Help Support HomeBuiltAirplanes Forum by donating:

  1. Feb 6, 2013 #21

    Jay Kempf

    Jay Kempf

    Jay Kempf

    Well-Known Member Lifetime Supporter

    Joined:
    Apr 13, 2009
    Messages:
    3,671
    Likes Received:
    935
    Location:
    Warren, VT USA
    Springs can keep the stick centered to give a sense of center and force increasing with distance from center only. But Gs building with stick deflection load the seat of your pants proportionally so I wouldn't say that there is no feedback. There is just no feedback in the stick. I do not think that one couldn't learn to fly a system like that Many small airplanes with side sticks have very little actual feel or force feedback. They have tiny control surface widths and so little force. Full flying tails have virtually no feedback so people add servo tabs to them to give designer feedback.

    Anyway, my point is that you can fly sims, models, games with no force feedback. It can be learned. Pull back on the stick the Gs build up quickly. Pull back softly and the gs build slowly. One thing about digital controls is that you can put in reduced control volumes and limits that are triggered by G limits or speed limits. That isn't what I wanted to start with but it could be done. Accelerometers are fairly straight forward now.

    One could add modes to the system in phases. Mode switch between an established mode and a proposed or mode that is being tested until the debugging is done.
     
  2. Feb 6, 2013 #22

    Jay Kempf

    Jay Kempf

    Jay Kempf

    Well-Known Member Lifetime Supporter

    Joined:
    Apr 13, 2009
    Messages:
    3,671
    Likes Received:
    935
    Location:
    Warren, VT USA
    I wouldn't use a wiper pot servo either. I would use a redundant position and drive circuit actuator. Still, off the shelf stuff. Hall sensors are all the rage now. In microelectronics optical position sensors are cheap and easy as well. Not a lot of resolution is needed. Speed and holding position are important.
     
  3. Feb 6, 2013 #23

    Pops

    Pops

    Pops

    Well-Known Member

    Joined:
    Jan 1, 2013
    Messages:
    7,125
    Likes Received:
    6,020
    Location:
    USA.
    I built a homemade autopilot a few years ago and installed it in my single seat Cub. I used giant scale model airplane servos to operate trim tabs large enough to control the airplane but weak enough that I could override the autopilot with the controls. I have plugged the RC receiver into the servos and operated the servo tabs on the ground with the transmitter while setting in the seat, but never controlled it in flight with the transmitter. The autopilot had roll (wing leveler), with electric trim, Pitch( lock on a pressure alt.) with electric trim. Also could set a ROC or decent and fly a standard rate turn rt and Lt. Never did hook it up to the GPS. After a few months getting tired of watching the autopilot fly the airplane,and not having the fun of flying it, I took it out. :)
    So I guess it could be a fly by wire.
     
    Last edited: Feb 6, 2013
  4. Feb 6, 2013 #24

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,379
    Likes Received:
    2,304
    Location:
    Rocky Mountains
    Just to be fair to the FBW throttle that left me stranded - when I used the word "pot" I was not meaning that it used a typical wiper/resistance potentiometer for position sensing, only that it has the position sensor in the pedal assembly.
     
  5. Feb 6, 2013 #25

    autoreply

    autoreply

    autoreply

    Moderator

    Joined:
    Jul 8, 2009
    Messages:
    10,732
    Likes Received:
    2,542
    Location:
    Rotterdam, Netherlands
    Virtually all sailplanes except for the really old ones have centering by springs, nothing else. Control feel (pitch) between 40 and 120 kts barely differs, roll is mostly internal friction, not aero forces.
    Most folks adapt without even realizing it. The only problem they run into is the lack of noise at high speeds, which can make you fail to notice you're already going really fast.
     
  6. Feb 6, 2013 #26

    Jay Kempf

    Jay Kempf

    Jay Kempf

    Well-Known Member Lifetime Supporter

    Joined:
    Apr 13, 2009
    Messages:
    3,671
    Likes Received:
    935
    Location:
    Warren, VT USA
    That's kinda the impression I have. Most really modern side stick aircraft have very little stick force and take some getting used to in terms of force of the stick vs. speed and Gs.
     
  7. Feb 6, 2013 #27

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,447
    Likes Received:
    3,245
    Location:
    Mojave, Ca
    Though I have yet to fly it, a buddy tells me the Extra 300 with aileron spades has zero force or feedback. If you throw the stick hard over, it stays there until you return to center by hand. Just like in the hangar.
     
