Assuming there are no mechanical linkage problems, what would cause a plane flying in high alpha knife edge to rock up and down. Acts as though the rudder is stalling... recovering and so on. Condition worsens with greater rudder deflection. The plane in question has a huge rudder to fin ratio, long tail moment.

I have noticed full scale aerobatics planes with triangle stock attached to the trailing edge of the rudder. Whats the purpose of this?

Kevin

I have NO direct experience with this myself. However, I can think of a couple of ideas.

My first thought was that your rudder was having problems with the servos being overloaded or with flutter when you use the rudder in that extreme of a manner, and maybe you should increase servo torque and stiffen the linkage and maybe the rudder itself.

Your question about the triangle stock reminded me of something I saw the other day on a 40% Extra. The pilot had slit fuel tubing lenghtwise and taped it to the TE of the rudder. This apparently fixed a tail waggle problem in (if I recall) downlines. He didn't have an explination of why it worked, just that it did work and it had been suggested to him to try. Looked funny, but worked well. Since t's easy to try and undo, you might try it.

teh triangle stock you are seeing is a tailbrace. it simply strenghtens the tail section and is not of use for smaller planes. i would agreee that your control rod is either flexing, or your servos is getting overloaded. to increase the torque, try going to a 6 volt pack. if not, get a higher torque servo, the plane in question i assume is an aerobatic plane. try doing the same thing at 3/4 throtle and tell me what happens

do you have a gyro on the rudder?

Schmleff

There is an aerodynamic effect that causes control surface "hunting" on sharp afterbodies. Sharp trailing edges cause the surface to hunt. John Monnette (of Sonerai / Monex / Sonex fame) had terrible problems with his Monex being unstable. After many other changes, he discovered that it was the sharp trailing edges on his control surfaces that was causing it.

Schmleff

Holds KE fine at 3/4 to full throttle with little rudder deflection. Its when I slow it down and add more deflection that things get strange. Flew it again tonight...acts like when you apply rudder to pitch the nose up, the fuse has a reaction and wants to bounce back to level flight. I suppose it could be the very long tail moment, lots of fuselage side area behind the CG. This is only a 24% plane. Have a JR 8411, 155oz servo on kevlar pull-pull, no gyro.

Here's a picture of the fuselage, and the big rudder...is this considered a sharp trailing edge?

Kevin

is that a kit? If not (or if so I suppose) is the airfoil shape symetrical on the rudder?

Nice color choice!

Schmleff

Its a rounded airfoil tail, if thats what your asking. I think a flat plate tail is also considered symetrical? Its the same on both sides.

Plane is a Colombo Anderson 24% 300L kit. Balsa sheeted foam surfaces. I'm calling it a Extra stretchy.

I'm starting to think its a very sharp trailing edge, considering how long it is. I could try a fuel line on the trailing edge next flight.

Kevin

That fuel tubing might act as an "energizer" and keep the airflow from separating from one side of the surface. In short it sounds like your rudder might be so big that you are actually stalling it. It recovers and stalls again, causing the rocking you see. This just my guess though!

Hello Kevin,
You have already told me that your engine can turn an APC 17x6 at 9000RPM (51mph speed). At 3/4 throttle, your plane would fly at 38.25 mph speed .
From your picture, the rudder deflection seems to be around 40 degres from the center?
If we put those number in servo calc software we have:
Rudder
Average control surface chord (cm) 23
Average control surface length (cm) 36
Optional: Maximum available servo torque (oz-in) 155
Maximum deflection of servo arm from center (degrees) 42
Maximum deflection of control surface from center (degrees) 40


Maximum required torque at maximum airspeed (oz-in) 155

That rudder is so huge that we would need 2 servos in parallele to maintain 45 degres deflection at full speed. Amazing!
I think your servo is not enough strong to maintain its position at 3/4 throttle.[8D]
Scou

I forgot to write the calculation is based at 40mph maximum speed wich is 3/4 of full speed. At 50mph, we would need 242 oz-in of torque with the same rudder and servo deflection.
Scou

Scou,

What servo torque calculator you using? I can't move the rudder(phisically with my hand) against the servo's torque, for fear of breaking something. This servo is very strong, and the servo arm is 1" shorter than my rudder horn, so I should have a mechanical advantage. I just doh know???[sm=confused.gif] besides, we can't be putting "two" 2oz rudder servo's in a 1/4 scale plane[sm=spinnyeyes.gif]

Kevin

Hello Kevin,
Send me your email in my PM and i'll send you the servo calc excell software. Very useful software to size servo...
A cheaper solution would be to play with different prop and see the flying performance. A perfect prop for 1/4 scale is 18x6, 18x8 or 18x10. Maybe is to strong for the G26 but you can try with 17x8 or 17x10 prop and see the difference???

