 |
| Author | Post |
|---|
PatriciaJB
Administrator
| Joined: | Tue May 15th, 2007 |
| Location: | |
| Posts: | 303 |
| Status: |
Offline
|
|
Posted: Thu Mar 27th, 2008 10:27 am |
|
Hello Everyone,
This thread is for a general , wide ranging and ongoing discussion of manned ornithopter design. eg: Wing configuration,design, human and engine powerd, take-off methods etc.
Over to you now, thoughts and comments are invited.
Patricia
Last edited on Mon Mar 31st, 2008 12:59 am by PatriciaJB
|
murray
Member
| Joined: | Thu May 24th, 2007 |
| Location: | |
| Posts: | 163 |
| Status: |
Offline
|
|
Posted: Sun Mar 30th, 2008 09:32 am |
|
For those of us who have used the forum for some time to rant about aspects of ornithopter design, I suppose this thread is intended to elicit a summary of what aspects and principles each of us considers to be important. Here is my two bob''s worth of prejudice, obsession and fevered imaginings:
Operational factors:
- Even in multi-wing configurations, ornithopters will undergo rhythmic linear and angular accelerations under power. Pilot and passengers must adapt to this motion as to a ship at sea or a horse.
- Since the frequency of flapping is likely to be more comparable to a galloping horse than a rolling ship, only while gliding would an ornithopter provide a stable platform for the usual preoccupations of flying, such as scanning the sky ahead, navigation, instrument checks and radio communication.
- Aircraft structures are rated for a “g” loading safety margin. With an engine powered flapping mechanism insensitive to gusts and manoevers, the increased downstroke loading will erode that margin or else require heavier structures than a propeller driven aircraft of the same takeoff weight.
- Engine powered, piloted ornithopters are thus unlikely to be as efficient, controllable or comfortable as propeller driven aircraft for extended flights.
- the rhythmic power demand of flapping flight is ideally matched to muscle power in a rowing action, requiring total involvement of the pilot-athlete in the force, movement, control and feel of every stroke. This “feel” will be required not only to maximise efficiency but also to allow protective “give” of vulnerable structures, joints and linkages with gust or manoeuvring loads. Compared to engine-powered ornithopters, muscle power will require less “g” margin hence a lighter structure (which is just as well !). ,
- Muscle powered ornithopters will be fun but unlikely to achieve flights long enough to be troubled by navigation or discomfort, except perhaps in soaring mode.
- The practicalities of storage and transport to test sites for mockups and prototypes favour folding or demountable wings, especially if already articulated for flight.
- I am attracted to a tailless or bird-like minimal-tail design if it could achieve the required stability and efficiency. Birds seem to manage OK!
Design principles:
- Modern materials such as aligned-fibre composites together with the improved accessibility of aerodynamic theory are the reasons I am encouraged to try things that frustrated far cleverer pioneers of flight.
- an ornithopter must, when the flapping stops, glide stably and controllably to a safe landing.
- gliding flight is much simpler to analyse with conventional aerodynamic theory than flapping and affords a starting point for estimating power requirements, as crucially required in designing for human powered flight. Glide calculations can also be used to model the upstroke and downstroke of flapping flight for rough efficiency estimates.
- induced drag is the primary determinant of minimum power demand and all human powered designs must begin with a strategy for its minimisation, whether by large span, wingtip “feathers”, sweep and dihedral articulation or tricks yet undemonstrated.
- Although parasitic drag has only secondary influence on minimum power at low speeds, there are gains in aerofoil performance and induced drag at higher speeds where parasitic drag rapidly becomes dominant. Every conceptual design should therefore start clean, minimising interference drag and avoiding features such as wires, struts, loose fabric or open cockpits unless subsequently required as a weight-saving compromise.
- Because powered flapping loads are superimposed on the steady supporting lift load, monoplane ornithopters require an energy storage system to absorb the energy released on the upstroke and efficiently return it to assist the downstroke. A dihedral spring is required, loaded to 1g at the mid-stroke position that gives the most efficient stable glide, thus supporting relaxed-muscle glide in the event of cramps, exhaustion or problems with the power linkage.
- The design process has many repetition loops as compromises are worked out between weight, performance, stability, control, structural and ergonomic factors, with cost constraints at every stage on theoretical calculations, models, trial structures, ergometers, mockups and full-size test models.
- Unconventional features such as my minimal-tail aspiration must be reconciled with stability and performance at each iteration.
- Design must start somewhere and my instinct is to avoid detailed drawings until minimum power estimates for the chosen induced-drag reduction principles appear within the pilot’s capability and the structural implementation of required articulations is demonstrated by models, mockups and load tests.
Testing and learning to fly:
- Once a potentially viable design emerges I see the next step being an aerodynamic mockup of the whole wing and stabiliser, articulated and sprung but perhaps initially lacking an active flapping mechanism.
