I've put some thought into this myself because with a little bit of custom software you could do this with a TUT board.  So, yes it's possible, probably even practical in the sense that you could make the model do what you specify, at least at speed.

As to whether it'd be practical for stunt -- maybe not.  Think how fast the airplane rotates in response to elevator with the airplane stopped.  That's because the sensitivity of the airplane's rate vs. elevator displacement is zero at zero speed, low at low speeds, and high at high speeds.  So you'd need some sort of a speed-sensitive gain from handle to rotation.  Also, stunt planes are generally designed to be quite stable, with way-far-forward centers of gravity compared to any other form of aviation.  This should translate into a marked tendency for the tail to weather-vane into the direction of the airflow, which you cannot do with a TUT or anything else unless you add vertical flow sensors to the mix (which you could do with a TUT, if you wanted to go there).

I think that if you did it you'd want to have a linear relationship between flap and handle, and a nonlinear relationship between handle and pitch, to give you very little rotation for the amount of up and down you get during level flight and straight segments, yet still get lots of rotation in corners.  This is more or less the "Igor flap" concept, just tunable in electronics.

I'm not trying to drive you off of the idea -- I think it's a worthwhile thing to try, particularly because it would allow you to move the CG back until the airframe's pitch stability is zero or even negative.

It's just not something you want to slap together.

Tim, I didn't see your comments as negative.  I was just considering the idea as a thought experiment.  The fly-by-wire concept is intriguing, but I doubt I'll ever try it.

Guys,

I'm curious about a variation on the electronic bellcrank concept.  Instead of handle movement commanding a certain elevator position, what if handle movement commanded (proportionately) a certain pitch RATE?  When the handle is neutralized, the model would maintain its current pitch attitude until another correction is received.  Would this be possible or practical, and would it make for smoother flying?  

Not sure but isn't this the same as Kim's fly by wire setup a few post's down?

If you have a good stunt plane and know how to trim it better than I do, you can do this aerodynamically with traditional control linkage.  I flew an Impact once (not mine) that looked like a poor simulation of a stunt plane. The engine made the same sound everywhere, the airplane went the same speed everywhere, and I saw only the profile of the plane: never any wing, never any oscillation.  A handle input resulted in a pitch rate proportional to input.  It was weird.  
[/quote]


I believe there is only one  person (other than Howard) who knows what you are describing.  

Fellow Ones and Zeros CL Flyers-

I am considering trying something along these lines (no pun intended) as well. The feedback potentiometer (pot) of a gutted R/C servo with ball bearings on the output shaft could serve as the bellcrank/encoder. They make huge output arms about the same size as a traditional bellcrank. The pot would provide analog input to a microprocessor and from there it's just software to generate (3) R/C servo pulses : 1 for each flap and 1 for the elevator. Via software, do some R/C tricks like exponential throw on any/all of all (3) control surfaces. Change the Flap/elevator ratio exponentially (?). Keep pushing the C.G. aft as expo helps with level flight stability.

Anyone already tried this? (and failed)? or wants to join me in this? I could do Arduino, or MPASM or maybe just C. Probably want a processor that has a 12 bit A/D.

Cheers,

Steve Manganelli

I hope we are talking theory and sport flying.  This could not be used in any sanctioned competition.  It is fun to play with but nothing, in my lifetime will make better corners than a clear head, well-trimmed plane and a firm grip on the handle.

Ken

Oh, and Practice, Practice, Practice

Of course it is possible. However, I don't think this will lead to smoother flying.

Consider level flight. If the altitude of the plane fluctuates due to air turbulence, you have to correct that by counteracting that with two inputs: up/down to get back to the original altitude and then correct again for the pitch attitude. That is exactly what you do when using the traditional way.

Consider flying a looping. You don't want a constant pitch rate at all, you want a constant radius of the flight path. Initially, during the first half of the looping the airspeed of the model will decrease, with a corresponding reduction of the radius of the flight path if the pitch rate remains constant, and the opposite in the second half. So, you have to correct the pitch rate continuously to compensate for this, which again, is exactly what you do in  the 'normal' way.

