| Author |
Message |
|
|
 Posted: May 01, 2012 - 10:57 AM |
|
Joined: May 01, 2012
Posts: 5
|
|
Hey I am a mechanical engineering student and I need some help.
I'm trying to develop a angle of attack sensor for a glider. I need help with a device to measure the angle. I was thinking of using a hall effect sensor. Do you think it will work or does anyone have any other ideas that might help me? |
|
|
| |
|
|
|
|
|
|
Posted: May 01, 2012 - 12:41 PM |
|
Joined: Sep 04, 2002
Posts: 21271
Location: Orlando Florida
|
|
| How about using a 3 axis accelerometer as a tilt sensor? pitch is the atan2(y,z) and roll is atan2(x,z) where the x.y.and z is the volts out converted to Gs. |
_________________
Imagecraft compiler user
|
| |
|
|
|
|
|
|
Posted: May 01, 2012 - 03:23 PM |
|
Joined: Feb 19, 2001
Posts: 25921
Location: Wisconsin USA
|
|
|
|
|
|
|
|
Posted: May 01, 2012 - 03:40 PM |
|
Joined: Dec 11, 2007
Posts: 6849
Location: Cleveland, OH
|
|
Welcome to the Forum.
Stable flight or aerobatics? i.e. What angle does it need to cover, normal flight and approach, or "unusual attitudes"?
Magnetic field orientation isn't parallel to the surface of the Earth, either. At the poles being parallel to the mag lines would put you in a nose dive!
Uncle Bobs the man, see his post above. One might add a gyroscope and use the accelerometers to auto re-set the gyro during stable flight.
This sounds like one of those projects that could range from a simple testbed and a few lines of code to a PhD project implementation.
JC
Edit: I missed Theusch's post. The rotary encoder would be fine for a system that isn't accelerating, but that brings us back to the question I asked above. Stable flight, normal attitude, or not? If one is using the AOA for a stabilized final approach I would think that would work reasonably well. If one looked at the instrument on the Base to Final turn the info might not be what one one expecting it to report.
JC |
|
|
| |
|
|
|
|
|
|
Posted: May 01, 2012 - 05:00 PM |
|
Joined: Feb 19, 2001
Posts: 25921
Location: Wisconsin USA
|
|
|
Quote:
The rotary encoder would be fine for a system that isn't accelerating,
??? OP asked for
Quote:
angle of attack sensor for a glider. I need help with a device to measure the angle.
I don't think an angle changes regardless of speed or acceleration. Now, your questions are certainly valid if developing a control system for autonomous flight. But that isn't what was asked. (Hmmm--do we know whether this glider is full-sized/manned or a model?)
[Full disclosure: I'm not an airplane person, real or model.]
Aren't the simple old-time attitude indicators/artificial horizons nothing more than a weighted sphere in liquid? |
|
|
| |
|
|
|
|
|
|
Posted: May 01, 2012 - 07:51 PM |
|
Joined: May 01, 2012
Posts: 5
|
|
It will be a full scale glider in normal flight conditions. The sensor will however have to be adaptible so that it can be used on different sailplanes. Thanx for the help any other ideas that come to mind? I'm thinking of designing a pivoted vane angle of attack sensor. Therefore I need to get an idea of which sensor to use before continueing.
I like the idea of the accelerometer. Will it be able to "remember" the zero angle of attack angle? |
|
|
| |
|
|
|
|
|
|
Posted: May 01, 2012 - 08:38 PM |
|
Joined: Jan 03, 2006
Posts: 4417
Location: Hemel Hemsptead, UK
|
|
A three-axis accelerometer can be used to derive local vertical if it's stationary or moving at constant velocity, but if you're climbing, diving or turning there is temporary acceleration which causes local vertical not to be aligned to the middle of the earth.
But for constant direction flight, you should be able to get what you need.
Another approach - also using accelerometers - is to use one of the model aircraft autopilots and monitor its correcting output. It's the same approach, but all the hard work is done! |
_________________
Neil Barnes
www.nailed-barnacle.co.uk
|
| |
|
|
|
|
|
|
Posted: May 01, 2012 - 08:59 PM |
|
Joined: Dec 11, 2007
Posts: 6849
Location: Cleveland, OH
|
|
|
Quote:
I don't think an angle changes regardless of speed or acceleration.
True, the AOA may not change much as a function of velocity or accelaration, but the readings from a MEMs accelerometer are certainly a function of both angle and acceleration.
