Way cool, flies so smooth. Maybe a pony in the pile for future foot-launch aircraft? We are having discussions currently about making the winglets on my wing fly-by-wire.
Pitch axis has proportional angle of attack error, pitch attitude error, and pitch rate feedback. The yaw axis is proportional sideslip angle error and yaw rate feedback. Roll axis is proportional roll angle feedback only. There are no integral loops used or gain scaling. This is the simplest controller I could come up with and it could be made better in the future.
Could you asymmetrically adjust the span loading for yaw control? Adjusting the span loading will in turn change the upwash and downwash across each wing, so the induced drag would shift and change the yaw moments. The Prandtl bell shaped lift distribution with it's higher downwash in the middle and upwash at the tips is something you could potentially design in to help reduce/avoid adverse yaw associated with roll.
To obtain a pure yaw input using this approach would require a highly distorted span load because lift and roll moment would have to remain unchanged. I don't think this will work well unless coupling between the axes can be tolerated.
@@smorrismlbco Certainly easier to model it that to make it work in practice. The effect of control surface deflection on profile drag would also make things harder to juggle for a wide range of control movements. The use of bell shaped lift distribution would probably make this juggling act simpler as the tips could provide proverse yaw. If the craft was stable in yaw then you might even be able to get away with just two inboard control surfaces, one per wing. However, this is whole different experiment than craft that is unstable in all axis and needing to actively manage stability like the one you've just built.
@@robertosfield BSLD is used to minimze adverse yaw on swept wings that already have some mechanism for yaw stability like wing sweep. BSLD alone will not stabilize this design in yaw.
@@smorrismlbco I wrote "if the craft was stable in yaw" in this context assuming one would need to add vertical surface to enable yaw stability. I think it might be possible for bell shaped lift distribution to provide a little yaw stability in combination with appropriate polyhedral with a straight wing - think gull wings and how the lift distribution changes on yaw and how this then could provide helpful the yaw moment if done right.
@@robertosfield In my opinion, straight wings are different from sweptback wings... The yaw stability is mainly provided by the sweep instead of dihedral. Merely dihedral cannot stablize the wing, especially with a large moment of inertia.
The pots come from a Hitec HS 422 servo. They have low enough friction to work at low airspeed with a large vane. I read them using the computer's A-D channels.
Depends on aspect ratio and airfoil shape. I have a stable flying wing(no sweep back) with a somewhat lower aspect ratio which is easy to control by manually adjusting a flap, ailerons, and throttle. My airfoil is unlike the traditional reflexed airfoils seen in the literature. I haven't experimented enough with aspect ratio to know how large I could make it.
I don't have enough instrumentation on the model to measure it and I have not flown it with a stable CG setting. Theory predicts that the efficiency of the stable design can be close to the unstable design at a single flight speed but will degrade at all other conditions. If you include slow speed flight, the difference will be significant because the stable model will have a much lower CLmax.
![Blox Fruits Dragon Rework Update [Full Stream]](http://i.ytimg.com/vi/EqDAp8udhm0/mqdefault.jpg)
Way cool, flies so smooth. Maybe a pony in the pile for future foot-launch aircraft? We are having discussions currently about making the winglets on my wing fly-by-wire.
I'll be bugging you about some good baseline control gains for a similar wing that has a few extra joints....... beautiful build as always Steve
You know how to reach me, thanks!
Beautiful build. Those airflow sensors are ingenious!
Very cool! Is it a simple proportional controller between the windvanes and the control surfaces? Or is it a full PID controller?
Pitch axis has proportional angle of attack error, pitch attitude error, and pitch rate feedback. The yaw axis is proportional sideslip angle error and yaw rate feedback. Roll axis is proportional roll angle feedback only. There are no integral loops used or gain scaling. This is the simplest controller I could come up with and it could be made better in the future.
Now this is a great build! WELL DONE!
This is cool, using technology as a tool to extend human capabilities!!
amazing !
Could you asymmetrically adjust the span loading for yaw control? Adjusting the span loading will in turn change the upwash and downwash across each wing, so the induced drag would shift and change the yaw moments. The Prandtl bell shaped lift distribution with it's higher downwash in the middle and upwash at the tips is something you could potentially design in to help reduce/avoid adverse yaw associated with roll.
To obtain a pure yaw input using this approach would require a highly distorted span load because lift and roll moment would have to remain unchanged. I don't think this will work well unless coupling between the axes can be tolerated.
@@smorrismlbco Certainly easier to model it that to make it work in practice. The effect of control surface deflection on profile drag would also make things harder to juggle for a wide range of control movements.
The use of bell shaped lift distribution would probably make this juggling act simpler as the tips could provide proverse yaw. If the craft was stable in yaw then you might even be able to get away with just two inboard control surfaces, one per wing. However, this is whole different experiment than craft that is unstable in all axis and needing to actively manage stability like the one you've just built.
@@robertosfield BSLD is used to minimze adverse yaw on swept wings that already have some mechanism for yaw stability like wing sweep. BSLD alone will not stabilize this design in yaw.
@@smorrismlbco I wrote "if the craft was stable in yaw" in this context assuming one would need to add vertical surface to enable yaw stability.
I think it might be possible for bell shaped lift distribution to provide a little yaw stability in combination with appropriate polyhedral with a straight wing - think gull wings and how the lift distribution changes on yaw and how this then could provide helpful the yaw moment if done right.
@@robertosfield In my opinion, straight wings are different from sweptback wings... The yaw stability is mainly provided by the sweep instead of dihedral. Merely dihedral cannot stablize the wing, especially with a large moment of inertia.
Very cool! I was excited by the alpha-beta probes, but it looks like those are just potentiometers? Any extra information about that?
The pots come from a Hitec HS 422 servo. They have low enough friction to work at low airspeed with a large vane. I read them using the computer's A-D channels.
I wonder how long a really really good pilot could keep it in the air without the stabilization?
less than 1 second. It's too unstable to be controlled by hand
Depends on aspect ratio and airfoil shape. I have a stable flying wing(no sweep back) with a somewhat lower aspect ratio which is easy to control by manually adjusting a flap, ailerons, and throttle. My airfoil is unlike the traditional reflexed airfoils seen in the literature. I haven't experimented enough with aspect ratio to know how large I could make it.
How much more efficient is this compared to a normal aircraft, with the center of gravity in front of the neutral point?
I don't have enough instrumentation on the model to measure it and I have not flown it with a stable CG setting. Theory predicts that the efficiency of the stable design can be close to the unstable design at a single flight speed but will degrade at all other conditions. If you include slow speed flight, the difference will be significant because the stable model will have a much lower CLmax.