- Tethered glide (or kite) tests should be used to explore stability in all axes, especially if multi-panel articulation or unconventional stabiser mechanisms are involved. Video monitoring of all such experiments should be used to interpret instabilities and possible structural failoures etc.
- The unpiloted mockup could be mounted on a car-roof mounted platform. Alternatively a smooth windy hiil top would serve although this testing environment will be very dependant on suitable winds and probably a source of great frustration.
- To the nose should be attached via a horizontal tether rope to a short upwind pole to overcome drag. Under the intended C of M, the structure is attached to a heavy chain ballast to load the wings progressively rather than with a sudden jerk in the event of a gust or pitch-up excursion. A length of ballast chain equal to the height of the nose tether pole should approximate a pilot’s weight, with excess chain lying on the ground in reserve.
- Once the mockup is demonstrated to glide stably on the tether unpiloted, the flapping mechanism and pilot support is added. I would chicken out of piloted tests on a car-roof platform, preferring the fustration of waiting for suitable winds on the hilltop.
- Flying lessons could then begin in the manner of a fledgling bird. Retaining the nose tether and the ballast chain, now attached via a spacer rope as long as the tether pole height, training would begin with glide attitude control then gradually introduce a gentle flapping action to test its stability effects and to learn coordinated control movements. Flapping amplitude would be progressively increased until it absorbed real effort while maintaining control, something like learning to row a boat. Some instrumentation ( eg. tether tension, anemometer ) would be helpful at this stage to measure achieved and required power output.
- Supposing that we got this far without discovering fatal weakness in the structure or the pilot's bones, with the power estimates still looked good, it would be time to transfer lessons from the mockup to an airworthy prototype, complete with streamlined fuselage.
- Tethered flapping would be continued in the airworthy prototype, with real athletic training for power, timing, feel and control while testing the structure and stability. Hopefully by this time the pilot would be able to make way upwind, keeping the nose tether slack to demonstrate the capability of controlled level flight.
- Next step: free flight over a flat paddock in zero wind, initially low glides from a bungee launch then higher, introducing restrained flapping, finally putting some effort into it to see how far the glide can be extended. Don’t forget to video all these flights; it may be important to understand a near or actual crash.
- After several extended, crash-free flights, all video recorded with distance markers, contact the TV news!
|
DDavis
Guest
| Joined: | |
| Location: | |
| Posts: | |
| Status: |
Offline
|
|
Posted: Sun Mar 30th, 2008 01:33 pm |
|
I’ve been reading through this and other excellent forums on ornithopter development and like to suppose my thinking will contribute. I am not however, a qualified engineer and am relying largely on intuition. Much of what I suggest would need to be tested. This letter is one adapted from others I recently posted in Flapping Wings Forum, with some additional thinking.
Wing clapping… I read of the proposal for a four winged biplane ornithopter, clapping its wings on each side together in the horizontal plane. I don’t see this working very well. In nature it is the upper sides of the wings that meet, in this design it is the upper side of the lower wing against the lower side of the upper. The lower wing may well benefit from the clapping effect, but the airflow around the upper will be going the wrong way and destroying its lift on the upstroke. Added to this is the stress on the wings structure and the increased weight from having to make them more robust.
I’ve seen no detailed information on Vladimir Toporov’s design, but gather it had two wings in dragonfly style beating as a pair of rocking see-saws. This looks good in that it should be easy to get an alternate twisting of the wings going, but if pedal powered I doubt if the pilot could effectively operate them. A two winged design like this might perfectly suit a two man crew using a rowing action, but I don’t believe one person could effectively power it by any method.
I like his basic design, and particularly take note of an observation by another researcher that with the wings being continuous from one side to the other, no spring device is required to hold the wings level without effort from the pilot. It is also likely that a less serious damping device would need to be included, as unexpected gusts are unlikely to have much effect.
I gather that Toporov used pedals to power his design, and here is where I believe he went wrong. Pedals, like crankshafts, do not efficiently transfer power in a direct manner at all points of their travel. Most mechanical arrangements are going to fall short here.
For muscle power, I like the system the Japanese team settled on; their rowing method gives them 100 kilos of effort. Even though intermittent, I suspect this is rather more than any amount of pedalling could provide.
(I don’t like any other aspect of their design. It is a big clumsy monster like the pedal powered craft, and its flappers are merely for propulsion. Far from being an ornithopter, it is just an alternately-powered fixed-winger.)
For the mechanism, I suggest the simplest will also be the lightest and most effective. To each wing root I would have a 30 degree segment of pulley attached, over the arcs of which the cords moved by the rower will stretch. A total wing movement of 15 degrees up and 15 down means that these segments can be contained one in front of the other within a 60 degree triangular area just behind the pilot, and fully within the fuselage. The beauty of this is that all of the pilot’s effort is applied to the wings at all points in their travel. Nothing is lost to indirect transmission.
