Quantcast

Freewing construction

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
Hi,
I have recently discovered this forum, having spent some considerable time on the Rotary Wing Forum very profitably. However, since I am designing/building a gyro/fixed wing hybrid aircraft, the rotary guys were only able to help for the particularly gyro-aspects of the project.

I am currently trying to figure out the best way to build wings, and it seems that my thinking is very similar to ideas expressed in recent threads dealing with spar-less wing design.

I plan to have my wings machined in a local CNC shop out of polystyreme. Since the wings are not rectangular, but gently curved with upturned wingtips, this seemed to me the only way to achieve the required precision. However, since having this sort of work done isn't cheap (and I may want additional sets of wings further down the track), I thought that the wisest way forward would be to:

(1) Use the basic wing plugs to pull female molds from (top and bottom skins)
(2) Sacrifice the foam wings by first slicing out the "spar", and then by slicing out the ribs.
(3) Armed with my top and bottom skin molds, all I now need to do is to use the cut out parts as molds for both spar and ribs. The spar would be made according to the Marske method of Graphlite (pultruded carbon fibre rods) spar caps laid in a female mold and vacuum bagged into place with a layer of fibreglass. This would give me a "U" shaped spar. Place two back-to-back for a very light and extremely strong "I" beam spar which should fit between the top and bottom skins perfectly. Similarly with the ribs.

This should give me sufficient molds to create as many wings as I wish should I screw up the first one..

I might also add that I'm going to be employing the "freewing" design, so there is little (if any) torsional stresses on the wing in flight. And the free pivoting wing also limits possible G-forces to about 1.5G (this is what I have read - and can't currently confirm this).

However, some of the many imponderables at the moment consist of things like:
What wing area will I require?
If I wish to cruise at X knots, what part will wing geometry play in this?

I recently came across a very poorly photocopied article on the freewing (very little technical information is available, which doesn't make life easier...). It indicates that by moving the pivot point of the wing forwards or backwards, one can trim it to fly at a given airspeed. This seems a bit strange to me. Can anyone shed any light on this? I've attached the grainy photocopy FYI.

Regards,
Duncan
 

Attachments

Last edited:

Peter V

Well-Known Member
Joined
May 24, 2006
Messages
139
But how do you propose to move the pivot point? That diagram is intended to show where the best position is for your pivot point in relation to your maximum airspeed (plus you'll need some torsional resistance on the pivot to increase incidence and low speed performance).
A freewing is very sensitive to CG, so trim it by moving your seat forward or back ;) :gig:
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
But how do you propose to move the pivot point? That diagram is intended to show where the best position is for your pivot point in relation to your maximum airspeed (plus you'll need some torsional resistance on the pivot to increase incidence and low speed performance).
A freewing is very sensitive to CG, so trim it by moving your seat forward or back ;) :gig:
Hi,
Thanks for the feedback. I wasn't thinking of moving the pivot point - I was asking how to work out where to put it in the first place. Once it has been set, it is going to be there for good or bad, and I just wanted some input on how better to understand the principle.

My understanding is that if you want to design your aircraft to fly at a trim speed of, say, 100kts, then you would put the pivot point at position X. If you wished to fly at a higher trimmed speed, for the same power, you would set the pivot point some distance forward of position X. My main concern, if this is correct, is how to work out position X to begin with.

The grainy photocopy seems to indicate some scaled measurement of where to place the pivot point - but the derivation of the scale is unknown to me.

Regards,
Duncan
 

Peter V

Well-Known Member
Joined
May 24, 2006
Messages
139
Unless you're making a Spratt 107, that data won't be any help to you.
You'll be needing the original NASA studies:
Fun reading for engineers. Will require 1 beer per page to make sense to normal people.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
Hi,
Thank you very much for the references. I'll sit down this evening (suitably fortified with beer) and give them a go.

Question: Why would the data be applicable only if I were building a Spratt 107? I've ordered plans for the Spratt 103 - just so I can see how he manages the wing control mechanisms. I presume this aircraft will obey the same principles as the 107?

Essentially, any free-wing aircraft will behave similarly, surely? But I'll read your referenced reports and see what emerges from the woodwork.

