Other flapping wing designs
In designing of ornithopter models there are mainly two major tasks, the development of the drive technology and the development of the flapping wing. In general, the wide interest lies in the drive systems and components. But the main problem of the development of such aircrafts are the flapping-wings. In this field of design desire differs very widely from reality.
Below, the attempt is being made, to give a rough overview about the physical characteristics of known flapping wings. But this collection doesn't claim to be complete.
- 1. The bird wing, the ideal
- 2. Membrane flapping wings
- 2.1 The sail as archetype
- 2.2 Simple membrane flapping wings
- 2.3 Simple membrane flapping wing
- 2.4 Active twisting by spar rotation
- 2.5 Aeroelastic twisting by spar torsion
- 2.6 Flying wing ornithopter
- 2.7 In tandem
- 2.8 Thrust-Wing
- 2.9 Oscillating stretched wing
- 2.10 Rotating wings
- 2.11 With non-twistable arm wing section
- 3, Profiled flapping wings
- 3.1 With artificial feathers
- 3.2 With inclined hinge of the hand wing
- 3.3 Twisting by tilting the leading edge
of the wing
- 3.4 With stepped twisting
- 3.5 Twisting by stroke movement of the
- 3.6 Servo controlled wing twisting
- 3.7 Shearflex principle
- 3.8 Oscillating wing tips
- 3.9 Shell wing
- 4. Notes
- 5. Related Links to other flapping wing designs
1. The bird wing, the ideal
Naturally, the great archetype for technical flapping wings is the living bird wing. His great effectiveness due to his manifold possibilities to move purposeful and to change the shape will certainly be unobtainable in aero modelling for a long time. This is also true for his weight distribution and his sensor technology.
In this drawing by K. Herzog the anatomic subdivision of the bird's wing in arm- and hand section is pictured. It can also to be used advantageously when describing technical flapping wings. The longitudinal parts of these wing sections are rather different depending on bird species.
2. Membrane flapping wings
Membrane flapping wings especially are changing the direction of chamber in the hand wing section according to the flapping direction. This way, they can produce much thrust and achieve steep climbing flights (flying with Thrust). But up to now they are less suited for gliding flights and for flying with lift.
2.1 The sail as archetype
A sail - though in other circumstances - has about the same function as a flapping wing. It shall generate as much thrust as possible under changing approach flow directions.
By material selection, layout, division into parts, sail trim and rig tuning the sail characteristics can vary in wide ranges. Battens give the sail more stability and an optimal shape. A lot of descriptions with sophisticated tips about the fabrication of the sail and its practical use can be found.
Indeed, a lot of systems of membrane flapping wings have been developed, but detailed information about them is barely available
2.2 Simple membrane flapping wings
pinion feather by
(ca. 1937) obviously was optimized for thrust generation. Therefore, he increased
the chord in the outer wing area. But this pinion feather was not intended
for generating lift at the same time. It is merely a propeller for changing the direction of rotation.
Tim was the first in
mass-produced rubber powered flapping wing model - with
simple membrane flapping wings - invented by Albertini
Prosper and de Ruymbeke Gérard (France 1969).
The membrane printing of Tim in the marginal picture was drafted by
Under the designation
Tim Bird this model is
available in trade till today.
2.3 Simple membrane flapping wing with battens
Here a famous Membrane Flapping Wing, equipped with small
for stabilisation of the membrane, developed by A. Pénaud
more information at related link 1
2.4 Active twisting by spar rotation
This is a membrane flapping wing by Erich v. Holst (1943) with drive-controlled wing twisting in the arm wing section by spar rotation. Only the rib at the end of the arm wing (number 9) is fixed to the spar. It is linked with a crank drive which effects the stroke movement as well as the rotary movement of the spar.
The twisting in the hand wing section happens largely passively. In addition, a transition from cross to longitudinal battens can be seen. In spite of alternating profile chamber direction during a flapping cycle a relatively purposeful increase of wing twisting tip wards is made possible. The bird models by K. Herzog (1963) follow this scheme, too.
2.5 Aeroelastic twisting by spar torsion
The flapping wing model of the Czech Cenek Chalupsky (1934) was flying steadily without a tail unit. Its achieved climb power is still considered remarkable today.