If using Toporov’s basic design, I suggest the wings be linked such that each of two rowers power both wings through half their full cycle. If the Japanese figures are correct this craft could have a constant thrust of 100 kg’s if the wings can make efficient use of it… more than many an ultralight engine and propeller. My feeling is a redesign of this aircraft with a two person crew using a rowing action might perform surprisingly well.
This suggestion will surely raise some eyebrows… could muscles somehow approach the performance of an engine? I note that fixed wing aircraft do not require an engine matching the weight of the plane in thrust to stay airborne, unless climbing vertically with unreduced speed.
A propeller moves a small volume of air very quickly, whilst an ornithopter wing ‘moves’ a large volume slowly. Fanjet engines get their high power from moving a large mass relatively slowly compared to the exhaust of their core, so maybe there is something in this.
I recognise that a flapping wing does not achieve lift by beating-down on the air, nor does it by merely providing propulsion. In level flight the tip of a bird’s wing is following a sine path that is longer than the path taken by the body of the bird, and this necessarily means that it is travelling faster.
In beating its wings, the bird is accelerating them and generating rather more lift than it would obtain by gliding alone. Some of that is directed forwards for propulsion, and some of it for lift. On the upstroke the wing is mostly soaring, with its attitude adjusted so as to reduce its drag.
On takeoff, a bird will sweep its wings well forward on the downstroke, and almost directly backwards on the upstroke. This gives it a helicopter-like lift alternating with a powerful rearwards fanning action to help it get up to airspeed quickly. (This is similar to the action of many insects, such as flies.)
I note Patricia’s design for producing more complex movements than merely flapping up and down. Birds approach the simple stroke once at cruising speed however, so perhaps it is a refinement that may not be necessary for an artificial craft intended to operate only above a conventional stalling speed. I like the design though, it’s clever.
For my design I am not interested in emulating all of a bird’s abilities. I am happy to launch my craft by winch, and keep things simple. If I have the answers for efficient level flight here, the rest can follow later.
All human powered aircraft built to date have been large, fragile, clumsy and slow. Nevertheless, some are efficient enough as to fly on only 220 watts, well within the reach of any good cyclist. (an athlete can approach 1 KW for brief periods) I feel that something much smaller and more nimble, and being small, inherently less fragile could be the go. About the wingspan of a sailplane, perhaps up to 18 metres.
I choose a pterodactyl-wing, with a narrow streamlined leading edge as a strong but flexible spar, and a cloth membrane stretched behind as with a hang glider. Rather than try to emulate the kinked wing of a pterosaur, it would be flexed in a single sweeping curve from the base to the tips, as it were a fishing rod.
The general plan is for a sailplane-like craft, with high-mounted wings sprouting from just behind the pilots head. As the pilot is going to be the heaviest part of the package, the wings will need to be swept forwards to keep things in balance. Hence the pterodactyl appearance.
A yachtsman observed that the pterodactyl wing worked like a mast and sail, naturally flexing towards the tip under stress in a way that would well suit an ornithopter.
Very large wings such as on the Gossamer Albatross would be hopelessly unwieldy for a muscle powered ornithopter. A flapping wing needs to be structurally strong, and capable of withstanding fatigue. It will be constantly twisting and oscillating, and this places far higher structural demands on it than a fixed wing craft might get away with. It will be difficult to build lightly, and so the smaller the wing area I can get away with, probably the better.
A pterodactyl-style wing tapers to a point and most of its area is inboard. This is going to make flapping the wing easier, as the narrow, increasingly high aspect ratio and fast-moving tips can cut through the air with ease. The large area inboard will provide plenty of lift during both of the strokes simply through passive gliding, carried along by the more active outer areas.
Many differing opinions on the size and weight of the larger pterosaurs abound, but it seems reasonable to work on a wing loading of about 2 lbs per square foot. The highest performance sailplanes are often up on 8 lbs, suggesting that pterosaurs may have found flying easy.
I want an all-up weight with pilot of 100 kilos or less, and so a wing area of only 150 square feet is likely to be enough. With modern materials, I expect it to be easy to build an aircraft this lightly, and still be tough enough to survive a few knocks.
For control, I have in mind a simple butterfly tail. I think an all-moving type will give me a lighter design with simpler controls and less drag in operation than likely with elevators and rudders.
For lateral control, and also to assist in twisting the wings between power and upstrokes, I suggest relying on narrow trim tabs on the outer trailing edges. Being flexible, the wing itself would twist. Being trim tabs, they would operate oppositely to ailerons and also would add almost nothing to the weight.