Regards,
Duncan
 

Peter V

Well-Known Member
Joined
May 24, 2006
Messages
139
That article was based on a study of the Spratt 107 and is particular to its 23112 airfoil. As long as the 103 has the same wing and hinge point, the article may apply. :nervous:
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
Hi,
Excellent. Yes, my airfoil is also the NACA23112, so that is good news. The reason for the airfoil choice is that it has zero (or at least very low) pitching moment.

Thanks for your help. I've printed out the second of the two reports (112 pages!) using the friendly office laser printer... So I'll be giving this a go tonight. I might come back to you with a bunch of questions - I hope you're up for it!

Cheers,
Duncan
 

Peter V

Well-Known Member
Joined
May 24, 2006
Messages
139
Well, I'm no expert on freewings, just someone with an interest. Unfortunately, there arn't too many people with any expertise in the area. I'd talk to Bernard Geffray (spratt103@gmail.com) he designed the Spratt 103 and with 5 prototypes under his belt, he should be of some help, espcially if you know some French. :ponder: Friendly guy, and as you probably bought those plans from him, he'd be obliged to help. :grin:
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
Hi,
Well, I spent some time reading the NASA report you pointed me to, and it wasn't all that tough, to be honest. Of course, I had to skip the mad mathematical doodlings. And I had to just gloss over those paragraphs which *looked* like they were in English, but were obviously written by someone from a parallel universe.

I found it extremely informative, and it has completely changed the way I'm thinking about free-wings. For those reading this post without prior knowledge of the free wing design, here is a quick summary:

Background:
A free-wing aircraft differs from a conventional airplane in that the wings are not fixed in angle of attack relative to the fuselage. They are attached to the fuselage on bearings, enabling them to rotate freely along the span-wise axis. In flight, the pitch of the wings is determined solely by the direction of the local airflow. Hundreds of aircraft employing the free-wing design have been built, and fly extremely well. Their main advantage over conventional aircraft is that they exhibit over 60% improvement in gust alleviation. Why? Because in a gust, the wings pivot, not the fuselage.

The report
Essentially, the report found the following:
  1. There are two basic design options for freely pivoting wings: (1) both wings pivot in tandem (2) both wings pivot independently of each other.
  2. When both wings pivot in tandem (ie L and R wings are coupled in pitch) the aircraft displays significant gust-alleviating properties (over 60% improvement over fixed-wing aircraft of identical config)
  3. When the wings are allowed to pivot independently, two serious issues arise.
    1. Severe rate of spiral divergance, with the time taken to double the amplitude being less than .81 seconds. By comparison, the miniimum recommended time to double spiral amplitude is 12 seconds (ie 15x greater) in the approach condition, and 20 seconds (ie almost 25x greater) in the cruise condition. (With differential freedom removed, the time to double the amplitude in either mode rises to 163 seconds - well above the accepted minimum.)
    2. Severely reduced roll damping, making the aircraft almost non-flyable. (With differential freedom removed, roll damping is identical to a standard fixed-wing aircraft.)
As can be seen from the above, it would be foolhardy to design independently free L and R wings, but eminently sensible to couple the L and R wings.

What are the benefits of the coupled free-wing design?
  1. First (and most basically) it allows a designer (ie me) to forget about optimum wing AoA relative to the aircraft. Just mount the wings, and allow aerodynamic forces to deal with the AoA issue.
  2. The most obvious benefit is the gust-alleviation properties of the freely pivoting wing. This would allow all the benefits of high wing loading, but all the comfort of much larger wings. And in transport-based flights (ie long cross-country flights) personal comfort is important.
Limitations of the NASA report
Of necessity, the report dealt with a narrowly defined set of design criterea. Namely: NACA23012 airfoil, use of ailerons to influence wing pitch relative to airflow.

Choice of airfoil
The NACA23012 airfoil was chosen because (1) it has a relatively high max lift coefficient (2) very low (almost zero) pitching moment, since it is a slightly reflexed airfoil and (3) lots of experimental data relating to this airfoil exists already. The very low pitching moment is important, because with increased speed, the tendency is for a wing to pitch downwards, increasing speed, which in turn feeds back into the situation which rapidly becomes divergant. With fixed-wing designs this can easily be counterbalanced, but with a free wing, control authority is quickly lost. A number of early Pou de Ciel (Flying Flea's) were lost when their pilots could not pull them out of dives. The solution is to use a zero pitch moment airfoil.

