Solving forward sweep problems by using tapered thickness

Help Support HomeBuiltAirplanes.com:

Doggzilla

Banned
HBA Supporter
Was just thinking about forward sweet wings and something occurred to me.

If the CG of the aircraft is anywhere near the center of the wing, making the inboard portion of the wing thicker would make the wing vastly more statically stable.

Since the portion behind the center is the inboard section of forward swept wings, making the inboard wing thicker has the effect of more lift/authority from the rearward section at a given AoA.

Therefore, tapering the thickness should make a forward swept wing statically stable instead of unstable and prone to divergence, no?

Dana

Super Moderator
Staff member
A thicker airfoil won't necessarily produce more lift at a given AOA, camber more significant. Moving the wing center of lift forward also won't necessarily make the aircraft more stable than moving the CG forward. Also consider that a thinner section at the tips means the tips may stall first.

Doggzilla

Banned
HBA Supporter
For forward sweep that is a good thing.

Forward swept wings suffer from divergence which makes them unstable and prone to backwards tumbles. Tips stalling before the rear/center of the airfoil means it would not suffer divergence at high AoA near stall.

Center of lift would shift backwards during stall instead of forward.

wsimpso1

Super Moderator
Staff member
Log Member
Was just thinking about forward sweet wings and something occurred to me.

If the CG of the aircraft is anywhere near the center of the wing, making the inboard portion of the wing thicker would make the wing vastly more statically stable.

Since the portion behind the center is the inboard section of forward swept wings, making the inboard wing thicker has the effect of more lift/authority from the rearward section at a given AoA.

Therefore, tapering the thickness should make a forward swept wing statically stable instead of unstable and prone to divergence, no?
What "problems" with forward swept wings and stability are you attempting to solve?

The lift slope of ALL foils (in 2D measurement)is about 0.1096/degree. Thickening or thinning the airfoil does not change this. Going to 3D airfoils, where we have tips and leakage around the tips, the entire wing's lift slope is reduced in relation to the aspect ratio of the wing. An aft swept wing tends to act like it has a little less AR than one with its 1/4 chord line perpendicular to flight direction. I suspect that a forward swept wing will act like it has a little more AR, but in neither case will that have much influence upon stability.

If you have a problem with stability in pitch, we need to address issues of where the wing and other surfaces are in setting the neutral point. If the aftmost CG is not adequately forward of the neutral point, you can either move the aftmost CG forward or shift the neutral point aft. The two prominent ways to shift the NP aft are to select more tail volume or shift the wing aft. This applies to the forward swept wing just as well to any other wing. Shift it aft for more stability in pitch. The center of lift and pitching moment for a forward swept wing is found the same way it is found for all other wings...

Billski

Jay Kempf

Curmudgeon in Training (CIT)
Interesting idea. Front load an airfoil with lots of camber at the tips. Farther back at the root. Don't think that really works in the end but interesting thought. Center of area will probably drive the overall dynamics.

BBerson

Light Plane Philosopher
HBA Supporter
The tip chord can be made smaller than the root chord with ordinary planform taper. This would make those forward swept tips stall slightly sooner than a constant chord. Not that a few degrees sweep are any problem.

Topaz

Super Moderator
Staff member
Log Member
...Forward swept wings suffer from divergence which makes them unstable and prone to backwards tumbles.
Presuming the wings don't rip off the aircraft - the most-likely outcome - what you're describing is an aeroelastic effect, and not stability effect per se. The stability of the aircraft is indifferent to whether the wing is swept forward, swept aft, or not swept at all. It's very useful - and important - to keep the two concepts separate. Making the root section thicker won't really change the lift curve slope of the wing in that spanwise region, meaning the effect you're hoping for won't happen. Billski is setting you straight on this - calculate your design's stability strictly through the positions of the neutral point and the center of gravity, which is almost entirely dependent on planform. If you also have aeroelastic effects driving unwanted pitching moments, those need to be analyzed and dealt with separately.