It is likely I will need to be altering the incidence of the wing for up or down strokes, though I suspect it might not need to be much once at cruising speed. A lightweight spring should be incorporated to take the stress of holding the wings level, without effort from the pilot. This would also serve to absorb the upstrokes and add zest to the downstrokes. A method of locking the wings at a small dihedral should be included for when restfully soaring and for high G manoeuvres.
If this basic arrangement can be made to work, it might be worthwhile developing two other methods of control. The inner part of the wing spar could have spoilers above to assist in lateral control, and help keep it on the ground when landing. Landing flaps mounted not on the trailing edges, but hinging down from this same leading edge spar and close to the centre of lift, as with the Me 163 Komet… could help slow landings and also add drag.
(Very curiously, though this design worked beautifully on the Komet, nobody seems ever to have tried it again. Northrop mucked around for ages trying to get trailing edge flaps working on his flying wings, yet for landing at least, the answer was simple and had already been proved.)
Control is likely to be a problem during an active flapping takeoff or even just low to the ground. Electric servos might be the answer. They could perhaps be controlled by voice, via onboard computer. With today’s electronics, such systems need not add much weight nor even be expensive. It shouldn’t be difficult even to automate the takeoff, so the pilot can devote all of their effort to powering the wings.
Even with the relatively simple wing of a pterosaur, we must recognise that we will not easily match their level of control. Pterosaurs had muscular control over the tension in their fingers, and they had muscle fibres running through the wing membrane allowing fine control of the surface. If the membrane connected to the rear legs as most studies suggest, they could easily vary the tension of their trailing edge. Their brains were large in the areas that provide fine control of muscles and balance. Their pteroid bone probably gave them control of the forward membrane connecting to the neck, allowing its use as a leading edge flap. Most likely, they flew beautifully.
It might be that a smooth wing surface is not the way to go. Golf balls fly better with dimples and many of the Australian aboriginal boomerangs were made with shallow grooves along their wings. These have been shown to have the same effect. Despite much money having been spent on boundary layer control, I am not aware of any attempts to apply this method to aircraft. I don’t suggest dimples as they would make it difficult to keep clean and free of ice, but grooves would be easy to maintain.
It wasn’t engineering that discovered the effect of dimples; it came from the observations of experienced golfers. The boomerangs grooves were likely designed by people just as clever, and perhaps even more so. Play with a stick in a smooth running shallow stream, with the sun behind you. The turbulence created is clearly seen on a smooth sandy bottom. A child playing with a boomerang in a stream may well have discovered the usefulness of grooves… with something nearly as good as a wind tunnel at their disposal.
The future of ornithopters I believe will be as sportscraft. Someday they might even feature in the Olympics, alongside other sports such as rowing.
David Davis tigersnake@vic.australis.com.au
|
Richard
Member
| Joined: | Fri Feb 29th, 2008 |
| Location: | |
| Posts: | 153 |
| Status: |
Offline
|
|
Posted: Mon Mar 31st, 2008 12:41 am |
|
David said: ''It might be that a smooth wing surface is not the way to go. Golf balls fly better with dimples and many of the Australian aboriginal boomerangs were made with shallow grooves along their wings. These have been shown to have the same effect. Despite much money having been spent on boundary layer control, I am not aware of any attempts to apply this method to aircraft. I don’t suggest dimples as they would make it difficult to keep clean and free of ice, but grooves would be easy to maintain.''
Hi David,
I think BL tripping devices are (EDIT--not always necessary at these Re numbers ; depends on airfoil chosen though. END EDIT....playing around with fussy laminar flow foils is also probably a waste of time. BL suction is probably extravagant too as a flapping wing has a natural tendency towards lower profile drag than a fixed wing due to inhibition of the growth of boundary layer thickness caused by flapping. See this paper for similar remark
http://www.ornithopter.org/archive/Lippisch.pdf
Richard
Last edited on Wed Apr 9th, 2008 10:36 pm by Richard
|
murray
Member
| Joined: | Thu May 24th, 2007 |
| Location: | |
| Posts: | 163 |
| Status: |
Offline
|
|
Posted: Mon Mar 31st, 2008 04:14 am |
|
Hi David,
I agree with many of the ideas in your summary but have a few comments:
DDavis wrote:
...
Wing clapping… I read of the proposal for a four winged biplane ornithopter, clapping its wings on each side together in the horizontal plane. I don’t see this working very well. In nature it is the upper sides of the wings that meet, in this design it is the upper side of the lower wing against the lower side of the upper. The lower wing may well benefit from the clapping effect, but the airflow around the upper will be going the wrong way and destroying its lift on the upstroke. Added to this is the stress on the wings structure and the increased weight from having to make them more robust.
Yes the extra wing loading would worry me too.
DDavis wrote:
I gather that Toporov used pedals to power his design, and here is where I believe he went wrong. Pedals, like crankshafts, do not efficiently transfer power in a direct manner at all points of their travel. Most mechanical arrangements are going to fall short here.