Use of ailerons
In my opinion, this is the Achilles heel of the design as described and discussed in the report. While it acts perfectly well in the coupled-wing scenario, it is inadequate and potentially fatal in the independent wing scenario.

If one examines the Spratt 103, for example, one discovers independently pivoting L and R wings, but no ailerons. Each wing is coupled directly to the control system (joystick). In a left turn, for example, the pilot moves the stick to the left, pitching the R wing up and the L wing down, effecting differential lift, and hence a turn. However, while the wings are free to rotate independently, they are nevertheless linked back to the controls, where the pilot has final authority.

The "direct-control" freewing concept
Flying a "direct-control free-wing" is an interesting experience for the pilot. The stick does not sit passively in his hand. Rather, he can feel it moving, back and forth, left and right as the wing movements feed back into his hand. Rather like the feedback one gets from riding a horse. As the horse puts his head down and increases pace, so the stick tugs forward as the airplane increases speed. Similar feedback is available to the pilot in L and R turns. At no time are the wings able to run away with themselves in a divergent spiral.

One limitation of the direct-control design is the fact that L and R wings, while able to pivot freely, are nevertheless coupled to each via the controls. The L wing can't move independently of the R, and vice versa. This will reduce the gust alleviation properties of the design slightly in cases where a gust strikes one wing only. But it is a workable compromise.

One final observation regarding the use of ailerons

Coupling the wings directly via a mechanical link to the controls gets rid of the need for ailerons altogether, since the entire wing acts as a great big aileron if the pilot pulls back on the stick. This needs to be carefully geared so that the stick pressure (and throw) is adequate for both slow and high speed flight. But doing away with ailerons has practical advantages which are quite compelling.
  • One can build a perfectly laminar flow wing, since there will be no breaks in the wing surface for hinges etc.
  • Cheaper, simpler and quicker to build a wing with no ailerons
  • An aileron-less wing looks extremely sexy :gig:
So there you have it. I hope (if you're interested in the free-wing design concept) you have found this post interesting. I certainly plan to build a direct-control freewing for my airplane, and when I have something worth looking at, I'll be taking some photos and posting them here on the forum.

Kind regards,
Duncan
 
Last edited:

Topaz

Super Moderator
Staff member
Log Member
Joined
Jul 29, 2005
Messages
14,178
Location
Orange County, California
Nice! Thanks for the summary. It's a neat concept.

How does one trim for different airspeeds? Some sort of trim flap on the wing?
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
Nice! Thanks for the summary. It's a neat concept.

How does one trim for different airspeeds? Some sort of trim flap on the wing?
Hi,
To be honest, I don't know. I'm still thinking that one through. One possibility is to move the pivot point forwards/backwards. Not sure if this is possible, but it might be. If so, the problem is neatly solved with still no aileron interference or trim tab either, for that matter.

Duncan
 

Peter V

Well-Known Member
Joined
May 24, 2006
Messages
139
Thanks for that report Duncan. Nice summary :)

I've given a bit of thought to trimming a freewing ultralight. I was actually serious about moving the pilot seat back and forth :para::ponder: The pilot is by far the heaviest part of the aircraft, so employing a skulls boat sliding seat controlled by a worm gear or such, would likely be sufficient. I was also pondering moving fuel around like a modern yacht employs water balast, but that would be tricky as it's an ever diminishing amount of fluid.
Moving the pivot of the wing back and forth is a bit scary. Problem of how to build a mechanism that you'd stake your life on as it's holding the wing to the aircraft.

Oh, and one type of free-wing not explored is where only a partial amount of the wing is free, maybe 30%, with the remainder being fixed. Probably a safer compromise for a larger light aircraft - I'd be nervous about large heavy wings flapping freely, after seeing a few manned ornithopter attempts :speechles
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
Hi guys,
One of the problems about contemplating the "path less trodden" is the fact that there are few, if any, roadmaps available.