Tips stalling before the rear/center of the airfoil means it would not suffer divergence at high AoA near stall.
Strictly true, but then you'd also lose virtually all hope of roll-axis control as the stall progresses, and the likelihood of asymmetric tip stall in the "real world" means the airplane will drop a tip in most stall scenarios. These are exactly the reasons we want to avoid "tip stall" and have the wing stall from the root outwards.

Forward-swept-wing divergence issues are best dealt with by tailoring the structure of the wing to not diverge. That's usually done by adjusting the layup schedule of a composite wing so that the flexural center of the tip section is ahead of the quarter-chord point of the airfoil. Under loading, the tips then will either maintain a constant angle of incidence, or decrease their incidence slightly under load.

Dana

Super Moderator
Staff member
Strictly true, but then you'd also lose virtually all hope of roll-axis control as the stall progresses, and the likelihood of asymmetric tip stall in the "real world" means the airplane will drop a tip in most stall scenarios. These are exactly the reasons we want to avoid "tip stall" and have the wing stall from the root outwards.
This.

cheapracer

Well-Known Member
Log Member
I actually like Zenith's solution on the CH650 which is opposite, the spar is straight (constant height), and the wing tapers in on the rear edge, gradually increasing the wing's thickness percentage to the tip.

Doggzilla

Banned
HBA Supporter
If I understand correctly the idea is to make the area behind the spar larger so that when it’s under heavy lift it bends nose down instead of nose up. The X-29 had something similar.

I was considering having elevators on the outboard section to provide canard like control ahead of the CG, and having ailerons mid wing closer to mid chord so they are aligned with CG and have a better yaw moment. Probably a tail elevator on the inner trailing edge that works opposite if needed.

The outboard elevators would produce a force on the trailing edge, helping to counteract aero elasticity bending the leading edge.

Outboard split flaps or emergency spoilers would also probably serve as a good way to reduce the risk of divergence as well.

All of these would clearly have to be tested on models to judge performance.

Slightly off topic, I have considered using fixed tricycle landing gear as an inverted V tail, just spread wide. Combining them into one part like Rutan did for his tip gear.

Can’t use tip gear on forward sweep for obvious reasons, but tricycle gear is actually in a very good position to serve as additional control surfaces.

Or just to act as a good source of static stability. It doesn’t necessarily have to be a moving surface.

I don’t really trust small models because of how different the Reynolds numbers make them handle, but I would definitely like to try out something large.

With so many different configurations it’s going to have to be modular as well.

Aesquire

Well-Known Member
As Topaz states, there are 2 different "divergence" issues .

Aeroelastic divergence, which can be dealt with by changing the torsional stuffness, not the area behind the spar, as I understand it.

Pitch stability "divergence" where the root stalls first, pitching the wing nose up as the tips are still flying. As Topaz says, you don't want tip stall either. Because the elevators are in the center section, so they have the greatest leverage, a center section stall does more than move the lift forwards, it also kills the up elevator aerodynamic see-saw effect, pitching the nose down. You want to tailor those opposing effects to get the pitch stability results you want.

A little washout in the wing helps with reducing tip stall/spin, and the Aeroelastic divergence issue.

I like the inverted V tail gear/vertical fin idea. Forward swept wings have the YAW stability issues the reverse of swept wings. On a swept wing the wing yawed back has less frontal area than the leading wing giving a kind of dihedral restorative effect in yaw. Forward swept wings have the opposite and need a vertical stabilizer to counter act that.

I'm doubtful about elevators outboard. To get a canard effect you'd need a lot of sweep. In most forward sweep designs, the tips end up with the trailing edge fairly near the CG. So you'd have little leverage in pitch with elevators near the tips.

Aesquire

Well-Known Member
Re: wing tip landing gear.

Technically you could make tip mounted gear in tail dragger configuration like a Rutan Quickie, but forward swept wings usually have more dihedral than a straight wing, so tip mounted gear would be long and higher drag, and the drag is in a lousy place. So I agree it's a bad idea!