- wing span
- cane spar
- 3.1 kg
- 2 m
- [109 oz]
- [79 in]
- [394-590 in]
Each flapping wing of this ornithopter has two spars. The straight, bending resistant spar (H1) transmits the power of the stroke motion. The bended torsion elastic spar (h3) determines the magnitude of the wing twisting.
Both spars cross approximately in the center of the semi-span. At the cross point they are movably interlinked. For the torsion elastic spar (h3) not to bend backward too much a string or an elastic thread is apparently tightened between the tips of the spars.
During downstroke of the wings the lifting forces are increased. The spar h3 and the wing are twisting. The magnitude of the twisting acts in accordance with the magnitude of the lift force and the stiffness of the spar. It therefore happens aeroelastically.
Additionally to the twisting the tip of the spar h3 bends upwards in the downstroke. As a reaction it bends downwards at the other side of the cross point - thus, in the section of the arm wing. Thereby, the camber of the airfoil is increased a little. Thereby, an adaptation to the requirements of an effective stroke motion takes place (please look at Piskorsch Adolf (1975), Pressluft-Schwingenflugmodell Chalupsky) and the vidio of the related link 2
2.6 Flying wing ornithopter
Here an ornithopter without a tail unit, developed by Jean-Louis Solignac (France, 2000).
The flapping wing model has a very simple and light driving mechanism and is powered by a rubber motor. With a wing span of 15 cm (5.9 in) it has a weight of only 0.6 grams (0.021 oz [US]). The airplane performances are amazingly good (for the construction of the flapping wing model please look at related link 3.)
The particular about this flapping wing is the down cambered airfoil shaped by battens. Thereby it flies in a stable attitude without a tail unit. This can theoretically be explained with the shifting of the center of pressure of thin airfoils. It can be tested in the adjacent experiment with a paper airplane. The cross-section of this paper airplane equates to a down chambered airfoil.
2.7 In tandem
Ornithopter with two sets of flapping wings based on a Dragonfly, developed by Erich von Holst (1943).
Here, for simplifying the mechanism both opposite halves of a wing are rigidly fixed to a unit. Thus the center of pressure of the model is fixed between the two wing units.
In such tandem arrangements with wings flapping in opposite directions the vertical pendulousness of the fuselage should be avoided. This, however, bears the disadvantage that the backmost flapping wing is in the turbulence wake of the front one. Only for very small wings and at very small Reynolds numbers this may be beneficial. This is a model by Horst Händler (1988).
By mechanisation of a Dragonfly's flight principle.
Erich von Holst has developed his
(1940s) with two in the opposite direction rotating three-blade wings (please
look at E. v. Holst with a video of the Thrust-Wing).
The flapping angle in one stroke direction constitutes 180° or 360° for a complete flapping cycle. Three instead of two wing blades per rotor offer a constant supporting force.
In contrast to a propeller at the Trust-Wing also can be generated a considerable lift force perpendicular to the direction of flight. One must only increase the rate of advance of the Thrust-Wing - similar to a normal flapping wing - and fly with a small positive angle of attack of the Thrust-Wing axis (The rate of advance is the ratio of airspeed to circumferential velocity at the wing tip.).
This is a fine example for an innovative transfer of biological principles
of a flapping wing in engineering. But the specialism
not exist at that time.
The Thrust-Wing model
ENTOID (2007) by Velko T. Velkov has two parallel rubber powered rotors
(please look at the related link 4).
2.9 Oscillating stretched wing
Thrust also can be produced by raising and lowering a stretched wing in flight. But thereto the lift or the normal force during the upward motion must be smaller than during downward motion. The bigger the difference, the better for the thrust (please look at the principle of flight/vector diagram). Furthermore, a continual alignment of the angle of incidence is normally necessary.
Here a strikingly simple generation of an accordant oscillating motion of the wing by using an eccentrically pivoted rotating mass consisting of the mainspring and the gear. In this case the wing is aeroelastic twistable. The idea was coined by W. B. Mituritscha (probably from Russia, 1953).
Unfortunately, a forward and backward motion of the wing occurs along the way. However, this can be avoided by a second counter rotating mass.