I don't know.. like pedal cranks, rocking wings too have a "dead spot" in their motion at turnaround. I don't know what gearing and phase Toporov used but he probably optimised these things.
DDavis wrote:
For muscle power, I like the system the Japanese team settled on; their rowing method gives them 100 kilos of effort. Even though intermittent, I suspect this is rather more than any amount of pedalling could provide.
I don't know what you mean by "100 kilos of effort". Presumably you mean a 100kgf force between hands and feet, but this must be redirected to the wing actuator and may be varied arbitrarily by leverage. The quantity that means something is the energy (= force x distance) in each stroke, as this is conserved (minus frictional losses) through whatever pulley or lever coupling is employed.
DDavis wrote:
For the mechanism, I suggest the simplest will also be the lightest and most effective. To each wing root I would have a 30 degree segment of pulley attached, over the arcs of which the cords moved by the rower will stretch. A total wing movement of 15 degrees up and 15 down means that these segments can be contained one in front of the other within a 60 degree triangular area just behind the pilot, and fully within the fuselage. The beauty of this is that all of the pilot’s effort is applied to the wings at all points in their travel. Nothing is lost to indirect transmission.
I'm not sure how the rowing action is connected to those sectors; sounds like at least a cable and pulleys would be required and these would have some friction losses.
DDavis wrote:
This suggestion (100kg thrust) will surely raise some eyebrows… could muscles somehow approach the performance of an engine? I note that fixed wing aircraft do not require an engine matching the weight of the plane in thrust to stay airborne, unless climbing vertically with unreduced speed.
Once again, force is not the basis on which to compare muscles with engines. I can generate huge forces with a crowbar but it won't help me fly. The real comparison is in power (=energy/time = force x distance/time).
DDavis wrote:
For lateral control, and also to assist in twisting the wings between power and upstrokes, I suggest relying on narrow trim tabs on the outer trailing edges. Being flexible, the wing itself would twist. Being trim tabs, they would operate oppositely to ailerons and also would add almost nothing to the weight.
An interesting idea that would not only reduce structural loads but by opposing the lift-induced camber would tend to stabilise the local angle of attack. As with all reflexed sections however you pay a price in L/D for that stability.
DDavis wrote:
Many differing opinions on the size and weight of the larger pterosaurs abound, but it seems reasonable to work on a wing loading of about 2 lbs per square foot. The highest performance sailplanes are often up on 8 lbs, suggesting that pterosaurs may have found flying easy.
"Easy" flying means low power requirement, for which wing loading is not the only, or even the main determinant. Induced drag is, which is why the sucessful propeller drive HPS's have such large span. You have to use other tricks like winglets or wing-grids if you want to cut back on span.
DDavis wrote:
A lightweight spring should be incorporated to take the stress of holding the wings level, without effort from the pilot. This would also serve to absorb the upstrokes and add zest to the downstrokes. A method of locking the wings at a small dihedral should be included for when restfully soaring and for high G manoeuvres.
I wouldn't lock the wings at all, just rest on the 1g spring for restful soaring. If you want high g manoevers you should apply the load with muscles so as to directly feel the spar and linkages creaking.
DDavis wrote:
Even with the relatively simple wing of a pterosaur, we must recognise that we will not easily match their level of control. Pterosaurs had muscular control over the tension in their fingers, and they had muscle fibres running through the wing membrane allowing fine control of the surface. If the membrane connected to the rear legs as most studies suggest, they could easily vary the tension of their trailing edge. Their brains were large in the areas that provide fine control of muscles and balance. Their pteroid bone probably gave them control of the forward membrane connecting to the neck, allowing its use as a leading edge flap. Most likely, they flew beautifully.
Nice features, but I would like some wind-tunnel assurance that you can get a decent section L/D ( ie >100) with a pterosaur-like "mastwing" section having membrane TE.
DDavis wrote:
It might be that a smooth wing surface is not the way to go. Golf balls fly better with dimples and many of the Australian aboriginal boomerangs were made with shallow grooves along their wings. These have been shown to have the same effect. Despite much money having been spent on boundary layer control, I am not aware of any attempts to apply this method to aircraft.
I think Richard covered this one; dimple or groove turbulators may be useful for delaying flow separation in very specific circumstances at low Reynolds numbers. Otherwise they increase drag. Many aircraft owners get them for free due to hail stones and it does no good at all when used indiscriminately.
DDavis wrote:
The future of ornithopters I believe will be as sportscraft. Someday they might even feature in the Olympics, alongside other sports such as rowing.
Bring it on!
