Balancing ailerons. How?

[Joel K. Scholz]

I just finished flying my wing to destruction, due to wing flutter.I am rebuilding a beefier wing, but I would have more confidence in it if I can figure out a simple way to balance the ailerons (elevons). http://www.youtube.com/watch?v=oCpP_o6ZtlI

[Rodney]

The main thing is to have very stiff (no soft balsa or structure that can be easily twisted or deformed by external pressure. Next, good tight hinges and pushrod assemblys with absolutely no slop or play. Make sure the pushrods can not bow or buckle under airloads. Next, hard to do, have as much mass in front of the hinge pivot point as you have behind it. Sealed hinge gaps help, not by reducing flutter but by eliminating the turbulence that might induce the flutter with flow through the gap. Just remember, everything will flutter, even an iron bar, if the stimulus is right. You just shift the resonance point into a higher frequency by stiffening up the structure and eliminating any possibility of sloppy or soft hinges and control push rods.

[Brandano]

That looked like the entire wing fluttering, though it could have been induced by the elevons. Maybe that initial dive in the ground did more damage than what it looked like? In any case a sturdier and heavier leading edge may improve things. Also, the full span ailerons will be more prone to flutter due to their intrinsic flexibility, they will tend to twist and flutter like a flag. Maybe you should strengthen them as well, or find a way to drive them from their center. the most common way to balance ailerons is to add balancing "horns" near to the wingtips, but on long thin ailerons like these some weight should be added in front of the hinge line further inboard as well.

[pmackenzie]

Ouch.
To me that did not look like aileron flutter.
The frequency was too low. Ailerons usually "buzz"
I think that was the whole wing twisting.

Pat MacKenzie

[Joel K. Scholz]

ailerons were sealed and linkage secure.I had moved the control horn further out on the wing, but will move to the middle of the elevon. I am not clear on how you balance the aileron in front of the hinge line. Like a stuka? The ailerons themselves were normal solid 1 1/2" aileron stock, with a triangular balsa piece to form a hinge lin flush with the top surface of the wing. Would it help to hinge them on the middle of the trailing edge of the wing?Below is the link to the build log in flying wings forum. 86"Horton


http://www.rcgroups.com/forums/showthread.php?t=782118

[flypaper 2]

Used to have the same thing with the original Zagis called the Zagi flap. You could put it in a shallow dive at full throttle and it would do the same thing. It was mode of foam so it didn't blow apart. As soon as you throttled back and slowed down it quit. Heavier spar to the wingtips cured it but took the fun out of it. :D
Nothing to do with the ailerons. Just make the wing more rigid.

Gord.

[Brandano]

There's several ways, the Stuka's ailerons were designed more for simplicity than for efficiency and are maybe overkill, but on a dive bomber flutter sure must have been something to think of during the design. All setups rely on having the same amount of mass in front of the hinge as behind it. On full size planes this is obtained by embedding steel, lead, tungsten or even depleted uranium in the leading edge portion of the control surface, the one that is embedded within the wing profile. Other smaller plane use horns protruding above and below the control surface, and ahead of the hinge, opportunely weighted. In some cases these are shaped like blades or closed loops, perpendicular to the hinge, and punching through the wing or fixed portion of the rudder or elevator. Sometimes the counterweights are shaped like aerodynamic profile and act also as aerodynamic counterbalance to reduce the amount of force needed to move the control.

[flypaper 2]

On the P51 Mustang, that protrusion on the leading edge of the rudder that extends into the fin, is a weight and aerodynamic balance to control flutter.

[nmasters]

If balancing is your only option then it all boils down to one thing. You need some lead in front of the hinge (preferably as far ahead as possible). A frise nose or a balance horn (http://rwebs.net/avhistory/history/Zeke32.htm#Elevators) is a handy place to stash the counterweight. A frise nose also helps with adverse yaw in upright flight but makes it worse while inverted. So if you plan on flying upside down you should keep that feature to as short a section of the aileron span as possible. Another way I’ve seen counterbalances attached to ailerons on full size ‘wings is an aluminum tube sticking forward with the end of the tub either filled lead shot and epoxy or a streamlined bulb molded onto the end. The second type can be longer than a frise nose and therefore you can balance the surface with less lead but it needs room to move. The tube has to stick down at an angle equal to the maximum down deflection of the eleven to provide clearance. Here’s a good picture of the counterweight on a MitchellWing (http://www.mitchellwing.com/images/pics/Tom's%20Mitchell%20Wing%20.jpg). This type is easy to adjust and effective but prone to snag things on the ground in a wing low landing. I remember one guy called it his rabbet spear.

Another type has the same tube sticking forward but instead of a lead bulb it has a vertical bar that can move up into a small hole in the wing.