  8. Feb 6, 2013 #28

    Jay Kempf

    Jay Kempf

    Jay Kempf

    Well-Known Member Lifetime Supporter

    Joined:
    Apr 13, 2009
    Messages:
    3,671
    Likes Received:
    935
    Location:
    Warren, VT USA
    Yup,

    So here's another possibility: if you have a computer interface system you could have a safe mode or sport mode, and you could have an aerobatic mode. Safe or sport mode would go like this: if the pilot let go of the stick and it returned to center and there was no input for say 2 seconds the aircraft would return to straight and level. If you switched to aerobatic mode it would just continue in the direction that it was last heading. If you wanted to tie throttle into that there are more combinations. That gets more into intertial navigation sorts of complexity but there are probably simple ways to do it safely with less than Airforce level of accuracy in the sensors and a relatively low sample rate between decision events in the control system to keep it simple.
     
  9. Feb 6, 2013 #29

    Himat

    Himat

    Himat

    Well-Known Member

    Joined:
    May 5, 2011
    Messages:
    2,793
    Likes Received:
    644
    Location:
    Norway
    Fly By Wire, light airplane style.

    It would take determination and faith in the technology. Some good engineers and maybe more important, a pilot that trusted an electric powered and electronic controlled system as much as a mechanical system in a light plane.

    It have been done, actually on a higher level: http://www.homebuiltairplanes.com/forums/rotorcraft/11340-very-exciting-first-manned-flight-electric-multicopter.html

    This is not just electric actuators on the control surface instead of mechanical ones, it’s fly by wire with a control computer doing all mixing and dynamic stabilization of the vehicle.

    The military drones don’t fall out of the sky in droves. Not even the small simple ones. R/C planes mostly crash because of pilot interference. I do think that reliability is ok if operated by a competent engineer.

    R/C model airplanes are absurdly strong. Maybe, but I have folded more than one set of wings.:) Actually I do think that pilot induced oscillations might be more problematic. In the 1950 when fighter planes first got boosted controls a lot where crashed due to this. SAAB got the same problem, pilot malfunction, which had to be rectified within the FBW software.

    Maybe the weight question here is similar to weight of composite design vs. metal design. Design a composite airframe with “metal” thinking and it’s at least as heavy. Design as composite with “composite” thinking from the start it is lighter. This might apply in this case too.
     
  10. Feb 6, 2013 #30

    rtfm

    rtfm

    rtfm

    Well-Known Member

    Joined:
    Jan 3, 2008
    Messages:
    3,190
    Likes Received:
    571
    Location:
    Brisbane, Australia
    Hi Autoreply,
    OK, I'm intrigued. What would be the value of splitting your control surfaces like this? How would this be more effective?

    Do I assume that in a roll, for example, one would deflect the outside flaperon panels more than the inside panels? Or vice versa? This is non-intuitive to me. Likewise with multiple rudder panels. How would you design/plan their differential deflection, and how would deflecting the panels to greater or lesser relative degree improve the rudder effectiveness? And the same question for multiple elevator panels. I just don't see how this would be beneficial.

    Sexy, yes. But more effective?

    I can quite see the value of being able to deflect all panels simultaneously for TO and landing. Huge effective flap area. But how does one use aileron control with everything deflected like this?

    Differential inner and outer panel deflection to act as air brakes on landing: get this...

    Regards,
    Duncan
     
  11. Feb 6, 2013 #31

    autoreply

    autoreply

    autoreply

    Moderator

    Joined:
    Jul 8, 2009
    Messages:
    10,732
    Likes Received:
    2,542
    Location:
    Rotterdam, Netherlands
    That's not something I can explain in a few words. Nevertheless a try.

    Theory (pure academics) say a fully morphing wing is optimal. Local deflection is roughly proportional to the spanwise station to the power 1.5 or so. Open-class sailplanes come pretty close. A modern open-class sailplane has 4 spanwise surfaces. Flap action is roughly mixed in (inboard to out) 100%, 75%, 50%, 0%, while aileron deflection is 0%, 50%, 100%, 100%. The outboard panels only move upwards (to avoid tip stall).

    In reality, breaking it up in 2 or 3 sections (drooping ailerons basically) approaches the ideal pretty well. Above everything else it allows you to design a wing where the wing stalls at every spanwise station at the same time (bad, dropping tip=>spin). That's great for max performance, except when you're likely to stall, namely with flaps. Dropping flaps (reduces local stall aoa) is a healthy barrier against tip stall, so then drooping ailerons half as much as your flaps gives you both maximum lift and a healthy protection against tip stall.



    But my preference for many small stations has another reason. Redundancy and simplicity. The individual "flaperon" sections are small enough to easily cope with a mechanical/electrical failure of one of them. Sure, you'll need a lot of actuators and trailing edge surfaces. But you can move away from the mechanical complexity. It also relieves you from some other nasty issues like flap/aileron bending, torsional stiffness etc.

