Go Down

Topic: Anemometer (Read 2267 times) previous topic - next topic

ulrichard

May 29, 2010, 05:10 pm Last Edit: May 29, 2010, 05:48 pm by ulrichard Reason: 1
I have two Anemometers that I want to use for a small weather station:
http://flytec.ch/sensoren.htm  (has 2 wires in the cable : white, brown)
http://www.aircotec.com/cms/front_content.php?idcat=49 (has 3 wires in the cable : blue, brown, black)
Now I tried to find information on how to interface them on the internet. But I don't find any reference to something that looks similar. Some of the bigger ones say they're optical, while some seem to have reed switches, still others might be inductive.

I will contact flytec and aircotec, but it's weekend, and maybe someone on the forum can point me to the right direction before they will answer.

ulrichard

I found nothing relevant on google.

Started measuring the 2 wire one.
The voltmeter can't detect any resistance. Tested both directions. Still and rotating.

3.3V  ->  Anemometer  ->  R15K  -> GND
Measured with the ocilloscope between both ends of the anemometer.
Switched the wires of the anemometer.
No change in the voltage if it's still or rotating at different speeds.

tkbyd

Some anemometers work by closing a switch once per revolution.

Others work by using the cups to drive a little generator. The faster the wind, the more voltage produced.

And still other approaches exist... see below.

Regarding your three wire sensor: It may have an anemometer with a Hall effect sensor detecting the passage of a magnet, once per revolution. If so, that is a neat anemometer... be careful not to damage the sensor with connecting things to it. Or you may have one of the "fancy" ones.

Two esoteric approaches are used....

a) Put a voltage through a wire with a high temperature coefficient. The current will tell you how hot the wire is, the temperature of the wire will depend on...
i) How hard the wind is blowing on it.
ii) How warm the air in that wind is.

b) It is possible to have several ultrasonic trasponders "squeeking" at detectors in a small unit which measures the time taken for the sounds to travel different directions through a horizontal plane. Measurements of Doppler shifts, and some hairy geometry calculations will tell you the wind speed and direction.

Bottom line: Hope that your 3 wire sensor isn't one of the clever ones. If it is, it ought to be worth something to someone on eBay with one of the Aircotec modules!


retrolefty

#3
May 30, 2010, 06:15 pm Last Edit: May 30, 2010, 06:16 pm by retrolefty Reason: 1
Anemometer is just a generic term to mean measuring wind speed. I've seen many different methods use in projects over the years.

Use a small DC motor as a generator and measure the developed voltage to determine speed.

Use a magnetic sensor to read the spinning cups and time the rpms to determine wind speed.

Use four sensitive RTD temperature sensors along with a reference sensor (not subject to the wind) and with enough software, wind speed and direction can be calculated. No moving parts required.

Use four ultrasonic sender and receiver units and determine wind speed and directions. Again no moving parts required.

I'm sure there may be even more methods, but you get the idea.

Lefty

ulrichard

#4
May 30, 2010, 06:33 pm Last Edit: May 30, 2010, 06:54 pm by ulrichard Reason: 1
It's not so well visible on the photos, but they have a small fan inside. The word by word translation from German would be something like wing wheel, but I guess that doesn't make sense in English.
If I blow through, they keep spinning for a while. They have saphire bearings with very small resistance.
If it was a small generator, I should be able to measure something with the Voltmeter/Ocilloscope, right?

My best guess at the moment is a hall effect sensor.
The three wire device could match something like this:
http://openenergymonitor.blogspot.com/2009/09/hall-effect-sensor-circuit-diagram.html
But how would I find out which wire is which. Are there any standards?

ulrichard

#5
May 31, 2010, 10:12 am Last Edit: May 31, 2010, 10:13 am by ulrichard Reason: 1
Just got a note from Flytech that the sensor uses not a hall effect sensor, but an Oscillator with 1MHz.
No idea how that works...

markB

lefty
Quote
Use four ultrasonic sender and receiver units and determine wind speed and directions


Our commercial ones use three, and are great until you get rain showers passing thru and it throws up errors....great.

