Go Down

Topic: Measuring wind direction and utilizing negative feedback control? need advice (Read 2134 times) previous topic - next topic

PriceyTrash

Hey tech heads,

I've only been using Arduino a few months.  I've been learning pretty quickly, but I need advice on a project! 

I need to measure the direction of wind using a small custom wind vane, and then get an "intake" pipe to face in that direction. The catch is that the intake pipe (and wind vane) is attached to a hexacopter.  As of now, I have a couple ideas in mind.  I am using an Arduino Uno.


First off, I was thinking about using the LSM303 digital compass sensor on top of a wind vane (with slip ring so cords don't tangle).  A second LSM303 sensor would be attached to a fixed location on a hexacopter.  Can I send the output from both of these remotely to an LCD screen so the person flying can see the directions and line them up?

Second idea,  I could attach LSM303 sensor to the top of the wind vane.  Then attach the "intake" pipe to a continuous rotation servo.  I know the position of these cannot be controlled, but could I code the system so the servo stops rotating when the "intake" pipe is lined up with the LSM303 sensor of the wind vane?

Third idea,  Have a rotary encoder attached to the wind vane.  The rotary encoder could control a small stepper motor that is attached to the intake pipe.

Any help is very appreciated. Thanks everyone.

Grumpy_Mike

I am dubious about the whole project, once the thing is flying the wind will be greatly affected by the movement.
Nevertheless if I had to do it I would use a magnet on the vane and a fixed magnetic rotary encoder underneath it. You get very good resoloution for not much money.

Robin2

Is the intake pipe fixed to the hexacopter - so that aligning it is achieved by rotating the whole hexacopter?

...R
Two or three hours spent thinking and reading documentation solves most programming problems.

KenF

Since both items are on the quadcopter you only need to interest yourself with their direction relative to the quadcopter.  

Surely though, the obvious solution is a mechanical linkage between the two.

PriceyTrash

-Grumpy_Mike - The wind vane will only be used when the hexacopter is hovering.  Your idea with the magnetic rotary encoder, how could I use this to then get the intake pipe to face the same direction as the wind vane?

Robin2 - The intake pipe can either be fixed to the hexacopter, or mounted such that it can rotate on a servo or stepper motor.

KenF - I only briefly thought about a mechanical connection to the two.  It would be nice, but I am not sure that I could keep the friction low enough that the wind vane could still turn at low wind speeds if it is connected mechanically to the intake pipe.

Grumpy_Mike

Quote
how could I use this to then get the intake pipe to face the same direction as the wind vane?
You said it yourself:-
Quote
The rotary encoder could control a small stepper motor that is attached to the intake pipe.
However, what is the intake pipe and why does it need to be pointing into the wind?

PriceyTrash

Grumpy_Mike - Ah, I just wasn't sure if the stepper motor with the rotary encoder would work.  The wind vane will not be facing the same direction as the intake pipe initially. Say the wind vane is pointing 60 degrees initially, while the stepper motor (and intake) is at 0 degrees.  Upon starting the program, can I get the stepper to move to that 60 degree position. And so on and so forth when the wind changes?

To answer your question, the "intake" is really a much more complicated device that is going to be used to collect aerosolized bacteria.  If the intake is pointing towards the wind, the efficiency factor of collection is much higher.

Robin2

Robin2 - The intake pipe can either be fixed to the hexacopter, or mounted such that it can rotate on a servo or stepper motor..
I don't know much about hexacopters (apart from the spelling) but it seems to me if it is easy to rotate the 'copter it would be simpler to have the intake fixed to it so that you have only two differentially moving items - the wind vane and the ('copter and intake).

If the intake can rotate separately you have to manage the position of 3 things relative to each other.

Of course it might be easier still if the windvane was attached to the intake and they could, together, rotate separately from the 'copter.

What about using a large windvane and no electronics?

I doubt if a stepper motor would be suitable in an airborne vehicle, they are heavy and very power-hungry.

