Working as a BYU electronics shop tech, I occasionally get to help out the guys at the MAGICC lab. A PhD student came in a few weeks ago with a problem: He wanted to make an Alpha-Beta probe. Encoders of the accuracy he wanted are too heavy and too expensive to put on a 5 pound flying wing, and the multiple Pitot-tube option was not his game to play either, so he was wondering if we could make encoders like the big expensive ones he was looking at, except for being big and expensive.I got to thinking about P-factor in flying tail-draggers and high AoA flight: Props that are presented with a relative wind that is off-normal to the plane of rotation produce asymmetric thrust. Using that principle in reverse, small pager motors could be mounted to a Pitot-probe, each offset by 90-degrees. It is possible to work up different configurations, redundant placements of similarly oriented pager motor "generators" and the like, but the basic idea is that as long as the props on the pager motors are uni-directional (no camber/bias for rotation in forward verses reverse), and as long as the props are kept out of dirty air or asymmetric dynamics due to the Pitot-probe, these should make a simple Alpha-Beta measurement system.As to taking the measurement: My hunch is that the ratio between the potential differences made by the motors resolves the alpha and beta components of the airflow. That or the current generated when a load is place accross the terminals of the pager motors.So has this been tried before? What are some issues?
As posted there, here is an update on my progress. I've hacked a Airhogs micro chopper thing that my brother donated to me in the name of science. The little pager motors on the frame are absolutely perfect! The props are just flat blades, no camber to mess up readings taken when the prop disc is flat to the relative airflow. Next up: to attach and solder the pair to a pitot-tube.
I think you would have serious problems with weight and accuracy, though.
What level of accuracy do you need? A very lightweight solution would be to use freely-pivoting vanes, and point a camera at them from the airframe. (The vane support would provide a reference -- the vane could have some thickness at the tail end, to help with visibility.
Over limited ranges one could use themistors in three tubes oriented a few degrees aprat the flow over the thermistors can be measured by measuring their temp after self heating and the three mneasurments could be combined to give you your aoa.....
Another approach that might keep the weight down would be to use vanes that are hooked up to mechanical linkages through the probe mount back to stripped down servos inside the airplane. Just use the servo pot to get an analog voltage off of the vanes (strip the gears and motor out). Obviously the genius is in the mechanics becuase it idealy should be friction free. Got to love the BYU guys always pushing the limits
I bet inertia. I think alpha/beta change frequently enough that when you go back to "neutral" your props will still be spinning. Why not just use simple "wind vane" sensors with a high rez optical encoder? I have seen AOA sensors done before years ago by charles river R/C using this method. I bet sideslip could be measured in the same way.
Comments
http://diydrones.com/forum/topics/angle-of-attack-sensor
As posted there, here is an update on my progress. I've hacked a Airhogs micro chopper thing that my brother donated to me in the name of science. The little pager motors on the frame are absolutely perfect! The props are just flat blades, no camber to mess up readings taken when the prop disc is flat to the relative airflow. Next up: to attach and solder the pair to a pitot-tube.
http://home.alphalink.com.au/~derekw/anemometer_diy/anemain.htm
I think you would have serious problems with weight and accuracy, though.
What level of accuracy do you need? A very lightweight solution would be to use freely-pivoting vanes, and point a camera at them from the airframe. (The vane support would provide a reference -- the vane could have some thickness at the tail end, to help with visibility.