MORE ON BIRD CONCEPTS

This is a series of letters between enthusiasts wanting more information on the use of "tip feather" type control systems for both models and full sized aircraft.  Each of the letters has been answered by Bob Hoey, and we appreciate him taking the time to help with his comments and observations.  Click here for more of Bob's work.


February 1, 2008

Hello Bob.  Love your bird models. 

I have your plan for the Turkey Vulture and have yet to build it.  Might try to have it ready for Soar Utah this September.
.......I have
some questions about designing birds:
           
<>I assume the structure for all your birds is basically the same with the exception of the wing-tip controls.  On the seagull  and pelican, did you place the wing servos next to the body as you did on the vulture?  I would think the complex dihedral/annhedral/swept spar would make it difficult to run a carbon rod from the fuselage to the wing tip.   How did you do it?  Was the servo placed at the outside wing beak?
            Did you use the same airfoil for all your bird models? 
  I am considering designing a albatross with a conventional structure (wood).   I think that the basic structure of the vulture might work.  I would redesign the wing to be a two piece affair for transport.  Also, I would like to keep the wing-tip controls like your birds and not use ailerons.  Are the twisting wing tips on your seagull effective?
            Thanks for  reading this and thanks for your time. <>

Jeff Meskey  <Jeff.Meskey@SSAMarine.com>
SSA Utah

Hi Jeff,

You sound like an "experimenter" like me. An Albatross should be a good, but challenging, project. I can probably give you some tips that will help out.  First I'll answer your questions directly.
        Yes, the basic structure is the same for all, except for the wing spars, which I'll describe later.
        The Seagull has servos at the wrist joint, and carbon torque rods from there to the tip aileron (like the vulture). It works OK, but goes against the desire to keep the wing inertia as low as possible (to keep the roll oscillations down).
        The Pelican has the servos at the center section of each wing with a standard, flexible push-pull tube that operates a bellcrack near the tip. That is what I would suggest for your Albatross. There is a little more slop, but it works just fine.
        Until recently I have used the same airfoil for all the birds - undercamber with reflex at the trailing edge. It works well for low speed thermal soaring but the undercamber provides a lot of drag at higher speed. I recently tried an airfoil that has the same upper surface as the Turkey Vulture model, but the bottom surface is flat between the leading edge and the forward edge of the trailing edge piece. This removes the undercamber, but retains the reflex at the trailing edge which is necessary for pitch trim. It flies just fine and seems to have better penetration at higher speed. Should work OK on the Albatross.  (which, of course, is a pretty high wing loading, fast bird).
        I agree with a 2-piece wing for transport, and also to allow you to change the dihedral by bending the connecting plate or tubes.
        The seagull tip ailerons work just fine and will allow you to select a trim bias position that will eliminate adverse yaw that you would get from ailerons.
        Now some suggestions;
                (1) You should probably build in 3 or 4 degrees of washout in the outer wing panels outboard of the wrist joint. (I have 4 degrees on the Seagull)
The "sweet spot" for the seagull model is with the tip aileron at a -7 degrees incidence with respect to the wing center section. ( best coordinated turns)
                (2) The wing spars for the Seagull and  Pelican were built up as follows; The main spars (full depth) were cut from a sheet of 3/16 (or whatever, balsa) to the shape of the wing dihedral (front view). A 1/16" by 1/4" cap strip was glued to the top and bottom of the spar. A strip of .007" by 1/4" carbon ribbon was glued to the 1/16" cap strips. When the structure is completed, the wing is sheeted, top and bottom, with 1/16th sheet balsa forward of the main spar to the leading edge.  This traps the carbon ribbon between the cap strip and the sheeting which produces a very strong, light weight  "D" spar wing and helps retain the complex dihedral shape over the span of the wing.
        A little bit complex, and very difficult to build without warps, but no one said it would be easy!!
        All birds are pretty short-coupled in pitch- especially the albatross.  And it doesn't have a very large tail. The reflex in the wing arifoil will be pretty important to allow you to trim it.
        I have found that starting with the little round disks (mounted near the wing trailing edge, angled in about 20 degrees) helps to provide some level of directional stability during the early flights. The final tweeking (eventually eliminating the disks) should be done by adjusting the total wing dihedral. It only takes a couple of degrees to change from an unstable airplane to a marginally stable, but nice flying model.
        I have attached some photos to help with the above comments. I have photos of the construction phase of the Seagull and Pelican model, but they have never been published.  (Click on this link then go to the bottom of the Seagull and Pelican pages to view the images.)
        If I can help in any way, let me know, and by all means, let me know how your Albatross flies!!
 