Less speed and more vertical thrust make sens with aerobatic planes.
Scou

Whats interesting about this scenario is that I've been flying a 1/4 scale Aeroworks Edge 540 for a few years now with a 90oz rudder servo. That plane will high alpha KE at very slow speeds. Same 17-6 prop. max rpm 9000. Plane weighs the same, but smaller rudder. I'm about to try changing out my 8411, might have a bad one....

Kevin

That plane will high alpha KE at very slow speeds There is big diffence on the air speed coming to the rudder between slow speed and 3/4 throttle!
Daniel

Scou,

How are you coming up with the average rudder control surface cord length of 23 cm?

My rudder measures 35 1/2cm tall at the hinge line
18cm at the bottom cord, back of hinge line
10cm at the top cord, back of hinge line

I think that gives me a average cord of around 14cm

Need to remember the rudder area forward of the hinge line is not loading in the same direction as the rear.

What do you think??

Kevin

I just ran the numbers on servo calc.... coming up with max torque of 95 oz/1" at 50 mph. Much closer to what most manufactures would recommend on a 1/4 scale plane. Am I missing something??

Kevin

Id try the fuel tubing on the TE of the rudder. I know that one some of the composite aerobatic planes, you will see this wiggle. My FiberClassics does it as well, but it is usually in high speed flight, and I know it has plenty of torque on the surface. Try the tubing, I know that is a fix on the 40% FCs and such.

Hello Kevin,
Maybe you are right by using only the average moving surface cord in your calculations. I did mine 'quickly' with non moving+moving rudder surfaces that give me 36cm average cord instead of 14cm.

Why don't you send an email to Ragz if he has the same knife edge effect with his Extra?
Keep me posted...
Scou

To me something is not being calculated into the loading is the fact that the rudder has a balancer, which would reduce the servo load . At a certain point( I think its about 25%) the area ahead of the hinge line would actually start to force the rudder away from neutral. One can tell by moving air over the rudder without any linkages attached to it.
When 3D was evolving MAN had a pic of Mike McConville's modified MW 300 with a huge counterbalance which I copied building mine. I had the tail wiggle on windy days and an engineer friend told me what I stated above.
Still, rudders like tons of power even with a balancer.

Fuse is probably stalling.

Mike

Tipover,

Can't offer any insight to your knife edge waggle, but can comment on the triangular stock on the trailing edge. I currently fly a G-202 in IAC full scale competition and use the equivalent of triangular stock on my ailerons. Actually, I use P-strip (it looks a little better and is easier to apply and remove), which is an extruded adhesive-backed rubber strip available at any hardware store and intended as a door sealer. A 12" long piece attached to the top and bottom of the trailing edge of the aileron and the trailing edge of the wing just before the hinge line, provides a "centering feel". You can actually feel a "notch" at the center position of the stick. I have mine located at the outer end of the aileron, but I've seen them at the mid-span as well as the inner end of the aileron and all seem to provide the same effect.

Some aerobatic aircraft have very thick trailing edges on their control surfaces; the Sukhoi for example, has ailerons with a trailing edge nearly 2cm thick. The purpose, I'm told, is to eliminate a vagueness around the center position. I would guess the triangular stock you saw on the rudder has the same purpose.

Thanks to everyone for your responses. The full scale input is especially interesting. One thing missing from our models is the control surface feedback you only get while hanging onto the controls in a full size plane.

Got in a couple more test flights earlier this week. First up was a 1/4" balsa stick taped up the back side of the rudders trailing edge. No real improvement in KE angle of attack control, but it did dampen much of the waggle out . Plane seemed to track better in general. Most noticeable was how well the down side of a stall turn tracked. Also noticed less pitch coupling to the belly while in KE.

Next up was a rearward cg shift of about 1/4". This change along with the trailing edge rudder stick was indeed heading in the right direction. The plane would now maintain KE at about half to 3/4 throttle. Adding power, it would climb. So, it seems that cg is the major factor in how well this plane will high alpha KE. I have a feeling that because of the longer than scale tail moment, the KE cg is shifted farther forward from the canopy area than is normal for an Extra. Seems the long fuselage makes the plane so stable, that it resists any yaw control inputs. Could be the reason the designer used such a silly large rudder area? It looks impressive, but its really just balancing out the long fuselage, and just breaking even.

Kevin

I have an [link=http://www.geistware.com/rcmodeling/calculators.htm]excel program [/link] like that written by a guy when he was going to MSU. If you come across him, let me know. I need to get him to modify it slightly. It does have a couple of bugs.