That said I agree with the people telling you that it needs to be stiffer. A technique that’s appropriate to swept wings and doesn’t add very much weight is called “desweeping the spar”. Basically you just move the root end of the spar farther back along the root chord so that the spar sweep angle is less than the 1/4c line. The trailing edge of the D-tube sheeting should also follow this line so you end up with a tapered D-tube. The resistance of a tube to torque is proportional to its cross section area (or is that the square of the area? :p ). Anyway what’s going on is that although the flutter exciting frequency is being generated near the tip it’s got a lever twisting the root so you have to make the whole wing resist being twisted. This simple mod dose that and the only extra weight is a few square inches of D-tube sheeting. You can also counterbalance the wing the same as the elevons. Any extra mass ahead of the spar will raise the flutter speed.

--Norm

[flypaper 2]

Also why the engines on the B 52s are hanging out forward of the wing, to act as aerodynamic balancers to stop flutter. Sort of killing two birds with one stone. :p

[Joel K. Scholz]

Thanks gentleman, all this is very helpful.

[MarkusN]

Wings with sweepback are notorious for a coupled torsional / bending flutter. Ailerons can be a part of it, increasing the torsional input, but the wing also does it on its own. Hard wings help. So does a spar which has less sweepback than the wing. (i.e. which goes through the wing center rather far back.) The Braunschweig guys did this to de-fuse the rather nasty flutter tendency of the SB 13.

[nmasters]

Also why the engines on the B 52s are hanging out forward of the wing, to act as aerodynamic balancers to stop flutter. Sort of killing two birds with one stone. :p
Yep, and most other jets with under wing engines. The 747 had 700 lbs of DU in the nacelle nose ring of each outboard engine. The production models used cheaper lead. The MiG 15 also had a small chunk of ballast in the leading edge near the tip. I’ve heard that it would have taken about 10X the weight in structure to achieve the same flutter reduction. Dragonflys also have a dark spot called the stigma in the same spot apparantly it's a censory organ

--Norm

[biber]

Many modern fullsize gliders have lead in the outer wing portion's LE.
At least ASW 27 and ASH 25 have it.

[markdrela]

Wings with sweepback are notorious for a coupled torsional / bending flutter. This is called "body-freedom flutter", or BFF, which is a very big problem on swept flying wings. It's caused by the interaction of aircraft pitching and wing bending --- the wing torsion is almost irrelevant.
Here's another video in slo-mo: http://www.youtube.com/watch?v=nRit6tcNT4s

The way to get rid of BFF is to make the wing bending frequency be much larger than the aircraft's pitching frequency. There are two main ways to do this:
1) Add a fuselage to increase pitch inertia and slow down the pitching frequency
2) Stiffen the wing spar to speed up the wing bending frequency.

Solution 1) is not an option for a flying wing design, which leaves only option 2). On the SB13, the wing spar caps had to be increased to some crazy levels to suppress the BFF -- something like 5x or 8x the normal weight compared to the sparcaps of a straight-wing glider. This spar weight penalty is a very unattractive feature of swept flying wings.

[JetPlaneFlyer]

Mark,
Wow... so simple and obvious once it's explained! I'll certainly make a mental note of 'BFF' for future reference.

Adding weights in the wing as suggested earlier would therefore be likely to be couner productive as it would bring the natural frequency of the wing in bending down and 'BFF' would manifest itself at lower airspeed :(

I guess carbon fibre spar reinforcement would be the way to go on these models rather than traditional spruce spars? It may be hard to get enough wood into the wing without increasing thickness.

Steve

[Joel K. Scholz]

Mark I think your video shows exactly what happened to my plane.

[markdrela]

It may be hard to get enough wood into the wing without increasing thickness. It's worse than that. For a given airfoil thickness, bending frequency is dependent on the stiffness/mass ratio, and not on the stiffness alone. So once the wing is "mostly spar", adding more spar material will not increase its frequency. If you double the stiffness you will also double the weight, with no increase in the ratio and no increase in the frequency. So the only solution then is to switch to a spar material with a greater stiffness/mass ratio, e.g. carbon.

[markdrela]

Adding weights in the wing as suggested earlier would therefore be likely to be couner productive as it would bring the natural frequency of the wing in bending down and 'BFF' would manifest itself at lower airspeed Correct. One thing that does help a bit is to move the heaviest bits (servos, battery, motors) to the bending nodes of the wing, typically about 1/3 out on the semispan. The nodes are the points which don't move up and down while the wing vibrates in bending. Any mass at a node point is "invisible" to the vibration, and will not act to reduce the frequency.

[macboffin]

The best fix is undoubtably a very stiff leading edge "D" cell, ie leading edge sheeted back to main spar, which has vertical grain webs tying top to botton spar cars. The provision of a cross-spar at 90 degreees to fuselage also a very good detail.

[Texas Buzzard]

First applaud you cameraman. Secondly, the stills were excellent for seeing the construction.

Mark & nmasters comments were very good and all the others were of equal importance.

You asked a question, "How do I balance the ailerons?" Someone said the ailerons will "buzz" - I have had that happen at high speed, he is right.