Hoey



August 2, 2004
Hi Bob:

I  live in South Africa, fly radio control gliders, love watching soaring birds and am fascinated by soaring flight. I happened across your bird models on the TWITT page and really enjoyed your very realistic looking bird models and must say that you did a good job on the building side too, with those compound curves. What grabbed my attention was the use of the tip feathers to do roll control. I made a 1/4 model and then later a full scale 10m glider with a team and entered the Red Bull Gives You Wings Competition of 2000 and 2002, using winggrid as tips for reduced induced drag and trying to optimize for slow flight and short span. On this glider i used the front winggrid for roll control almost like a spoiler, creating drag and reduced lift on the side i wanted to turn too, 

http://www.winggrid.ch

figure 3 and 4 shows pictures of the model and the actual glider). I would like to have more information on your aileron/tip feather control system if any is available.
       I noticed that you put kinks in the tip feathers (bending upwards) to simulate those of real birds. From the information on the winggrid site (equal angle configuration), I learned that the front feathers carry most of the load, and therefore bend more than the other feathers, that is why the front feathers stand up more. This of course gives a positive pitching net effect. Another question is what airfoil you used for the tip feathers, camber and thickness?  On your models you have something like the "equal load configuration" where the tip feathers each carry roughly the same load. Maybe it would help to go for cambered airfoils, thin, and reduce the negative angles, as your feather airfoils might suffer from stalling at higher angles of attack.
        I see on your models that you are using reflex airfoils and an elevator/tail, almost like the Obelix model for pitch control - and would like to know what airfoil you used for the wing. I have previously looked at the Albatross on 

http://www.charlesriverrc.org/articles.htm

using positive pitching airfoils with more rearward cg and fly by wire pitch stabilization, I was wondering if you have considered trying this type of application on your models.
      Looking at the page on trim

http://www.charlesriverrc.org/articles/asfwpp/lelke_trimattitude.htm

especially Figure 11, I realized that if the cg position could moved using the trim position as reference – resulting in no trim deflection (cg is adjusted as the center of lift moved from slow flight to fast flight), drag could be further reduced. The other method to do this of course is to adjust the wing planform in flight - move the aerodynamic center (moving the lift center relative to the cg), like the Exulans

http://www.diomedes.co.za/exulans/ 

This of course is not a problem for a bird in flight, and I think that birds therefore do not need reflexed airfoils - resulting in less drag.
        I am thinking of being able to adjust tip feathers for different speeds, having them sweep back for faster flight, and at the same time retract the wing - shorter span, and vise versa, sweep the tip feathers forward and extend the wing for slower flight. I am also thinking of a way to change the angle of attack of the tip feathers in flight. I would still like to have the tip feather angle of attack in line with the camber line of the wing airfoil. But I need to finish the designs and test them to learn more, be it failure or success.
        Congratulations from my side for your achievements thus far. I would like to know of more as you progress - if you want to share it with me.

Tony Neerings
tony.neerings@iscor.com

--------------------------------------------------------------------------------

August 16, 2004


Hi Tony:

