Netduino home hardware projects downloads community

Jump to content


The Netduino forums have been replaced by new forums at community.wildernesslabs.co. This site has been preserved for archival purposes only and the ability to make new accounts or posts has been turned off.
Photo

Measuring Angle of Attack


  • Please log in to reply
13 replies to this topic

#1 Inquisitor

Inquisitor

    Advanced Member

  • Members
  • PipPipPip
  • 91 posts
  • LocationAtlanta, Georgia, USA

Posted 18 July 2011 - 02:31 PM

Once I get my Netduino+, I’ll be going through the simple stuff… one tutorial at a time. Eventually, I will get to the projects I really want to tackle. In that vein…

Question – I want to measure the angle of an object relative to the Earth. The Netduino+ and sensors will be on the object in question. I haven’t found a solution that meets all the requirements perfectly. I’m hoping someone has a better idea.

Details
• Object will be somewhere between 1 and 20 feet above a smooth floor.
• The angle relative to the Earth will be at most +/- 10 degrees
• I need the best resolution I can get… certainly less than 1 degree. 0.1 degrees would be great!
• I will need height also, but that can be separate sensors.

What I’ve Considered – I’ve ordered them to my best candidate based on my understanding of their limitations.

1. Using two of the ultrasonic sensors (http://www.sparkfun.com/products/9491) and measuring their relative height above the floor. Con: There resolution is 1 cm, so I’d need to have the two sensors over 18 feet apart to get my desired 0.1 degree resolution. I’d like to keep their distance apart less than 5 feet. Con: Expense… I’ll need about 20 of them… $900 Ouch!

2. Using two of the infrared sensors (http://www.sparkfun.com/products/242). Pro: Resolution is good enough. Con: Distance above the floor is too limited.

3. Using a gyro (http://www.sparkfun.com/products/9059). Con: It measures angular velocity and I’m under the impression that between factors like drift and the integration, I can not rely on it giving me repeatable results.

4. Using a two axis accelerometer (http://www.sparkfun.com/products/844) and calculating angle using a little Trig knowing gravity = 1.0. Con: Unfortunately the object may be accelerating up, down, forward and backward and I can’t see a way of removing those extra acceleration vectors.
Doing my best to keep the smoke in the little black boxes.
If my message helped you... how 'bout giving me a Posted Image
www.MessingWithReality.com

#2 JonnyBoats

JonnyBoats

    Advanced Member

  • Members
  • PipPipPip
  • 155 posts
  • LocationPhillips, ME

Posted 18 July 2011 - 10:22 PM

Have you considered measuring magnetic field?

#3 Inquisitor

Inquisitor

    Advanced Member

  • Members
  • PipPipPip
  • 91 posts
  • LocationAtlanta, Georgia, USA

Posted 20 July 2011 - 12:09 PM

Have you considered measuring magnetic field?


You know... I had a mental block... I was thinking liquid filled compass, 2D, in the horizontal plane.

But, you're right... From a high school physics standpoint, I recall... they are just lines of force and in 3D.

THANKS! That's the out-of-the-box thinking I was hoping for!

I'll dig into it tonight.
Doing my best to keep the smoke in the little black boxes.
If my message helped you... how 'bout giving me a Posted Image
www.MessingWithReality.com

#4 Mario Vernari

Mario Vernari

    Advanced Member

  • Members
  • PipPipPip
  • 1768 posts
  • LocationVenezia, Italia

Posted 20 July 2011 - 12:19 PM

I am sorry, but...are you meaning the angle with the Y-axis, being perpendicular to the floor? If so, why not a simple potentiometer (maybe a precision one), then read the resistance with some (simple) circuit? Cheers
Biggest fault of Netduino? It runs by electricity.

#5 Inquisitor

Inquisitor

    Advanced Member

  • Members
  • PipPipPip
  • 91 posts
  • LocationAtlanta, Georgia, USA

Posted 20 July 2011 - 12:48 PM

I am sorry, but...are you meaning the angle with the Y-axis, being perpendicular to the floor?
If so, why not a simple potentiometer (maybe a precision one), then read the resistance with some (simple) circuit?
Cheers


I appreciate your response... and keeping me honest! :blink:

I neglected to mention... the object is flying and I'm needing an accurate determination of its pitch angle (AOA)... hopefully in the 0.1 degree range.

Thanks
Doing my best to keep the smoke in the little black boxes.
If my message helped you... how 'bout giving me a Posted Image
www.MessingWithReality.com

#6 Fabian Nunez

Fabian Nunez

    Member

  • Members
  • PipPip
  • 10 posts

Posted 24 July 2011 - 12:33 AM

I appreciate your response... and keeping me honest! :blink:

I neglected to mention... the object is flying and I'm needing an accurate determination of its pitch angle (AOA)... hopefully in the 0.1 degree range.