...R
Two or three hours spent thinking and reading documentation solves most programming problems.

PriceyTrash

I don't know much about hexacopters (apart from the spelling) but it seems to me if it is easy to rotate the 'copter it would be simpler to have the intake fixed to it so that you have only two differentially moving items - the wind vane and the ('copter and intake).

It would be simpler this way.  It is easy to rotate the 'copter....manually.  Delving into the 6 DC motors to make it turn autonomously is not what I would like to do.  If I could get an output from a digital compass sensor onto an LCD screen near the flyer this could work.  Otherwise, the flyer cannot see where the wind vane is pointing when it is flying high.

Of course it might be easier still if the windvane was attached to the intake and they could, together, rotate separately from the 'copter.

What about using a large windvane and no electronics?

I doubt if a stepper motor would be suitable in an airborne vehicle, they are heavy and very power-hungry.

It could be. But the larger wind vane is more weight, like you pointed out with the stepper motor.  It would need testing, but I have a feeling it wont be sensitive to wind this way.  A light breeze may not change its direction.  A very small stepper motor wouldn't be bad, about .125 lbs or so....still not great.


I think what I can do is use a rotary encoder on the wind vane. It will sense the wind vane's direction.  Then I can use this output to control a continuous rotation servo that is attached to the intake, I think.

Robin2

But the larger wind vane is more weight,
i just meant a larger piece of polystyrene foam for the fin - negligible weight.

I believe the fin (vertical stabilizer) on a Boeing 777 is so big that it is not necessary to use the rudder to control yaw if one of the two engines fails on takeoff. And the engines are almost as large as a 737 fuselage.

...R
Two or three hours spent thinking and reading documentation solves most programming problems.

KenF

Could you perhaps tell us what your purpose is, within your application.  It's quite probable that there are better solutions that don't involve any fins or air intakes.

PriceyTrash

Could you perhaps tell us what your purpose is, within your application.  It's quite probable that there are better solutions that don't involve any fins or air intakes.
Their must be an intake, and there must be a way to sense wind direction. At think at this point I will attach the wind vane to the intake. Between them I will have a rotary encoder. The rotary encoder lets me know the relative position of the intake to the wind vane. The rotary encoder will control a continuous rotation servo at the base of the intake to turn it the direction of the wind vane.

Riva

Will the prop wash from the copter blades make rotating a collector to face the wind a moot point unless its opening is far enough above/below/outside prop wash effect?
Don't PM me for help as I will ignore it.

uxomm

Hi!
With my experience as a paraglider pilot i can tell (beside technical aspects): it is not easy to detect wind direction if you (and the meter) are on the flying oject. There is no reference because you cannot find out in an easy way whether the surrounding air is moving or not. It can be done in an indirect way for example with the help of gps data in combination with "defined" flying movements (like flying a circle or other) and measure the drift.

Udo

Edit:
Sorry, did not read the whole thread in detail: the wind direction will only be measured while the object is on the ground... OK, forget my arguments.
But another problem might be "ground effect": wind direction and speed are influenced much by vegetation, hills etc. on the ground. so on airfields the wind meter is placed with some distance (several meters) above the ground.

Always decouple electronic circuitry.

PriceyTrash

Will the prop wash from the copter blades make rotating a collector to face the wind a moot point unless its opening is far enough above/below/outside prop wash effect?
Yes, Riva, you're correct.  Members of my team have done Computation Fluid Dynamics on the 'copter we are using to make sure the collector is outside the prop wash.

Edit:
Sorry, did not read the whole thread in detail: the wind direction will only be measured while the object is on the ground... OK, forget my arguments.
But another problem might be "ground effect": wind direction and speed are influenced much by vegetation, hills etc. on the ground. so on airfields the wind meter is placed with some distance (several meters) above the ground.


You were right the first time Udo. The 'copter will be in the air, but it will be hovering! 

Go Up