Andy at TWITT forwarded your message.  It sounds like we have a lot in common, or at least are both pursuing the same interests.  One of my personal failings is that I'd rather "do it myself" than do web research and find out what others have done. A couple of years ago I DID spend some time looking at the "winggrid" concept. I came away with SOME understanding, but an awful lot of puzzlement. 
........My initial model testing was aimed at flying without any vertical surfaces, as the birds do. I used drag flaps under each wing, or spoilers on the top of the wing, to initiate turns. Both worked OK, but the spoilers were more linear.  My wing-tip ailerons did not evolve from the winggrid concept.  I did several tests on bird models with different wing tip configurations and found very little difference. Some were worse than others, but none provided any magic improvement over a standard airplane-type rounded tip.  I established a different notion about what's going on at a bird's wing tip. If you were to remove the six primary feathers from a bird's wing, leaving a squared-off tip, you would expect a strong tip vortex to form. That means that the air immediately outboard of the squared-off tip, (where the tip feathers would normally be) is in a strong upflow region as the air flows upward and back around the tip. 
........The tip feathers are merely blades stuck out into this upflow region, and the bird can adjust the local angle of attack of each blade to create additional lift (either up or down).  This theory helps correlate the observation of fairly large negative angles of incidence for the forward tip feathers on a soaring bird, yet they are bent upward indicating that they are creating lift. The significance of all of this becomes more obvious when you analyze the effect for differential deflection of the tip feathers for roll control. The forward feather is creating a local lift force that is canted forward due to the local upflow. That's thrust!!  Normally we consider that a downgoing aileron on the trailing edge of a wing will generate more drag and thus create adverse yaw which slows the turn. For the forward tip feathers the effect will be to create proverse yaw, which will help the turn.
........To test this theory I built a wingtip that had the 3 forward feathers mounted on a single spanwise axis. There was a change from adverse yaw to proverse yaw as the initial bias position of the ailerons was changed. The best bias position was with the forward feather at a whopping big -27 degrees of incidence relative to the chord of the rest of the wing.  I built a model of this wing tip and ran it in the NASA DFRC water tunnel. These tests confirmed that the flow over the forward tip feather was smooth at -27 degrees, but would stall (on the bottom) at -35 degrees, and stall (on the top) at -22 degrees. 
.........I will attach some photos of the tip aileron for my Turkey Vulture glider model. I am using the throttle control on the transmitter to adjust the bias position (both tip ailerons together), and the aileron control stick to deflect the ailerons differentially. There is a "sweet spot" for the bias that corresponds with the -27 degree front feather, that allows smooth, coordinated turns, with NO vertical tail.
........The feather airfoils are merely flat plates of 1/16th balsa, with rounded leading edges.  The wing airfoil is a computer-generated airfoil which produces a positive pitching moment of about .02. It has a lot of camber and therefore does not penetrate well. The intent for these bird glider models was to try to duplicate the soaring flight regime of birds, where fairly high camber and slow speeds are present. I'll attach an EXCEL spread sheet that has the airfoil coordinates.  I have read and understand the fly-by-wire pitch stabilization for the Albatross, but don't think that complexity is necessary for what I am trying to do.
........I recently did an experiment where I put three hinges on the wing - one at the centerline and one at each wrist joint, to try to adjust the wing sweep in flight. (Inboard panels rotate forward, outboard panels rotate aft) My primary purpose was to shoot for a high speed configuration that would be more like the birds use between thermals. The linkage got rather complex, but worked OK.  Unfortunately, I used a different airfoil which didn't have nearly enough reflex, and I misjudged the amount of tail deflection it would require for normal straight flight. (It requires full up elevator for both wing configurations). I will need to build another wing (ARGHHHH!) in order to continue the test. Notice that this triple hinge arrangement, if done properly, reduces the span, and also the area, but does not significanly alter the cg position. These birds are pretty smart!  Incidentally, the feather angles on my more successful bird ailerons are roughly the same angle as the upper surface of the wing airfoil, not the camber line.
........I did see some slight improvement in the glide angle when I added the wing tip ailerons to my Raven model. 

Stay in touch,

Bob Hoey
bobh@antelecom.net


July 20, 2004
Hello Everybody:

My latest brain spawn is in the Birdwing configuration.  It comes from the idea to get a BLUM (Basic Ultralight Motorized) in a garage without a lot of work. No need to fold the wings, no need to dismount and carry large parts (like Mike Sandlin did with his GOAT glider. He quickly changed his opinion).  Sources of inspiration are:

