have each of the masthead diodes biased differently so they progressively get brighter as you go around the circle, to make the transisitors conduct more/less. And summed out to give you a voltage level dependant on where the vane is and which transistor it turned on.

Resolution would be limited by how many pairs you can fit around the disk.

Just something from my deranged mind.

 

[Ross]

Bob, It could be done with a 360 degree potentiometer with no stop and a very small break at north. Then a calibrated milliampmeter display. Zero and 360 would mean the same .

 

[Bob Poff]

Bob, It could be done with a 360 degree potentiometer with no stop and a very small break at north. Then a calibrated milliampmeter display. Zero and 360 would mean the same .

I thought of that too Ross. But I wonder about the amount of torque a typically sized wind vane would provide to turn the pot. Especially after pot has been in the weather for a while and some crud deposits around shaft.

 

[Ross]

Typically that is why folks just buy them.

 

[wonkodsane]

Eventually I will build from scratch one of the three-transducer ultrasonic no-moving-parts critters that they're selling for a couple grand. The key word in the previous sentence is eventually.

If it is purely data and you make the driver software do the thinking, it should be much easier than trying to cram the logic into a circuit board and weatherizing a larger package. Mind you, I've only dabbled in the design at this point, so it's going to be a while before I'm able to do more than pontificate about it. I may find that parts for the instrument will be expensive enough to warrant just buying instead.

Any ideas from folks on whether an automotive mass airflow sensor would suit?

 

[Ray Bowles]

Thanks all for the responses to my question. There are a few ideas that have possibilities. After thinking about these I got thinking about my past drag racing experences. First starting with the basic distributor. 8 exit points out of 360 degrees. Now move to crank trigger systems. It will require wires but that's cheap and easly done. The camera is also a thought that could work. I will come back in a month with more answers.
Again, thanks for the input.
Ray

 

[Bill Roosa]

Ues the entrials of a mouse

It is a little known fact the mouse entrails can be use to divine wind direction and lots of other things too. If you take out the encoders you will find a ready made optical encoding wheel the uses an IR beam and IR detector to count the "makes and breaks" in a small plastic "window wheel" If memory serves they break a circle into 24-36 parts. Course you could use a gear arangment to get any granularity you want. Then all you have to do is build a standard counter program and make that relate to the actual position of the wind vane that drives it. One thing to consider, my Raynav has a calibration feature so you can fine tune the installation. It allows you to "offset" the indicator (change the count/point relationship)
BTW, this only works on wheel mouse entrails, the newer LED mice use a different sort of magic that is not suited for wind divination

 

[Ross]

It is a little known fact the mouse entrails can be use to divine wind direction and lots of other things too. If you take out the encoders you will find a ready made optical encoding wheel the uses an IR beam and IR detector to count the "makes and breaks" in a small plastic "window wheel" If memory serves they break a circle into 24-36 parts. Course you could use a gear arangment to get any granularity you want. Then all you have to do is build a standard counter program and make that relate to the actual position of the wind vane that drives it. One thing to consider, my Raynav has a calibration feature so you can fine tune the installation. It allows you to "offset" the indicator (change the count/point relationship)
BTW, this only works on wheel mouse entrails, the newer LED mice use a different sort of magic that is not suited for wind divination

Chicken feathers work on the shrouds so when I saw this post I thought Bill was pulling somebody's leg. My cat is quite fond of mouse.

 

[rejenk3]

If you are bound and determined ...There is the simple mechanical engineer approach that we used successfully for control rod drives. There are reed switches that close in a magnetic field. Take a disk and on the periphery attach the reed switches. Connect the wind vane to a bar magnet. Wire one side of the reed switch to a common and the other down the mast through LEDs on a board at the helm. The LED will light where the bar magnet is pointed. You probably only need to do 180 degrees since the N and S end of the magnet will close the switch, but you’ll know if the wind is fore or aft anyway. The reed switches are encapsulated in glass so weather proof. The current draw would allow AA to do the job.

 

[Ross]

If you are bound and determined ...There is the simple mechanical engineer approach that we used successfully for control rod drives. There are reed switches that close in a magnetic field. Take a disk and on the periphery attach the reed switches. Connect the wind vane to a bar magnet. Wire one side of the reed switch to a common and the other down the mast through LEDs on a board at the helm. The LED will light where the bar magnet is pointed. You probably only need to do 180 degrees since the N and S end of the magnet will close the switch, but you’ll know if the wind is fore or aft anyway. The reed switches are encapsulated in glass so weather proof. The current draw would allow AA to do the job.

How many wires from the mast head to the deck?

 

[Ed Schenck]

You will need a cable that has as many wires as you have reed switches. You should have at least eight I think. And, of course, eight matching LED lights. Then with the right magnet you will see two lights if, for example, the windex is between two switches. Now that I think back that is the way that the Heathkit worked.

 

[rejenk3]

Since the current will be negligable, 24 to 30 gauge is adequate. Check Jameco for multiconductor cables. The cable should only be about $15 and the reed switches $1.50 ea.
Yes you will need to play areound with the switch separation to get the resolution you want. If you get it right you will have twice the resolution as the number of switches, since lighting two says is between those switches.

Hey, I agree there are easier ways. But I don't think there are much cheaper for remote sensing at the masthead. Heck, I'd still go for the consumer weather station for wireless connection, wind direction and strength at $100.

 

[Ross]

Because the making of such an instrument will require a couple of printed circuit boards this link may be very useful.
http://www.riccibitti.com/pcb/pcb.htm