Same goes of course for all the  other manoeuvres.

You're right of course about flightpath vs pitch rate. I never considered my electronic bellcrank to command a pitch rate. I fly most of my R/C Pattern (Stunt) models with 30% expo on ailerons and elevator and a much more aft C.G. then normal CL position. Soft neutrals allows  you to sneak back to the right heading without the judges catching you (as often).

On modern fighters, the stick deflection indeed commands a rate. The computer takes the pilot's intent under advisement and then decides what control surfaces to wiggle. Could be stabilators, LE flaps, TE flaps, ailerons, spoilers,  rudder(s). Besides the pilot's intent, the computer uses airspeed, AOA, altitude, attitude, Gs in order to help decide what to wiggle. I'm sure I left a few out. I've never seen an AOA probe on any model airplane but the to make an airspeed pitot for R/C.

Cheers,

Steve M.

You're right of course about flightpath vs pitch rate. I never considered my electronic bellcrank to command a pitch rate.

So, all you will do is translate the input from the handle to a flap- and elevator deflection. Difference it will make is that te steering force on the handle is greatly reduced, I don't know if this is a benefit, you don't feel anything anymore.
Apart from this, you have to consider the extra weight you introduce with the servo's and electronics, and you will have to take care with the poisitioning of the servo's in such a way that you don't introduce a yaw moment to the plane.

I am still having trouble wrapping my brain around this.  If you are simply disconnecting the flap pushrod and replacing it with servo's I can relate to it.  So, forget the rules for now.  Rules change with technology even if this one is as chiseled in stone as you can get.

What you are talking about is *elimination* of control force feedback through the lines.  RC on a tether.  I don't think many of our expert level fliers will accept that.  We rely on that feel to do proper locking out of corners.  Reducing excessive pressure may be beneficial.  I am experimenting with spades right now.  I put on a set that was too large and experienced what you are trying to achieve - no handle pressure.  I was able to turn ridiculously tight corners, but the plane had no idea where to go after the corner.  No feedback would require 100% reliance on visual input and timing, and we simply do not fly that way, at least I don't.  I realize that spades are reducing the actual pressure on the flaps and your system will not do that but the feel at the handle will be the same.

Having said all of that I really like what you are doing, and I hope you are successful.  Just don't expect a rule change under AMA to allow for it in competition. 

Ken

Elimination of control force feedback is an unintended and probably undesirable side effect (?) However, I'm too new at CL to know what I'd be missing. How much "feedback" comes from twisted lines? I could start by clearing the lines, then pre twisting the handle X times.

Thanks for the Input,

Steve M.

Elimination of control force feedback is an unintended and probably undesirable side effect (?) However, I'm too new at CL to know what I'd be missing. How much "feedback" comes from twisted lines? I could start by clearing the lines, then pre twisting the handle X times.

Thanks for the Input,

Steve M.

It is not that kind of feedback.  With the modern pattern you only get a maximum of three wraps at any point and for most parts of the pattern less.  The friction from three wraps is virtually nil with the plane under power.  What I am referring to is the resistance that feeds back through the lines to moving the controls.  The more you move them the stronger it gets even to the point of no longer moving before reaching the full range.  Everybody uses that feedback in their own way, but I think most of us rely on it to some degree.  In my case it is an early warning to the plane starting to "wind up", whether I have "hit the wind" where I expected starting a maneuver or when I am reaching the release point in a corner.  It is all done subconsciously in microseconds which means it is super hard to do without and difficult to duplicate.

Having said that, what you have accomplished is fantastic and I hope you keep working on it.  "Proof of concept" is always the first step and you have done that.  Addressing the unexpected and solving those issues is the real work.

One + that I like is that you should have full control with very little line tension which is an issue we face in calm conditions and high winds.

Ken