Swept wings, otoh tend to have anhedral, but tip mounted gear puts landing loads way out on your wing spar, a bad idea, and in a motor glider design with a 45-50 foot wingspan taxiing would be a nightmare.

On the Quickie, Rutan was pushed into tip mounted gear by the desire to reduce weight and drag by the low power available in the Onan RV generator engine specified by the customer. Probably not his best idea ever.

BJC

Well-Known Member
HBA Supporter
I actually like Zenith's solution on the CH650 which is opposite, the spar is straight (constant height), and the wing tapers in on the rear edge, gradually increasing the wing's thickness percentage to the tip.
Jim Bede frequently suggested that E-AB airplanes with tapered wings should use the same basic airfoil, but increase the percent thickness along the span to have a thicker section at the tip, rather than use washout, to achieve a more pleasing stall.

BJC

cheapracer

Well-Known Member
Log Member
Jim Bede frequently suggested that E-AB airplanes with tapered wings should use the same basic airfoil, but increase the percent thickness along the span to have a thicker section at the tip, rather than use washout, to achieve a more pleasing stall.

BJC

Yes, I taught him well. I remember all those long conversations in the 60's on the Internet showing him what was what, "Build a small jet I told him, people will love you for it". Yes, I remember it well.

Because of my rib system, I can simply laser cut all those individual ribs out with ease, I no need no steekin form blocks.

Doggzilla

Banned
HBA Supporter
After thinking this all over the safest design is probably neither trailing or tip elevators.

Tip elevators would result in rolling if there is even slight yaw, and trailing edge elevators may stall, as others have stated.

I think either the inverted V tail or having the elevator behind the prop like the “Aussie plank” would be the safest to guarantee pitch control.

The landing gear has to be there anyways, might as well make it serve two purposes.

Speedboat100

Banned
Keeping a constant percentage thickness is also possible.

FSW solves nicely the CG problem vs overall weight in a very small plane where the wing other wise would be positioned under the pilots butt.

wsimpso1

Super Moderator
Staff member
Log Member
If I understand correctly the idea is to make the area behind the spar larger so that when it’s under heavy lift it bends nose down instead of nose up. The X-29 had something similar.
Having chatted with folks intimate with the X-29 project, I can tell you two factors made the highly swept wings of that airplane work:
• They did bias the coupling of bending and twisting by the lamination schedules of the wing components so that tip divergence was smaller than it would have been otherwise, but that coupling was nowhere near negated;
• They made the wings several times stiffer in both torsion and bending than would be normal, which greatly reduced the adverse deflections and increased the natural frequencies. The result was a rather heavy bird compared to the donor F-5, but it allowed the test program.
I was considering having elevators on the outboard section to provide canard like control ahead of the CG, and having ailerons mid wing closer to mid chord so they are aligned with CG and have a better yaw moment. Probably a tail elevator on the inner trailing edge that works opposite if needed.

The outboard elevators would produce a force on the trailing edge, helping to counteract aero elasticity bending the leading edge.

Outboard split flaps or emergency spoilers would also probably serve as a good way to reduce the risk of divergence as well.
In order for these to be effective, a couple things would be needed:
• The elevators would have to be flown through an autopilot to (while in normal control rules) prevent stalls and to (when in direct flight rules) respond to stalls by unloading the tips. Not a stick pusher, but fly-by-wire with stall recognition and recovery built in. In addition, you may induce significant instability with these forward elevators, so CG limits would have to be very carefully developed and then strictly adhered to;
• Condition to control surface deflection time may have to be quite fast if the system is to keep up with the airplane attitude and with the wing tip deflections while vibrating at its raised natural frequency. Control theory work will be substantial ;
• To test this with similitude, you will need to deliberately adjust the bending and torsional stiffness of the model to reflect that in the full size ship, then have the sense - control calculation - signal - actuator response time adjusted to scale with the model's wing tip natural frequencies.
All of these would clearly have to be tested on models to judge performance.

