At last! A good post in this thread. I haven't read all of it but I have never seen so much bad information in one thread. This and Dr. Drela's post are the only ones I would pay any attention to.
The internet has been most instructive. I have learned things my aero professor never taught me, my instructors never showed me, I never experienced in 30 years as a pilot, and I never saw in 32 years as a wind tunnel test engineer. Very amazing indeed.
On the internet, everybody is an expert even if they didn't stay at Holiday Inn Express.
ORIGINAL: AlexanderMikhailov
When speaking about aileron/flap flutter there is a lot of factors we have to take into account. Let me briefly summarize my own knowledge for European molded ships.
1. Linkage slop (linear). All holes should perfectly match clevises (0.01mm tolerance). Clevises should be glued to the rods. Linear servo shaft slop is nonexistent with BB servos.
2. Servo slop (angular). Some models like HS-(5)125 have significant angular play from the very beginning. Sooner or later it happens to every servo with metal gear. You cannot change that -- just take it or leave it.
3. Servo mount slop. In molded wings servos are attached to the skin. It does not matter how good the servo or the servo mount is glued to it -- skin plays a LOT! This fact is usually overlooked. The simplest way to reduce this play is to glue a thin (0.2mm) carbon/glass patch. The size is limited by your ability to extend it sideways through the servo opening. Anyway it should reach the wing spar. The ultimate solution is a large size servo base, which forms quite rigid structure when glued to the skin and wing spar (picture attached).
The worst thing is that we just reduce the play to some acceptable level, not totally eliminate it. Our final goal is to achieve the minimal control surface play with given sum of different slops. And now we start the most interesting part -- turn on our own common sense.
1. Linear travel. It is very easy. The bigger the radius, the bigger linear rod travel thus smaller linear slop effect. So both servo arm and horn should be as long as possible.
2. Angular travel. The rule of thumb is -- you should use the full range of angular servo arm movement. In this case you reduce negative angular slop effect.
3. Transfer ratio. The most tricky to understand. A lot of unjustified opinions. Just remember -- reduced transfer ratio makes control surface immune from the effects of both linear and angular slops.
So it is time to turn theory into several practical steps:
a. In molded sailplanes the limiting factor is a wing thickness. You should be happy if servo cover allows you to use hole #2 from the center.
b. Engage the full servo travel on your computer radio or servo tester. In case of digital Hitec servos I suggest to use Hitec HFP-10 servo programmer to set hardware endpoints to their respective maximums.
c. Take a full breath. Remind yourself that this moldie is a sailplane, not a 3D monster. You do not need big control surface movement, limit yourself to sophisticated adequacy. Speaking of ailerons, UP travel should be twice as DOWN travel. Speaking of flaps, 3-4 mm UP and 70-90 degrees DOWN.
d. Connecting pushrod. The common mistake is to connect "servo arm @ neutral (1'500us) to control surface @ level with profile trailing edge. Do not forget, in accordance with everything said we are going to use the full servo travel. So we are to find transfer ratio, when the servo endpoints correspond with control surface deflections as per previous step. If you are not good in trigonometry, make a kind of threaded horn and play with it. Or better yet, make a simple test bench from any piece of sheet material. Then print wing cross-section 1:1 scale and glue it to your test bench. Find rotation centers and bolt-on two servo arms. Find and mark servo arm and control horn endpoints according to wing cross-section. Connect them with temporary threaded pushrod/clevises at corresponding endpoints. Move to other side. See if control horn hole should be closer/further to the rotation center. Drill the new hole and check again. When the correct transfer ratio is found, measure the distance between rotation center and the hole. Do not forget that on the wing it should be a distance from a hinge, not a horn length.
e. Temporarily fix your horns to control surfaces and check, how good you were in trigonometry. If everything is OK, glue horns and clevises. Actually with some experience it was possible to skip step "d" but these beautiful moldies do not give you chance to UNDO if horn is glued out of place. I personally prefer to make very simple tool to hold the control horn with adequate and repeatable accuracy while epoxy cures.
f. Now we have control surfaces deflected at servo neutral. Ailerons are 1/2 UP, flaps are 35-40 degrees DOWN. Turn on your computer radio and use appropriate functions to set ailerons/flaps. In Futaba 9Z it can be done through conditions menu or better yet, through trim sets.
Rejoice. Your molded beauty will pay you back for your efforts.
Sincerely,
Alexander
Moscow, Russia
P.S. And the last but not the least. Landing with flaps fully deployed usually strips servo gear, but not in my case. When flaps are FULL DOWN, servo arm is coaxial with pushrod and no torque could be transferred to the output shaft.