The AOA is, IIRC, generally considered to the be angle between the wing's chord line and the surrounding air. Note that the actual reference is the surrounding air, NOT the plane tangential to the surface of the Earth.
In a significant cross wind, or while pulling a few G's in a tight turn and sliding in un-coordinated flight, the relative wind is not stationary air over the airport.
This actually matters if the goal is a "True" AOA for all attitudes of flight, (to be used as part of an imminent stall indicator), or if it is just a more simple AOA for final approach to landing indicator.
Dalpilot can probably correct the above for true aeronautical engineerng accuracy. The point, however, is that a simple magnetometer or Mems acceleromter probably won't do quite what the OP is looking for.
JC |
|
|
| |
|
|
|
|
|
|
Posted: May 01, 2012 - 09:16 PM |
|
Joined: Nov 22, 2002
Posts: 12051
Location: Tangent, OR, USA
|
|
The angle of attack can be quite different from the aircraft's angle with respect to the horizon, especially near stall. The gliders that I flew in a former lifetime had a small metal tab extending forward of the leading edge of the wing. It was positioned in such a way that when airflow changed (as it neared stall), it would flip up and close a little switch that squawked in the cockpit.
I really think that attempts to measure this based only on the attitude of the fuselage will result in dangerously misleading "information".
Jim |
_________________
Jim Wagner
Oregon Research Electronics, Consulting Div.
Tangent, OR, USA
"The only thing standing between us and victory is defeat" P.G.Wodhouse in Wooster & Jeeves series
|
| |
|
|
|
|
|
|
Posted: May 01, 2012 - 09:39 PM |
|
Joined: Sep 04, 2002
Posts: 21271
Location: Orlando Florida
|
|
| I think AOA sensors sniff a differential pressure coming from two ports in the AOA probe. I think. |
_________________
Imagecraft compiler user
|
| |
|
|
|
|
|
|
Posted: May 01, 2012 - 11:48 PM |
|
Joined: Nov 09, 2011
Posts: 400
|
|
I've only ever seen rotating vane based angle-of-attack sensors. And it seems to that would be the simplest approach - you only need a rotary position sensor to detect the vane's angle. I think pretty much any rotary encoder will do, although you should choose a non-contact sensor since you want the longest endurance.
- S |
|
|
| |
|
|
|
|
|
|
Posted: May 02, 2012 - 12:49 AM |
|
Joined: Sep 21, 2005
Posts: 2299
|
|
Keep in mind the AOA has no relation to ANYTHING other than the angle between the oncoming air and the wing chord. That is it.
I think the two common ways are mechanically (wind vane type), or measuring pressure differentials.
AOA gauges are not common on GA aircraft. I don't know much about large aircraft, but I think sometimes you will see wind vane types attached to the front of the fuselage.
I'm not sure, but I would guess differential type would be easier as you don't have 'mechanical' things to deal with. For $200 you could get a probe that would get you started quickly-
http://www.dynonavionics.com/docs/D10A_Feature_AOA.html
I don't know what kind of pitot tubes are used on gliders, but if these are similar, you could replace it with one of these. Run an extra tube, hook up a couple pressure sensors and you would then be on your way to gathering data.
(With mechanical vane type, you need to find a good place to mount- which is probably easy on a glider since you don't have a propeller screwing things up. The austriamicrosystems encoder listed previously would probably work good if you want to go this route, as it seems to have a high resolution.)
Of course, depending on what country you are in, you can't just start attaching things to an aircraft without paperwork (contrary to what most textbooks claim, paper makes aircraft fly). |
|
|
| |
|
|
|
|
|
|
Posted: May 02, 2012 - 02:55 AM |
|
Joined: Aug 19, 2003
Posts: 397
Location: Australia
|
|
|
Quote:
Aren't the simple old-time attitude indicators/artificial horizons nothing more than a weighted sphere in liquid?
Probably the very very first ones, but the mechanical units in widespread use today are spinning gyroscopes driven by air-suction or electricity. |
|
|
| |
|
|
|
|
|
|
Posted: May 02, 2012 - 09:35 AM |
|
Joined: May 01, 2012
Posts: 5
|
|
Thanks everybody!
Really apreciate the comments!
A rotary encoders sounds like the way for me to go!