    Add the increased lift and it's a pretty interesting idea. Just a reality-check; compared to a classic wing with 50% straight flaps and 50% straight ailerons you might be able to have a wing that needs 25-30% less area and thus weight. Himat's last remark is an excellent one, you really need a different (fundamentally different) mind-set to fully appreciate the possibilities. Everything changes.
     
  12. Feb 6, 2013 #32

    cavelamb

    cavelamb

    cavelamb

    Well-Known Member

    Joined:
    Nov 27, 2010
    Messages:
    304
    Likes Received:
    56
    Location:
    earth
  13. Feb 7, 2013 #33

    sachaknoop

    sachaknoop

    sachaknoop

    Well-Known Member

    Joined:
    Jan 30, 2012
    Messages:
    195
    Likes Received:
    11
    Location:
    Netherlands
    And this one, sorry I did not read this thread, but as stated in the beginning, all the stuff is there!

     
    Last edited by a moderator: Mar 20, 2019
  14. Feb 7, 2013 #34

    Dan Thomas

    Dan Thomas

    Dan Thomas

    Well-Known Member

    Joined:
    Sep 18, 2008
    Messages:
    4,858
    Likes Received:
    2,070
    Optical stuff relies on a light source (failure point) and must be sealed against dust. I would much prefer variable inductors: a moving core in a coil. They don't need a light source, they have no dust issues. They just need a small AC input and a current-measuring circuit.

    Dan
     
  15. Feb 7, 2013 #35

    Dan Thomas

    Dan Thomas

    Dan Thomas

    Well-Known Member

    Joined:
    Sep 18, 2008
    Messages:
    4,858
    Likes Received:
    2,070
    Not in my experience. We had a Cirrus SR20 whose empty weight was nearly as much as the gross of the Cessna 172, and it had only 20 more Hp and carried the same four folks a few knots faster. Not a good example of the "lightness" of composite construction.

    Dan
     
  16. Feb 7, 2013 #36

    autoreply

    autoreply

    autoreply

    Moderator

    Joined:
    Jul 8, 2009
    Messages:
    10,732
    Likes Received:
    2,542
    Location:
    Rotterdam, Netherlands
    Exactly. The Cirrus is an excellent example of metal thinking in composites. Apply "conventional design" to a FBW airplane and you'll also end up with an overweight contraption. 3 redundant connected servo's to one aileron for example are what you would come up with.
     
  17. Feb 7, 2013 #37

    cluttonfred

    cluttonfred

    cluttonfred

    Well-Known Member

    Joined:
    Feb 13, 2010
    Messages:
    6,365
    Likes Received:
    2,231
    Location:
    World traveler
    Aren't the structural design loads required for certification of composite structures higher than for metal structures? I know that used to be true and was part of the reason than the few certified production composite planes were sleeker but not generally lighter than their metal contemporaries.

     
  18. Feb 7, 2013 #38

    Jay Kempf

    Jay Kempf

    Jay Kempf

    Well-Known Member Lifetime Supporter

    Joined:
    Apr 13, 2009
    Messages:
    3,671
    Likes Received:
    935
    Location:
    Warren, VT USA
    Yup, but apply non-conventional thinking to the combination of both and I believe one could design an overall simplification of structures/parts counts and a weight reduction. Boeing did a lot of that in their new offering. They are going through growing pains but they'll sort it out. 30% reduction in weight is the number I have heard quoted. It is a lot simpler to just put a servo at a moving surface than to build a mechanical control system that spans the entire airframe.
     
  19. Feb 7, 2013 #39

    cavelamb

    cavelamb

    cavelamb

    Well-Known Member

    Joined:
    Nov 27, 2010
    Messages:
    304
    Likes Received:
    56
    Location:
    earth
    I take strong exception with your reasoning, Jeff.

    Are you thinking of something like RC model airplane servos? That won't do at all.
    Scaled up to have enough force and speed to handle control surfaces on a full sized airplane they would be huge - and heavy.

    Voice coils? Have you ever seen the power supply in an old CDC disk drive?
    And that's just to move a few light weight heads.
    NOTHING like moving an aileron or elevator.
    Dude!

    I think you need to rethink your thinking...
     
  20. Feb 7, 2013 #40

    cavelamb

    cavelamb

    cavelamb

    Well-Known Member

    Joined:
    Nov 27, 2010
    Messages:
    304
    Likes Received:
    56
    Location:
    earth
    Aircraft Cable Galvanized & Stainless Steel, Nylon & Vinyl Coated Cable, Cablelaid, & Strand | WorldWide Enterprises, Inc.

    1/8" steel cable weighs .029 pounds per foot.
    Double that for pull-pull arrangement and it comes to a whopping .06 pounds per foot.
    20 feet means about 1.2 pounds from wing tip to wing tip.


    Cost runs about a buck a foot - give or take.


    Now, you want to replace this with WHAT?
     

Share This Page

arrow_white