Murray
Last edited on Mon Mar 31st, 2008 04:23 am by murray
|
Jon Howes
Member
| Joined: | Sat Jun 9th, 2007 |
| Location: | |
| Posts: | 176 |
| Status: |
Offline
|
|
Posted: Mon Mar 31st, 2008 07:41 am |
|
Murray,
You responded to David:
A lightweight spring should be incorporated to take the stress of holding the wings level, without effort from the pilot. This would also serve to absorb the upstrokes and add zest to the downstrokes. A method of locking the wings at a small dihedral should be included for when restfully soaring and for high G manoeuvres.
"I wouldn't lock the wings at all, just rest on the 1g spring for restful soaring. If you want high g manoevers you should apply the load with muscles so as to directly feel the spar and linkages creaking."
Don't! A limit on the downward travel of a 1g sprung wing is essential when gliding (or perhaps very active roll control). If you do not have this then any significant reduction in load from 1g will give extreme and divergent roll instability since a sideslip will further unload the into-sideslip direction wing as the aircraft rolls towards the sideslip due to the anhedral, this aerodynamic reaction to this further reduces the g level giving more anhedral etc etc. This is not just theory, both my own models and those of John Mack, using 1g springs have done this. On my model the loss of roll stability was OK as a transient if the down flap limit was kept modest, in the glide condition it was hopeless and anything but relaxing.
My current approach is to allow some anhdral during flapping, but in glide mode a non-return device is used to lock the ings and limit their down travel to about 2 degrees of positive dihedral. Up travel is still allowed so the wings can float between +2 degrees and the upper flap angle limit. 2 degrees corresponds (as close as I can get it) to 1g so some springing it present. Don't even ask about the locking of the articulated bits......
Jon.
Last edited on Mon Mar 31st, 2008 07:42 am by Jon Howes
|
PatriciaJB
Administrator
| Joined: | Tue May 15th, 2007 |
| Location: | |
| Posts: | 303 |
| Status: |
Offline
|
|
Posted: Mon Mar 31st, 2008 08:21 am |
|
The 'Nightingale' Manned Ornithopter Chronicles 
My first manned ornithopter design [Nightingale 1] was loosely based on the wing of one of the largest pterosaurs 'Quetzalcoatlus northropi'. I say 'loosely based' because my thinking, like most of us, is that it is neither possible nor appropriate to try to exactly duplicate the features of a living creature in a man-made machine which is constructed with materials which have completely different characteristics from living tissue. My philosophy is to study all flapping wing creatures to understand their method of flight then to choose those features that not only could be applied after modification to a manned orni but that would also be capable of being actually built with the materials that we have available. This, combined with a study of all types of man-made wings or wing-like structures and their features and methods of construction and of course my experience as the first test pilot of the UTIAS ornithopter and resulting ongoing study of ornithopter flight formed a starting point for my first design.
I chose the pterosaur-like wing because: ... it's high aspect ratio wings would give it 'motor-glider' ability when needed, it was an articulated wing which would allow for greater range of wing kinematics, the pteroid bone gave me the idea for a method of controlling the inboard LE. The attachment of the membrane to the legs [ this varies and is still not agreed on by paleontologists] gave me the idea for TE control. The actinofibrils gave me several ideas and last but not least .... I thought that a pterosaur's wing was a relatively simple design , it would, I thought, be like building an articulated, flapping sail ...... how wrong I was !
The more I talked with pterosaur flight researchers, the more I realised that 'Quetzie's' wing was a highly complex affair which has major differences from a yacht sail. From the calculations of lift, drag, AOA etc through all parts of the stroke and at various wing stations which were given to me by one researcher, I realised 2 things:
[1] High aspect ratio 'motor-glider-like' performance is possible....but
[2] It is only possible if the wing is of highly complex design and the wing kinematics are very finely controlled. ie. the range of motion of the articulations must be precisely controlled to produce the exact amount of twist, flex, sweep, flapping etc required . Since this varies throughout the stroke cycle and also along the span and any variations between the actual wing and the wing in the calculations would lead to a dramatic loss of the calculated performance, I came to the sad conclusion that my first design was not feasible for a preliminary design at this point in time. It could , however, have future potential .
I changed the design to a 4 wing, tandem configuration [ Nightingale 2]
EPISODE 2 in my next post !
Cheers,
Patricia
Last edited on Wed Apr 2nd, 2008 10:32 pm by PatriciaJB
|
DDavis
Guest
| Joined: | |
| Location: | |
| Posts: | |
| Status: |
Offline
|
|
Posted: Mon Mar 31st, 2008 11:16 pm |
|
Thank you everybody for your thoughtful feedback! As I say, I am not an engineer... though I try to be well informed and have worked closely with engineers during a long career in the programming and setting of automatic lathes. Aircraft have been a lifelong fascination and I read up all that I can find on the subject.