Slightly off topic, I have considered using fixed tricycle landing gear as an inverted V tail, just spread wide. Combining them into one part like Rutan did for his tip gear.

Can’t use tip gear on forward sweep for obvious reasons, but tricycle gear is actually in a very good position to serve as additional control surfaces.
You are proposing main gear legs in the inverted V tail? If so, you are placing the "mains" very far aft. and the nose wheel will carry much of the airplane weight. This has a bunch of tails that you will find in model trials:
• Landing at speeds near stall speed with a fuselage level landing gear configuration will be in the sequence of tail contact - nose down accelerated rotation - nose wheel strike with considerable energy. A strongly nose up landing gear posture may be needed for controllable landings;
• The airplane will not be able to rotate for take-off in any conventional manner and will thus need to be in a strongly nose up posture to achieve takeoff at any speed reasonably close to stall speed;
• While on all three wheels, it will be in the same condition as a tricycle gear airplane being forced onto the runway at too high a speed - very little load on the "mains", lots of load on the nose gear, and a distinct tendency to swerve or ground loop. This has almost all of the characteristics of a tailwheel airplane plus:
• The "mains" (tailwheels) will have limited ability to keep the nose straight ahead because they are lightly loaded;
• The nose wheel will be ahead of CG and thus tend to drive accelerating yaw rotation - since airspeeds are still in the vicinity of stall speeds, this will not likely look like a ground loop so much as a big uncontrolled swerve to one side ending in wing tip strikes, leaving the runway, etc ;
• The nose wheel is not just a little ahead of the CG, as in most taildraggers, but may be a long way ahead of CG, greatly increasing all of the adverse effects.
In short; UGH!

Or just to act as a good source of static stability. It doesn’t necessarily have to be a moving surface.
What is to act as a good source of stability? The horizontal tail? Hate to be a grammar gripe, but you did not say, thus the question. The configuration you are suggesting with elevator effect in pitch axis control will likely be small and may have to be supplemented with an aft elevator too. The pitch and yaw damping available from the inverted V or Y tail will be useful and probably will needed for stability in pitch.

I don’t really trust small models because of how different the Reynolds numbers make them handle, but I would definitely like to try out something large.

With so many different configurations it’s going to have to be modular as well.
You will have a LOT more modeling similitude issues than just much lower Re. All of your control sensing and natural frequencies will likely be way off between model and full-size. Control theory work will be needed including lag between flight condition sensing and control surface response.

Let's ask again what this is supposed to achieve? How will it make for a better airplane? It will need to be quite a bit better to be worth the extra structural weight, weight and complexity of control systems, and development effort to result in a stable and working airplane.

Billski

Doggzilla

Banned
HBA Supporter
Okay so we are mostly on the same page. I’ll try and answer this without being too verbose.

When I said an inverted tail has greater static stability, I am referring to the F-16 belly stabilizers. This is the reason it only has a single tail. When at high AoA the belly stabilizers stick into the wind instead of reclining backwards like a conventional tail. Small surfaces with good airflow perform as well as large surfaces being choked behind the fuselage.

As far as gear goes, it would absolutely have to deploy forward to get better weight distribution. Probably flick open like a knife and have the wheels right ahead of the tail. This also allows easier shock absorber design.

The nose gear also may need to be more of a center gear like the U-2 if ground control becomes an issue.

As for wing tip stall control, the controls would need to be designed so they maintain lift even when the wingtips themselves have stalled. They would need to be designed to behave independently of the wingtips in order to maintain control.

Since they are only a fraction of wingtip area, allowing them to remain in control should not pose risk of tumbling backwards during stall.

Doggzilla

Banned
HBA Supporter
Or refer to post 15 for an alternative design

Speedboat100

Banned
This is the most famous forward swept winged aeroplane.