There are diverse proposals to generate an oscillation motion of the wing by a pilot who is flying in a hang-glider or another ultralight aircraft - for example by fast press-ups or knee-bends.
Entirely different model experiments with
oscillating wings shows Karl-Heinz-Helling with his
wing airplane (2008)
related link 5.
2.10 Rotating wings
To avoid the accelerating forces in the final stroke positions
flapping wings rotating on a cone-shaped shell where
sometimes built whose apex lies at the wing root.
The backward motion of the wing tip in the range of the lower end position is certainly disadvantageous for the lift.
Rotor Dragonfly (1944 and 1989)
the flight model by Horst Händler (1989). Both ends of the driveshaft are bended in Horst Händler's model. Thereon, the wings are attached freely twistable. The angle of incidence is guided by the upward pointing levers on the wings.
2.11 With non-twistable arm wing section
This is a membrane flapping wing by P. H. Spencer with a non-twistable arm wing section and passive twisting at hand wing section.
The arm wing is triangle shaped and has a large wing depth at the wing root. Arm- and hand wing membrane overlap in wing span direction. Obviously, the hand wing spar could make a little flap movement at the wrist. Later the hand wing depth was enlarged (please look at the construction of the
This flapping wing design of the
was developed by Percival H. Spencer (USA 1958) (please look at related link 6).
3. Profiled flapping wings
Range of application
Profiled flapping wings or double-sided covered wings may work with a very high efficiency. With their mostly relatively low flapping frequency and the small operating range of lift coefficient of a simple airfoil not much thrust can be produced. Not, at least, if the full lift must be generated concurrently (flying with lift). Therefore, profiled flapping wings are suited especially for a level flight, the gently inclined climbing flight and of course for changing to gliding flight.
3.1 With artificial feathers
To ease the twisting, the closed airfoil can be fanning-out. So far, this is particularly used for large manned ornithopters.
Adjacent, a flapping wing with staggered wing tips of the manned
1, developed by Walther Filter (1955, at the Hannover fair 1958).
The angle of incidence deflection of the
feathers designed as several
wings was controllable. (please look at external link 7)
Even for splay and straddle movement of the
there are old design proposals. In contrast, with
only with simple feather implementations experiments have been
A further example for artificial feathers is the
Icarus by Emiel Hartman (England 1959).
A more recent experiment with artificial feathers is to be seen
Birdman Georges Fraisé (France 2005,
please look at external link 8).
3.2 With inclined hinge of the hand wing
A special version of a flapping wing derives from K. Herzog (1963). With this wing, the rotation or the twist axis is not standing vertical to the stroke axis.
The arm wing should perform a flapping motion and a twisting motion at the shoulder joint. With rubber threads between arm- and hand wing the latter was pulled down a little (aeroelastic wing).
This is also an early suggestion for an articulated flapping wing with an additional flap movement of the hand wing.
The kink of the profile between the arm and the hand wing lies approximately at the same location as on the above mentioned membrane wing by P. H. Spencer.
3.3 Twisting by tilting the leading edge of the wing
The feature of the
pitch propeller by John
Drake lies in the twisting of the leading edge, not the
trailing edge of the flapping wing (England, flight tests
3.4 With stepped twisting
An approximate wing twisting can also be achieved by a stepped rotation of relative non-twistable wing sections.
The model EV4 (1979) was also equipped with such a rotation of single wing sections. But in this case, the rotation was controlled by the wing drive.
A typical representative of a passive stepped twisted wing is
Step-Twister with his foam wings (Depron) by Karel
Pustka (2004). The developing gap between the wing sections is
covered with a membrane.
3.5 Twisting by stroke movement of the auxiliary spar
Here, the wing twisting is generated by a phase-delayed stroke movement of the main and auxiliary spar - developed by Emile Räuber (France 1909).
This technology was also used at the EV2 (1976). In the margin, the wings with their two spars powered separately are to be seen.
The function is similar to the wing of a Dragonfly. Here, too, the phase-delayed flapping movement of the main and auxiliary spar determines the amount of the wing twisting.