Thanks

I agree with Mario, just use a simple pendulum constrained to only swinging in one plane (like the pendulum on a grandfather clock). The pendulum would swing from a simple potentiometer, and like he says there are simple ways to very accurately measure resistance. If you're worried about vibrations you could make it small and encase it in a box filled with nonconducting oil to dampen the vibrations. The viscosity of the oil would dictate how much damping you get.

#7 Inquisitor

Inquisitor

    Advanced Member

  • Members
  • PipPipPip
  • 91 posts
  • LocationAtlanta, Georgia, USA

Posted 24 July 2011 - 04:38 PM

I agree with Mario, just use a simple pendulum constrained to only swinging in one plane (like the pendulum on a grandfather clock). The pendulum would swing from a simple potentiometer, and like he says there are simple ways to very accurately measure resistance. If you're worried about vibrations you could make it small and encase it in a box filled with nonconducting oil to dampen the vibrations. The viscosity of the oil would dictate how much damping you get.


Auger in!

Well... unfortunately, a pendulum although simple, won't work in my case. Think of this as being a low-flying airplane...

  • It will be accelerating faster and slower - thus the pendulum's inertia will give a false reading and drive the plane into the ground or stall it.
  • It will have some speed say (30 knots) – thus the pendulum’s air resistance will give a false reading and drive the plane into the ground.
  • Every gram counts – Weights at the end of a pendulum, oil bath, etc… and the plane will never get off the ground.

Although complex computationally, if it works as I've been reading the magnetometer sounds like a good bet so far. I’ve got to get one and see. It has the benefits of (1) no inertia, (2) no air resistance and (3) light weight. Just got to see how it works in the real world.
Doing my best to keep the smoke in the little black boxes.
If my message helped you... how 'bout giving me a Posted Image
www.MessingWithReality.com

#8 BernardG

BernardG

    Advanced Member

  • Members
  • PipPipPip
  • 34 posts

Posted 26 July 2011 - 11:35 AM

I appreciate your response... and keeping me honest! :blink:

I neglected to mention... the object is flying and I'm needing an accurate determination of its pitch angle (AOA)... hopefully in the 0.1 degree range.

Thanks


Hum, if the object is flying, the AOA has nothing to do with an angle with the earth, but an angle related to the relative wind. So, unless what you want to measure is NOT the AOA, the different sensors you have been talking about so far are not appropriate. Maybe you need to read, or re-read this : http://en.wikipedia....Angle_of_attack

Remember that in a banking turn, that is with the wings at 90° to the earth, an airplane is still flying and can reach a critical AOA causing a high speed stall....AND THAT is the kind of flight step where accurately measuring the AOA might help to save the plane....

From what you say, we are talking about 2 different things, so you are not talking about AOA.

#9 BernardG

BernardG

    Advanced Member

  • Members
  • PipPipPip
  • 34 posts

Posted 26 July 2011 - 11:39 AM

I appreciate your response... and keeping me honest! :blink:

I neglected to mention... the object is flying and I'm needing an accurate determination of its pitch angle (AOA)... hopefully in the 0.1 degree range.

Thanks


Hum, if the object is flying, the AOA has nothing to do with an angle with the earth, but an angle related to the relative wind. So, unless what you want to measure is NOT the AOA, the different sensors you have been talking about so far are not appropriate. Maybe you need to read, or re-read this : http://en.wikipedia....Angle_of_attack

Remember that in a banking turn, that is with the wings at 90° to the earth, an airplane is still flying and can reach a critical AOA causing a high speed stall....AND THAT is the kind of flight step where accurately measuring the AOA might help to save the plane....

From what you say, we are talking about 2 different things, so you are not talking about AOA.

#10 Inquisitor

Inquisitor

    Advanced Member

  • Members
  • PipPipPip
  • 91 posts
  • LocationAtlanta, Georgia, USA

Posted 26 July 2011 - 08:34 PM

Hum, if the object is flying, the AOA has nothing...


FOR MY VERY SPECIFIC SITUATION - Everything I have stated is completely accurate... including AOA and it being “possible” to calculate it using the Earth’s gravity or magnetic fields (thanks JonnyBoats) as a relative datum.
Doing my best to keep the smoke in the little black boxes.
If my message helped you... how 'bout giving me a Posted Image
www.MessingWithReality.com

#11 Inquisitor

Inquisitor

    Advanced Member

  • Members
  • PipPipPip
  • 91 posts
  • LocationAtlanta, Georgia, USA

Posted 26 July 2011 - 09:04 PM

And even though BernardG’s comments were derogatory in nature, I will try to use them constructively. I’m quite secure with my aerodynamic understanding, but my electronics knowledge is limited.