I need to correct an error I made when I referred to Toporov's design as having 'two wings.' Somehow its wings being continuous from one side to the other turned each pair into single wings in my head... silly of me, as I am firstly a writer and like to be accurate in my choice of words.
I don't mind taking some risks in suggesting unconventional ideas, as hopefully the result will be in stimulating new thinking and ways of seeing things. So long as I don't appear as an idiot, that is...
I am currently attempting to build a sizable model to test my thinking, and not finding it easy as my only previous experience has been with control-line flying wings over 40 years ago. I am needing to learn quite different methods of construction.
Thank you Patricia for including me in your forum! I was having trouble finding the information I craved and now I have plenty to work through. I had thought it a vaccuum out there, but now I know there are some very good minds at work on the challenge.
(P.S. Patricia, I am still not having any luck with the emails. My own emailer is quite basic with no fancy settings, and I've had no troubles elsewhere. At least it works well with the forums, and for me at least, that is what matters the most. The full message I've been getting says that the address has 'permanent fatal errors' (!) and then the '571 spam source blocked' message. I don't understand it, but then all I really know about computers is how to use a keyboard.)
David Davis. tigersnake@vic.australis.com.au
)
|
PatriciaJB
Administrator
| Joined: | Tue May 15th, 2007 |
| Location: | |
| Posts: | 303 |
| Status: |
Offline
|
|
Posted: Tue Apr 1st, 2008 03:31 am |
|
Hi David,
I've set up 3 new email addresses because of the problem with my email rejecting incoming emails . Quite a few people have reported the same problem .
Any of these addresses should work:
pjb@ornithopter-pilot.com
flapper@ornithopter-pilot.com
pjbflap@gmail.com
Cheers,
Patricia
|
PatriciaJB
Administrator
| Joined: | Tue May 15th, 2007 |
| Location: | |
| Posts: | 303 |
| Status: |
Offline
|
|
Posted: Thu Apr 3rd, 2008 10:56 pm |
|
The Nightingale Manned Ornithopter Chronicles....and some of my thoughts on manned ornithopter design in general.
EPISODE 2
Engine-Powered vs Human-Powered.
Engine - powered ornithopters will, I think, successfully sustain flight before human-powered orni's because of the difficulty in providing sufficient human-power and in efficiently transmitting human-power to the wings and because the design of the wings themselves is so critical . All aspects of human-powered aircraft design are critical. The first sustained flight by a human-powered orni will be a far greater achievement than the first engine-powered orni to do so.
For these reasons and because my experience has been with engine-powered aircraft and orni's . Nightingale is an engine-powered orni. I think there is greater scope for future development than with human-power, though human-powered orni's, as I said, will be the greater technical achievement at first.
Well, that's it for episode 2 !
Episode 3 in next post....it'll be longer than this episode !!
Cheers,
Patricia
Last edited on Fri Apr 4th, 2008 02:41 pm by PatriciaJB
|
murray
Member
| Joined: | Thu May 24th, 2007 |
| Location: | |
| Posts: | 163 |
| Status: |
Offline
|
|
Posted: Sun Apr 6th, 2008 07:17 am |
|
Jon Howes wrote: Murray wrote:
"I wouldn't lock the wings at all, just rest on the 1g spring for restful soaring. If you want high g manoevers you should apply the load with muscles so as to directly feel the spar and linkages creaking."
Don't! A limit on the downward travel of a 1g sprung wing is essential when gliding (or perhaps very active roll control). ....
I see what you mean; good advice. I would have got around to thinking about the "tuck under" roll instability some day but my plodding preoccupation has been with structural strength in the downstroke and >1g situation. I will have to think further about how to make a minimum dihedral limit for gliding while allowing downstroke-override.
Murray.
|
murray
Member
| Joined: | Thu May 24th, 2007 |
| Location: | |
| Posts: | 163 |
| Status: |
Offline
|
|
Posted: Sun Apr 6th, 2008 07:39 am |
|
PatriciaJB wrote: The more I talked with pterosaur flight researchers, the more I realised that 'Quetzie's' wing was a highly complex affair which has major differences from a yacht sail. From the calculations of lift, drag, AOA etc through all parts of the stroke and at various wing stations which were given to me by one researcher, I realised 2 things:
[1] High aspect ratio 'motor-glider-like' performance is possible....but
[2] It is only possible if the wing is of highly complex design and the wing kinematics are very finely controlled. ie. the range of motion of the articulations must be precisely controlled to produce the exact amount of twist, flex, sweep, flapping etc required . Since this varies throughout the stroke cycle and also along the span and any variations between the actual wing and the wing in the calculations would lead to a dramatic loss of the calculated performance, I came to the sad conclusion that my first design was not feasible for a preliminary design at this point in time. It could , however, have future potential .
Patricia,
Control of twist and flex (ie camber?) etc is one thing but at each span station the local aerofoil section has to be good. Do you have a reference for those "exact" aerofoil sections reconstructed for QN?