Mark

ulrichard

I received a shematic from Flytec. It involves an LMxxx integrated circuit, a transistor, some resistors and ...  To produce a rectangle signal with varying frequency for use with the arduino.
Don't know yet if I'm allowed to publish that shematic.
I ordered the parts ant then I'll see.

ulrichard

So, I got the parts and first tried them on the breadboard. Worked immediately, but when I moved it to the PCB, it didn't. So, I added a resistor more to get the oscillating voltage pass trough the reference voltage of the OpAmp. On the Arduino side, I use the digital pins 2 and 3 as inputs and have them trigger a very simple interrupt function that does nothing more than increasing a counter. Once a second I read and reset the counter. I can easily calculate the km/h from that. As I have two of the sensors, for X and Y direction, I can use Pythagoras for the speed and arctangent for the direction.
But wait, here comes another problem. Using this frequency counter I only get the wind speed, but not if the wind blows from the front or the back through the sensor. Shit! The quadrant would actually be more important that the angle. Any Ideas how I could easily get the + or - sign indicating wind from front or back on the sensor? Preferably with only small non moving parts.
For more info : http://ulrichard.is-a-geek.net

Nick Velev

Hi ulrichard,
which one of the sensors you've used in your final schematic - the Flytec one or Aircotec one?

If it it the Flytec that you're working with, do you still need the Aircotec speed probe?

Cheers!

ulrichard

Hi Nick,

I use both (for X and Y). They actually work exactly the same. The one with three wires has one unused. People from Flytech told me that the Aircotec one is just a copy of theirs.
If you need one, I can ask the local paragliding school if they still have some left. They used to have a couple of old ones. But be careful if you want to use it with a recent flight computer. It didn't work with my XC Trainer.
I wanted to use the plug to make a second serial cable for track download, but internally, they use different pins. The ones used for serial are cut off from the sensor cable.

Nick Velev

Hi ulrichard,
I was hoping that you don't need one of them :)

Do you know which are the pins on the Aircotec round connector that are used for the speed sensor signal?

I want to check how different will XCT behave when supplied with a signal for TAS.
Especially in windspeed calculation which is in general OK with GPS only, but values shown are somewhat inconsistent and to my feeling not very precise.

A new Aircotec speed probe is quite expensive, but I think I can find a used one from Brauniger/Flytec.

I've always thought that the sensors inside is identical, but unfortunately the connector type is different.
So if I make a convertor to supply it to the correct pins on the unit connector, in theory it should work.

I'll be very grateful if you can help.
Cheers!

PS: Do you have any idea why your sensor didn't work with XCT? I can't imagine they've made a change in the way it is connected to the instrument between the past and new models. This should be a standartised interface.

ulrichard

Can't remember exactly, as there's nothing left from the cables on the plug.
All I can tell is this:
The two wires were connected to the two symmetric pins that  are farther apart. The other two that are closer to each other are used for serial.
The brown and blue wires are used from the Aircotec probe, while the black wire is not used.
The brown wires are the same on both probes. Instead of the blue one the Flytech probe has a white wire.
So, to make it short: Connect the two wires of a Flytech sensor to the two pins that are farther apart on the XCT plug. I don't know the correct polarity, nor if it matters.
Should work in theory...

Some years ago I connected one of the old probes in the flying school to my XCT, and it didn't display anything. So I ordered a new one, and it didn't display anything neither. On a friend's XCT it did, however. So, I sen't my XCT for a repair. They replaced the board. From then on the newly bought sensor works, but I never tried one of the old ones again on the XCT.

I don't think it will calculate the windspeed differently with an anemometer attached. The only thing it does, I think, is displaying and logging the TAS.


Nick Velev

Hi Richard,
thank's a lot for the info. I'll give it a try as soon as I have a chance to.
I hope to find something that can substitute the aircotec type plug and fits in the XCT connector. Then I can try to make some sort of adapter for the flytec probe.

In theory knowing both the speed that you travel through the air and groundspeed can help a lot wth windspeed calculation. I just don't know whether they've included TAS as part of their calculation algorithm or not. Ideally if you know both TAS + heading/direction and GS + track/direction you know everything about the wind and you don't have to circle at all to figure out its speed and direction from the drift.

If all this is actually not used by the instrument, then I don't need a speed probe at all. It will be quite boring just looking at a figure showing 40km/h most of the time, without any other benefit.

Best regards
Nick

Go Up