Peter, your thoughts about moving the pilot's seat back and forth is a way to move the CG, but not, I think, what the note I found on the Spratt 103 web site is suggesting. He seems to suggest that one needs to move the pivot point of the wing forward/backward - the point being that as speed increases/decreases the centre of lift likewise moves.

One of the things I'm trying to avoid if at all possible, is the use of any extra control serfices, like trim tabs or ailerons. Or as Mignet himself quaintly observes: "Ailerons - what a sad invention".

There seem to be two (and now that you mention it, two and a variation...) configurations:
  1. Linked wings which swivel in tandem, and a strong rudder to effect all turns (Flying Flea)
  2. Independent wings which swivel independently, and no rudder at all (Spratt designs - eg: Spratt103)
The Mignet design is simple and escapes the spiral divergance and lack of roll damping inherent in the 2nd design. The Spratt design affords almost zero yaw in turns yet very effective turn control - but is beset by spiral instability.

From where I'm sitting, it seems that the basic Mignet design (linked L and R wings + use of rudder to turn) is a "better" option. But it's a bit of a toss-up. Neither use ailerons, of course, but I notice also that the last Mignet design (HM 1100 CORDOUAN) *does* use ailerons (mainly for cross-wind landings). and also a trim tab set into the rear of the mid section of the wing.

Mmmm... early days yet, but the HM1100 may just be the better design choice. Don't know yet - still pondering the issues.

Regards,
Duncan
 
Last edited:

Peter V

Well-Known Member
Joined
May 24, 2006
Messages
139
But, if you do resolve to move the wings fore and aft during flight, what about the struts?
I think the 103 has been designed around a very restricted performance envelope. The engine might be your only form of trim.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
Hi,
Yes, I think you're right about the 103. It isn't the solution for any serious "travelling".

Moving the povot point of the wings won't involve moving the struts. They need to be fixed firmly into the fuselage. Rather, you'd have to mount the bearing on which the wings actually pivot on some sort of sliding mechanism. Shouldn't be that difficult. I can envisage the same sort of sliding mechanism you were advocating for moving the pilot's seat - except that it would be fixed to a beefed up spar. This would allow the wing to move back and forth a little but have the struts firmly fixed.

On the other hand, you could simply mount a tab on the trailing edge, the way the HM1100 does. This would effectively influence the wing's AoA.

What concerns me a little is the use of rudder alone for turns. Great when the air is still, but handling a crosswind landing could be interesting.

Regards,
Duncan
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
Hi,
This may be a silly question, but I certainly don't seem able to find an answer for it.

Question:
With a freewing, how does one calculate the CL of the wing? Put another way: At what AoA will the wing present itself to the oncoming airflow (eg: NACA23012 airfoil). Is this AoA dependent on the airfoil used?

Regards,
Duncan
 

Topaz

Super Moderator
Staff member
Log Member
Joined
Jul 29, 2005
Messages
14,178
Location
Orange County, California
Well, you can get the CL you need simply without knowing anything about the airfoil involved. The classic lift formula:

L = q*Sw*CL

... doesn't have any terms that relate to the airfoil, nor the configuration.

The second half of your question is, indeed, airfoil-dependant. The airfoil will generate a certain moment around the aerodynamic center under a given set of conditions, and if the countering moment from the suspension point of your fuselage is the same, the system will be in balance.

As I understand the Freewing concept (and that's fairly limited, so you'll want to check me on this), the design flies at one particular airspeed (actually, one particular 'q'), so once you've picked that airspeed, you figure out where to put the suspension point so that the weight of the aircraft balances the pitching moment of the wing under those conditions, when it's at the AoA needed for the right CL.

[Edit] Yikes, that was a grungy explanation... :depressed

Okay, let's try this:

Size your wing for the airplane so that it's flying at the airfoil's design CL at your chosen flight condition.

From that, you can calculate the angle of attack the wing needs to be at for that condition, and from that, the pitching moment the wing will develop.

Then, from the weight of the aircraft and the pitching moment the wing is developing, you can figure out where (chordwise) to place the suspension point.