Joel, from the video and from experience I DON'T THINK AILERON FLUTTER CAUSED THE EXPLOSION. The video shows a tremendous up and down movement extending through the entire wing. Alileron flutter doesn't cause that.

For the wing to be stable there has to be some downward force aft. This can be from tip washout and/or bending the ailerons up out at the tip. I heard on the video that lots of up-trim was put in. That is expected.

As the wing gained more speed the downward force of the tips became greater. The pilot dialed in a bit of up. The front center was exerting an upward force. These COMPETING FORCES caused the bending and destruction of the wing.

In the stills I see the leading edge is narrow (front to back) and you did not finish the trailing edge sheeting. I don't see the vertical sheeting between the sheeting on the trailing edge. This will give some stiffening. As others said a "D-Tube" Leading Edge will stiffen this wing a lot.

I see you relied on CF Tubes for the main spar in this wing. That eases construction, but a good old-fashioned I-beam Spar maybe with CF tape epoxied to outside of spruce spars. Sailplanes can be winched up and their wings don't fail. They use "D-Tube" and CF.

Finally, a change in the design will help. Use a D-Tube with CF. A few ounces being added will not hurt performance. It will help your satisfaction.

I added the diagram just to answer your question.

[Joel K. Scholz]

If you would like to see how I solved the problem, please visit this thread
.http://www.rcgroups.com/forums/showthread.php?t=782118

[BMatthews]

On a little 1/2A powered racer I had some years back the ailerons were so thin and flexible that it fluttered easily. I added some mass balances on stalks that stuck out ahead of the hinge line out by the 80% span points and the flutter went away. Or at least moved outside of the model's performace envelope. I used the same method used on some full sized aircraft for this but obvioulsy somewhat more clunky looking due to the size of the ballance weights. But it did work well. The P51 like horns built in would also work well if you design them in during the initial layouts.

If you go with mass balancing you don't need to move the balance point ahead of the hinge, you just need to get it ahead of the 25% MAC point of the control surfaces. On longer and more flexible controls you may need to split the mass between two such stalks or forward protrusions so each segment is balanced to avoid the flutter.

[Brandano]

These are off a BAE Lightning
http://www.binbrook.demon.co.uk/images/ailerons.jpg
Note the aerodynamic and mass balance on the tip portion and the blade shaped counterweight on the inboard portion, cutting through the wing

[Joel K. Scholz]

What an excellent idea.

[nmasters]

I'm not sure if this is the same flutter mode but it's worth seeing.
http://www.youtube.com/watch?v=MlbYmqi5paE

[vintage1]

Well, I missed this therad somehow.

Its patently obvious this isn't flutter at all.

On the last turn but one as it comes towards you, you can see the wing under the G loading suddenly acquires dihedral that is not built in. The spar has cracked. Then as a little G is applied the wing simply folds.

It needs more spar around the center section, and probably a strip of CF or glass as a cap strip.

The whole wing may not be torsronally stiff enough which might account for the strange behaviour during the roll.

I saw no sign of flutter at all - you generally get ailerons ripping off if they are fluttering, and you sure can hear them doing it in a high speed pass.

Which isn't to decry the excellent info here on aileron flutter, but it aint the main problem here.

Aside, what a beautiful model in the air. Well worth sorting out the structural stuff.

Details on the construction and where it failed may be well worth posting. I can just about tackle the maths of the main spar, if pushed..

[MCarlton]

Oddly, this looks exactly like the sort of harmonic pitch oscillation which plagued some of the "Mach1" research aircraft ie DH106.

It didn't look like flutter at all to me, I think V1 is right, but I would add the following;

How much washout is built into the wing?

Here's my probably nonsense theory;
On a light structure, bags of washout can do some strange things, as I mentioned in a previous post, it is entirely possible that at low AOA, the tips can be at a negative AOA and lose lift. This could even lead to a "negative" lift situation at the tips, thus increasing the "twist" further.

At this point, a proportion (depending on where the washout starts) of the wing could be stalled, especially right near the tips.

That stall condition won't be noticed immediately as much of the rest of the wing is still flying.

What it does mean though is that the effect of the washout as a pitch stabilising force is taken away and the model pitches forwards.

The whole wing goes into a negative AOA position at this point and airspeed reduces, allowing the tips to "untwist". Suddenly, normality is restored and lift is returned and this invokes a sudden "pitch up".

Sudden pitch up = Sudden increase in "G" + Light Structure = Crunch.

I don't know how your "elevons" are arranged, I am assuming constant chord at the tips? I would change that to a constant % of local wing chord.

I think that you could sort the structure out by stripping the covering away and sanding a shallow groove in the top and bottom of the spar, no more than about 1mm deep x 3mm wide. Then lay in some carbon tows wetted out with epoxy resin.

I would also think about covering the model in tissue/dope or something else which adds some strength. Plastic films are fine in their place, but there's no strength to them really and using them relies entirely on structural strength.

Just a few *off the top of my head* thoughts