Can a potentiometer be configured such that by moving through only 10 degrees, the resulting input into the Netduino’s analog port uses its full 10bits without too much “noise” or resorting to a mechanical gear and be “fairly” linear? Would anyone be willing to suggest a circuit that might accomplish this?

Thanks!
Doing my best to keep the smoke in the little black boxes.
If my message helped you... how 'bout giving me a Posted Image
www.MessingWithReality.com

#12 Jon Henry

Jon Henry

    Advanced Member

  • Members
  • PipPipPip
  • 43 posts

Posted 09 October 2011 - 11:43 PM

And even though BernardG’s comments were derogatory in nature, I will try to use them constructively. I’m quite secure with my aerodynamic understanding, but my electronics knowledge is limited.

Can a potentiometer be configured such that by moving through only 10 degrees, the resulting input into the Netduino’s analog port uses its full 10bits without too much “noise” or resorting to a mechanical gear and be “fairly” linear? Would anyone be willing to suggest a circuit that might accomplish this?

Thanks!



Bernards comments we a bit harsh, but did bare some truth. If I understood him correctly, he was saying that if a plane were in a 90 degree bank it could still measure AOA. That is true. An AOA measurement in a 90 degree bank however isnt very helpful. A critical AOA in a 90 degree bank wouldnt cause the pane to stall, to low an airspeed would and ever dropping altitude would. If you are in a perfect 90 degree bank, its virtually impossible to produce or maintain any lift unless you used the rudder as an elevator (dont think that would produce enough lift to keep that rock in the air).

From the initial post, I assume youll only need the AOA(pitch angle) when your straight and level. If this is the case, either a magnetometer or potentiometer will work.

The magnetometer would probably give the most accurate result but does require quite a mit of math and code to get it measuring correctly. If you have the know how and code space, that is the way to go.

The potentiometer would be the simplest route. It will however only measure accurately when your wings are level assuming the pot is mounted to measure the pitch axis. If the plane is fairly slow moving, a weighted lever(weight as far to the end as possible) and averaging (in the code, read the pot 5 or 10 times then take the average) should give you pretty good results.

in response to getting the full 0v to 5v or 3.3v (depending on the microprocessor used) from only 20 (+-10) degrees of movement, I would learn about op-amps. Figure out the voltage difference the pot would yield in that 20(I would go 30 to be safe) degrees of movement and figure out which op amp would amplify that range to the 5V or 3.3V that you need.

#13 Simon B

Simon B

    New Member

  • Members
  • Pip
  • 5 posts

Posted 10 October 2011 - 02:23 AM

Hi, Responding to the original post: There are a number of (more expensive) IMU boards/sensors that combine magnetic field sensors, a three-axis accelerometer and a three-axis gyro, and sometimes a GPS to produce a tilt-compensated velocity, angle and direction reading. These are specifically designed to be used for aircraft/drone control systems. They are often called "Six Degrees of Freedom" or "Nine Degrees of Freedom" sensors, and can be used as an AHRS (Attitude Heading Reference System). They can cost up to and over $100 though, which is quite expensive for a single board...you wouldn't want it to crash too many times :-) Check out this spark-fun product for $89: http://www.sparkfun.com/products/10724 Also, this sparkfun category: lots of boards that do exactly what you want: http://www.sparkfun.com/categories/160 Also, you might want to consider combining the IMU with thermopile sensors to work out the horizon line (and therefore the AOA), Thermopile sensors are a sort of calibrated infra-red camera. The sort of thing that is used in those new, instant in-ear thermometers: it 'sees' the temperature of an object by measuring the wavelength and intensity of the infra-red light it emits. These are typically used by drones to align themselves with the horizon by detecting the temperature difference between the ground and the sky, and have been used on an open-source Arduino based autopilot system called Ardupilot: http://diydrones.com...pilot-main-page Hope this helps. Cheers, Simon.

#14 Chris Seto

Chris Seto

    Advanced Member

  • Members
  • PipPipPip
  • 405 posts

Posted 10 October 2011 - 04:27 AM

Could you explain exactly what the application here is? I am having a lot of trouble understanding what you're trying to do. I think people's comments might come off as harsh because they are trying and failing to comprehend what you are trying to do here :) Is this on an RC plane or something? It is impossible to calculate AoA based on a gravitational measurement; I think your terms may be confusing people. :) If you are trying to calculate flight angle, you may find an IMU, such as one of the SFE 9DoF to be of help. Otherwise, true AoA is a purely aerodynamic measurement.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

home    hardware    projects    downloads    community    where to buy    contact Copyright © 2016 Wilderness Labs Inc.  |  Legal   |   CC BY-SA
This webpage is licensed under a Creative Commons Attribution-ShareAlike License.