Hopefully there was something like the information you were looking for in one of those! ;)

BTW - looking at the images you posted in another thread, did you ditch the rotor on this aircraft? I thought I read originally that it was going to be a gyrocopter with wings added to offload the rotor in cruise. If the rotor is gone, you'll need to chose your wing size and airfoil to some appropriate 'regular' flight speed that is fairly slow, since you'll be landing at that speed, too, without some kind of trim flap to alter the CM characteristics of the wing. This ought to be driving your airfoil choice - low drag at high Cl - but the limited number of 'good' reflexed airfoils out there may limit it even more, and be the ultimate deciding factor.
 
Last edited:

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
Hi,
Very nice answer. Thank you. I'll digest it at my leisure later this evening.

To answer your question re: gyro or Fixed wing. The simple answer is: both. Fitted with an appropriately sized wing, there is no reason why it can't be a FW aircraft. Remove the wing, fit the gyro bits (ie prerotator, mast, head and blades) and you have a gyro. Add to that the smaller wings, and you have a compound. Doesn't really matter - only that one has to size the various bits appropriately. Here's what the gyro versions look like:

Regards,
Duncan
 

Attachments

Topaz

Super Moderator
Staff member
Log Member
Joined
Jul 29, 2005
Messages
14,178
Location
Orange County, California
Wow. That'll be quite the aircraft. Heck of an engineering challenge, but as a compound you could size the wing for some nice cruise speed, rather than landing.

On the low-wing version, how would you keep the fuselage from swinging around upside-down under the wing after you off-load the rotor?
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,234
Location
Brisbane, Australia
G'day mate,
Yes, quite a challenge, but not, I think, *that* difficult. The one element which ties it all together is a bit of an accident, really. And that is the fact that the aircraft is built round a "strongback" - a carbon fibre "C" (see sketch below). This effectively provides a cradle for the pilot, as well as a very convenient place to attach a high wing and/or a rotor as well as a convenient place to attach a low wing and/or landing gear. In an of itself it is nothing special, but the trick is to have the top and bottom portions of the "C" protrude from the fuselage.

Suddenly the wings are not actually a fixed part of the fuselage, and can easily be removed and replaced with other items. All I have to do is to make sure each configuration fits in its own right. Then it is just mix and match.

As far as your question goes, at first I didn't understand what you were getting at. Then it occurred to me that the grand design might, in fact, be flawed. If I understand your question correctly, you are asking what is going to keep the aircraft oriented correctly while balancing on a tiny pivot point? What will keep it on top of the wing?

The answer is, I think, very simple. Consider the case of the high wing version: the plane hangs under the wing, but if it could be suspended in space directly from the centre of lift, it would, in fact, topple forward quite smartly - because the CG is about 5 or 6 inches ahead of the centre of lift. Yet, in flight, it doesn't. And the reason for this is because of the horisontal stab. Set at a slightly negative AoA it presses the tail down, just enough to keep the aircraft flying S&L. In fact, the plane doesn't even begin to nose over.

Exactly the same happens in the low wing version. The same H-stab works in the same way, keeping the tail depressed.

Although, having said this, I'm a little wary of having the weight of the plane balancing on the wing... Just as well I prefer the high wing version...

Having decided I now have a semi-decent reason to go with the high-wing, I now have to consider the possibility of a blade strike when configured like a gyro. If my recollection is correct, the blades can move left/right by 9 degrees each way, and flapping adds another 12 degrees or so. That's 21 degrees. Add another 9 degrees for safety, and we end up with 30 degrees required to keep the blades free of the wings below them. Which leaves room for a relatively small wing. At 3 deg AoA, 15ft NACA747a315 airfoil wings will produce just enough lift for a 120kt cruise with the Cd right in the middle of the drag bucket. Total drag only 20lbs. At 2 deg AoA, the cruise should be in the region of 135kts. Well, that's how I read Profili, anyway... :ermm:

But these are 15ft wings - 7.5 ft either side of the aircraft. A bit too long for comfort. So the proposal is to curve the wings downwards. Should look quite cool. I'm a little worried about the negative dihedral, but the pendulum effect of the high wings should counter that...

What do you think? Sort of like a arched over Raptor look...

Duncan
 

Attachments

Last edited:
Top