Furthermore, the Dragonfly obviously works with a strong spar at the leading edge. With the phase-delayed flapping movement of three spars the camber of the airfoil can be influenced, too.
The supports or linkages of the three spars at the body are clearly recognizable as dark, partly crossing structures at the back of the Dragonfly (please look also at related link 9).
In my experience, the twisting method is only suitable for small spans (about 1 meter). Otherwise the adherence to a certain twisting is undetermined by the elasticity, tolerances and friction of the components.
3.6 Servo controlled wing twisting
This is a lifelike and airworthy half-size replica of pterosaurs - a
Quetzalcoatlus Northropi (QN).
The aerodynamics of this ornithopter should equate the original as far as possible. The idea comes from the creative genius Paul MacCready Ph.D. (USA 1985).
The twisting of the wings was controlled by servos and the flight attitude was stabilized by backward and forward motions of the wing tips and nodding motions of the head.
SmartBird by the company Festo the twisting of each
wing side is controlled by a servo (2011). However, in this case
only the twisting of the hand section of the wing is controllable.
The inner arm section is not twistable. Its wing span is 2.0 meter, the weight 0.450 kg.
The wing covering is made of a 2 mm extruded Polyurethane foam plate which is formed to an airfoil (the cross section of the wing used at the model differs from the here shown wing ribs). Upper and lower side of this covering are not glued together at the rear end. So they can slide against each other and the wing is able to do twisting motion without folds (please look at the Shearflex principle in the next section).
On SmartBird it is noticeable that the wing mechanism imitates the bending between arm and hand section of bird wings. The accompanying images show a basic structure of the flapping wing mechanism during up and down stroke. You also can see an animation (2.1MB) of it.
For further details about the SmartBird project please look at related link 11
3.7 Shearflex principle
Here an aeroelastic twistable profiled flapping wing according to the Shearflex Principle. This system makes a relatively inelastic covering applicable. If the twisting along the wing is constant and not to excessive, the airfoil contour accuracy is therefore very good.
Here, the twist elasticity will mainly be determinate by the spar designed
as wing leading edge.
This system was invented by Professor James D. DeLaurier and Jeremy M. Harris (Canada 1994).
The ornithopter with its tripartition of the flapping wing is interesting, too. Jeremy M. Harris 1977 has applied it for patent.
On the adjacent photo James D. DeLaurier and Jeremy M. Harris can be seen with their remote-controlled model, 3 m in span and with combustion motor. A sustained flight was achieved 1991. A video is available (please look at related link 12).
3.8 Oscillating wing tips
The main spar of the
has no hinge, and instead flexes to produce the desired flapping motion.
By this way the wing tips perform an oscillating motion. Thereby the wing twists
passively under aerodynamic loads.
The principle of a wire powered wing oscillation has some marked advantages particularly for man powered ornithopters:
- fail-safe wing position for gliding
- applicable for large wing spans with its accordingly low induced drag
- very few moving parts
Todd Reichert has played an important role in developing of the Snowbird
and he also has flown it successfully. It was the world`s first officially
certifieded short flight of a human-powered ornithopter (Canada 2010).
For more information about the Human-Powered Ornithopter (HPO) Project
please look at related link 13.
3.9 Shell wing
A wing with an active wing twisting by a drive controlled spar rotation, developed by Albert Kempf (France 1998), please take a look at related link 14.
Apparently, the upper side of the wing consists of a cambered hard shell, which is shaped with foam on the lower side to a profiled airfoil wing.
A long thin plate with a cambered cross section
may be twisted easily and wrinkle-free. Also the aforesaid Shearflex principle
is using this property. This flapping wing category here is called
Truefly is to be seen in
the adjacent picture - an ornithopter with a wonderful flying sight. It
also was the first ornithopter which achieved strong climbing flights
with profiled flapping wings.
In the essay
Wing Designs (in German, PDF 1.8MB) can be found additional informations about
these flapping wing constructions.
In conjunction with the EV-models developed
flapping wings are to find on the site:
The wings of pterosaurs have worked differently than today's bird wings, please look at Wing-function of flying saurian
Ideas and suggestions for a further development are enclosed in the paper Lift during wing upstroke, version 10.0, 2015 - 2018 (PDF 1.0 MB).