Jim Cunningham posted pictures long ago of some massive fossil bones to prove that the QN wing did not have a thin sail-like membrane, but I have not seen a postulated aerofoil that we could run through XFOIL. It would have been grossly too heavy to have a double surfaced section filled with fat or muscle aft of the bony ( and possibly hairy) leading parts, so I assume it must have had a thin bat-like membrane, looking something like a sailwing but cambered. I have faked up such aerofoils adding a thin curved extension from the TE of a good double surfaced aerofoil, but the simulated L/D was not very encouraging.
Murray.
Last edited on Sun Apr 6th, 2008 07:40 am by murray
|
Richard
Member
| Joined: | Fri Feb 29th, 2008 |
| Location: | |
| Posts: | 153 |
| Status: |
Offline
|
|
Posted: Sun Apr 6th, 2008 07:55 am |
|
Jon said in Georges avatar thread -----The section on the first of the fifth generation gliders, the UP comet, was an approximate UI1720. This section is still used on most of the current rash of "topless" gliders. I have some test data but unfortunately not with realistic hang glider construction.
Murray,
I honestly hesitate to post because I fear I am missing the point but many or most hanglider double surface aerofoils are created by having two independent sails- one being the top surface sail and one the bottom surface sail. This allows such features as in-flight sail tension adjustment for different airspeeds.
Last edited on Sun Apr 6th, 2008 08:02 am by Richard
|
PatriciaJB
Administrator
| Joined: | Tue May 15th, 2007 |
| Location: | |
| Posts: | 303 |
| Status: |
Offline
|
|
Posted: Sun Apr 6th, 2008 09:41 am |
|
murray wrote: PatriciaJB wrote: The more I talked with pterosaur flight researchers, the more I realised that 'Quetzie's' wing was a highly complex affair which has major differences from a yacht sail. From the calculations of lift, drag, AOA etc through all parts of the stroke and at various wing stations which were given to me by one researcher, I realised 2 things:
[1] High aspect ratio 'motor-glider-like' performance is possible....but
[2] It is only possible if the wing is of highly complex design and the wing kinematics are very finely controlled. ie. the range of motion of the articulations must be precisely controlled to produce the exact amount of twist, flex, sweep, flapping etc required . Since this varies throughout the stroke cycle and also along the span and any variations between the actual wing and the wing in the calculations would lead to a dramatic loss of the calculated performance, I came to the sad conclusion that my first design was not feasible for a preliminary design at this point in time. It could , however, have future potential .
Patricia,
Control of twist and flex (ie camber?) etc is one thing but at each span station the local aerofoil section has to be good. Do you have a reference for those "exact" aerofoil sections reconstructed for QN?
Jim Cunningham posted pictures long ago of some massive fossil bones to prove that the QN wing did not have a thin sail-like membrane, but I have not seen a postulated aerofoil that we could run through XFOIL. It would have been grossly too heavy to have a double surfaced section filled with fat or muscle aft of the bony ( and possibly hairy) leading parts, so I assume it must have had a thin bat-like membrane, looking something like a sailwing but cambered. I have faked up such aerofoils adding a thin curved extension from the TE of a good double surfaced aerofoil, but the simulated L/D was not very encouraging.
Murray.
Hi Murray,
It was Jim Cunningham who did the calculations using his own program so I'm not at liberty to say much about it. The wing in the calculations was heavily based on Queztie's wing kinematics. [Jim has studied Quetz flight dynamics for many years and was also working on a flying replica at the time.]
My first wing design [wing1] was only loosely based on Quetz' and wasn't able to match the fine control needed for that type of wing and it was only capable of testing that one particular set of stroke kinematics. Since Nightingale's purpose is to test many wings and variations of wing kinematics , I decided a complete change of design was necessary.
Re: pterosaur wing membrane:
Jim posts on the 'Dinosaur mailing list ' - housed at the Cleveland Museum Of Natural History. You can find many of his posts on pterosaur wings and flight there.
Here's the link to their archive page, there's a search box at the top of the screen.
http://dml.cmnh.org/index.html
Cheers,
Patricia
Last edited on Sun Apr 6th, 2008 03:46 pm by PatriciaJB
|
PatriciaJB
Administrator
| Joined: | Tue May 15th, 2007 |
| Location: | |
| Posts: | 303 |
| Status: |
Offline
|
|
Posted: Sun Apr 6th, 2008 03:53 pm |
|
Hello Everyone,
Since this thread is the very essence of what this forum is all about, I've given it it's own section on the home page so that it's always visible there in the 'most recent post' list and can be easily and quickly located.
Cheers,
Patricia
|
Current time is 03:54 am | Page: 1 2 3 4 5   |
|
|
|
|