1) Pelican
2) Superfloater (ultralight glider)
3) Mosquito harness for hanggliders
      I use a easy to make contant chord central wing and I add smaller wingtip feathers to increase the span ... and hopefully the effiencienty of the square wing. To get the wing in the garage you need to flap the rear part of the wing upwards. It looks like a control surface, but it isn't. It is just a part of the wing that can be folded upwards to reduce storage size. This folding can be done by a easy piano hinge (I guess). No heavy or difficult hinge. If that part is folded upwards, the length of the airplane is reduce to 2,2 m!. OK, you need to remove the tail as well. But I saw that the Flying Flea tail is so easy to dismount and mount I want to use the same system. Again, not a great deal of work to create and so do at the airstrip to get it in flying condition. 
........The feathers ... well... I am still hoping to find easy mount and dismount systems in other sports. That I still need to work out (in the basic concept idea). But, you will get easy to handle small parts. The elevons are hinges at a small section of the rear spar of the wing. So ... the first part still has the last part of the wing, but there is another part connected to that part and that is the rest of the elevon.  Why this system? Well, I hoped to avoid complicated hinges this way. OK, it will not be a piano hinge, becauseit has to be dismountable. But ... It willstill be a easy hinge. The Flying Flea tail hinge still as inspiration.
........If all is dismounted you have a total dimension of 5 m by 2,2 m. Good to fit on a trailer and be pushed in a garagebox. I guess that the main design idea in this concept is: low wingloading (as the Vulture, Sea Eagle, Condor), higher efficienty of a rather square wing by those wingtip feathers, easy to make hinges and not a hard work in mount/dismounting the parts.
........I use the "cockpit" layout of the Superfloater. Anything easier ... send it to me. You will have noticed that the prop is rather large for this skid and wheel configuration. I hope to get the prop more efficient by giving it a wider diameter and to prevent the prop from touching the ground at take off.  You need to run while two bars (those green things at the bottom of the wing) are tilted downwards and support the engine and prevent the prop from touching the ground. Once in the air, you place your feet on the footrests on the skid and you retract the green bars. Landing? I suggest to use a folding prop and just land on the skid.
........I am still doubting what kind of airfoils I should use. Autostable and have a tail with not so much negative AoA (angle of attack) or use a more lifting flat bottom airfoil and have the pitch moment corrected by a "classic" V-tail. Euh ... I think more about the autostable airfoil. Reason: I hope to make elevons in that split V-tail.
    Now my questions:
1)   I noticed in the TWITT that Bob Hoey uses -27° to 0° AoA for his set of birdfeathertips of his Pelican model. OK, those will give not much lift. Can they be counted in as wingsurface? I do mean ... I need about 12m2 (+/- 129 sq ft). Without those tips I have 1.7m x 5m (=8.5 m2) of surface. Can I make the rest in those feathers?

2)   I am still wondering what would be the best and easiest system to make the BULM turn. Only elevons or is there a chance that it will never work without some kind of dragrudder?

........All comments are welcome. I will be off for some weeks soon and I hope to use your advise while drawing in my sketchbook during the holiday.
Koen Van de Kerckhove
nestofdragons@hotmail.com
 -------------------------------------------------------------------------------------
August 14, 2004


Koen:

Here are some off-the-top-of-the-head thoughts regarding your BULM Proposal. 

(1) The wing aspect rato looks to be about 3.7, which is low, but probably OK. The good news for a low-aspect-ratio flying wing is that the chord is fairly long which results in a longer moment arm for pitch stability and controllability. The bad news is that the angle of attack range is quite large and you will need a LOT of power to take off and climb. You will have to use a reflexed airfoil, and it appears from your drawings that you have done that.

(2) Don't expect anything dramatic by way of increased efficiency from the bird-feather wing tips. I don't think they are aerodynamically any more efficient than a normal tip. The trailing-edge-feather elevons that you show will produce a lot of adverse yaw. Since you show a rudder, you can link the two and probably create a reasonable turn capability, but I suspect you may have trouble if you also intend to use those surfaces for pitch control (tip stall, etc.). You might consider using the forward 2 or 3 feathers on a span-wise axis for roll control (as I have been doing on my bird models), and a separate, inboard trailing edge surface for pitch control.  The advantage of using the forward feathers is that the neutral or zero setting for both feathers can be adjusted by rigging to eliminate any adverse yaw.

(3) It looks like your placement of the pilot is a little too far aft. I suspect that your cg will be behind the best starting location of about the1/4 chord. Even if the airplane can be balanced by the engine and fuel, your cg will vary greatly with the weight of the pilot. Try to put his belly button directly under the 1/4 chord.

(4) Regarding the rear tip feather never being stalled, I suspect that is not true. The flow over the rear feathers depends on what is happening on the forward feathers. There is no magic formula for determining the relative angle of incidence of the individual feathers as you progress aft. What is right for one condition may be very wrong for another condition. (of course the birds can vary these individual angles at will!). What I am saying is that a stalled condition on any of the forward feathers (either from too much positive, OR NEGATIVE angle of attack) will mess up the flow over the
top of that rear feather.

(5) The overall configuration has some merit, and would be worth building an R/C model to establish the best control system, airfoil, cg  location, etc. You could also incorporate some of your hinging ideas to see how they work.

Good luck!!
Bob Hoey
bobh@antelecom.net
...2/13/08..............................................................................Back to Index