The rotating vane AOA sensor should be able to detect changes in the relative wind. When flying through a thermal the vane should deflect and increase the AOA. The reason i dont want to go with the pressure sensors are that the data aquisition system used on the flight testing, the air ports are used for other sensors. The mechanical sensor will be better as there are 2 analogue ports open. Any ideas on where to begin calculations for the vane element? |
|
|
| |
|
|
|
|
|
|
Posted: May 02, 2012 - 11:04 AM |
|
Joined: Sep 21, 2005
Posts: 2299
|
|
|
Quote:
When flying through a thermal the vane should deflect and increase the AOA
Maybe initially, briefly. You are in a fluid (air), and the glider (and wing) care not how that air is moving relative to the earth. You can be in a rising thermal, or a descending thermal, or no thermal at all, all at the same AOA. The only instrument that can tell you if you are rising/descending is a pressure sensor (altimeter, vsi, variometer, good gps, etc.).
It almost sounds like you are wanting to detect thermals, which is what a variometer is used for (and is probably already installed in the glider).
Quote:
The reason i dont want to go with the pressure sensors are that the data aquisition system used on the flight testing, the air ports are used for other sensors
Then you already have half the differential system, assuming you can get at that data. You can also tie in other sensors to that same pressure line, the other sensors will not care.
Maybe you need to specify precisely what your goal is before you get too far. |
|
|
| |
|
|
|
|
|
|
Posted: May 02, 2012 - 11:18 AM |
|
Joined: May 01, 2012
Posts: 5
|
|
the problem statement given to me was:
"For a body to have minimum drag it must be positioned accurately in the flow medium. The fuselage of the JS1 is designed form minimum drag when the stream lines are parallel to the cockpit centreline. The position of the centre of gravity, mass and the flap position however determines longitudinal angle of the fuselage with respect to the airflow. To further optimize the performance of the glider it is necessary to measure this angle
Problem definition
Design and execute a test to measure the in flight fuselage angle relative to the airflow."
As far as I know the AOA will differ when moving through different air currents. I am also unsure about the rotating vane sensor. Will the vane not flutter because of the oncoming airflow and will mass balancing the vane sort out this problem? |
|
|
| |
|
|
|
|
|
|
Posted: May 02, 2012 - 01:12 PM |
|
Joined: Jan 03, 2006
Posts: 4417
Location: Hemel Hemsptead, UK
|
|
Expressed like that, then the problem becomes easier.
The AoA will not change except for pilot inputs (push forward, ground gets nearer etc.) and momentarily entering and leaving a thermal.
A symmetrical airfoil pivoted ahead of its nominal centre of lift should track the actual airstream without fluttering though I'd expect a flat plate to flutter.
The angle of the chord of the wings provides your reference point - if the wing has flaps the chord and therefore the AoA will change when they're active, but that wouldn't matter in constant flight. |
_________________
Neil Barnes
www.nailed-barnacle.co.uk
|
| |
|
|
|
|
|
|
Posted: May 03, 2012 - 01:47 AM |
|
Joined: Nov 09, 2011
Posts: 400
|
|
The problem statement seems to be saying (in a long winded way) that the angle of attack sensor is used by the pilot to adjust pitch for best glide performance.
I know very little about the fluid mechanics involved, but I'd expect that you would control flutter by damping the rotating motion, rather than by adjusting the mass of the vane.
Do you have to design the vane yourself, or can you buy an off-the-shelf instance?
Just out of curiosity, will this be tested by actually flying a glider, or will it be done in a wind tunnel. Who gets to do the flying  ?
- S
ps: there's some good background material on angle-of-attack at John Denker's excellent "See how it flies": http://www.av8n.com/how/htm/aoa.html |
|
|
| |
|
|
|
|
|
|
Posted: May 05, 2012 - 09:41 AM |
|
Joined: May 01, 2012
Posts: 5
|
|
yeah, I have to design the vane my self
The equipment will first be done on a bech, but the actual test will be during flight |
|
|
| |
|
|
|
|
|
|
Posted: Jan 02, 2013 - 08:48 AM |
|
Joined: Jan 02, 2013
Posts: 1
|
|
Hi Everyone!
Lotterels, how far did you manage to go with your experiment?
Which rotary sensor did you use? was the measurement acceptable to your needs?
Could you make aerodynamic performance assessment?
Did it fly?! |
|
|
| |
|
|
|
|
|
FAQ
Search
Register
Log in to check your private messages
Log in
Maximize
RSS
