flaphoto

Flew my first flights today. Had two experienced pilots with me. The new bird flew great but had a flutter that only appeared to occur out of a shallow dive, not high speed and not under power. The other pilots said a couple of things about the linkage, but it appears to be slop free. One of the pilots pointed out that the servos did not appear to be locked in, in other words, they had play around neutral center when they should have been rock solid stiff. Ther servos are Hitec HS 512mg. No other problems, flew well and didn't even need to trim it. But, this is my first fixed wing and I could have overlooked something newbe-ish, so veterans, please help me out.
Attached are photos in case that helps.

flaphoto

are my servo horns too long? Should I use the metal ones that came with the servo instead of the plasic ones?

Thought of additional detail that may or may not have contributed. I had set up the radio so the flight modes (launch, cruise, land) would only operate the prop in launch position. And, when in launch position, the entire trailing edge was a a few millimeters down (both aileron and flap). I was flying around in this mode (launch with flaps and aileron slightly down) because I didn't want to have to start flipping switches if I suddenly felt I needed to power out of trouble. So, when it fluttered coming out of shallow dive, prop not powered, the aileron and flaps were still down a few millimeters neutral. Could this contribute to the flutter I experienced?

da Rock

Flutter most often comes from slop in the linkages. Way more often than from anything else. I've only run into a few servos that were sloppy around center, but that'd be the very first thing I'd fix and I'd fix it immediately. Are those servos new? If they're used, I'd swap them out in a heartbeat. Matter of fact, I'd quit reading this post right now and go do it. Right now.

There is something that should be mentioned here. Servo's usually draw the least current at center. And they also draw the least when they're under the least aerodynamic load. When your TX tells them to move that few millimeters down, they're no longer sitting at that sloppy center and probably not under the least aero load. Or maybe their gears are in an even sloppier orientation inside. Who knows...........

da Rock

I'd like to wax eloquent about a couple of things in your pictures.

I noticed that you have threaded ends on both ends of the connecting rods. Both clevis are threaded and free to move right?

Neither clevis can unscrew/screw because they're trapped by the horns BUT........ the rod can do whatever it wants, and just might. I landed a big 4Star the other day with the connecting rod on one aileron dangling in the breeze. So was the aileron. The rod had screwed itself deeper into one clevis and unscrewed from the other. The plane's owner had forgotten to tighten the lock nuts on the rod. The other aileron connector was almost out as well. Wonder what it'd flown like with both ailerons unhooked?

If I were you, I'd epoxy one clevis to it's rod. You've got way more than enough adjustment on one end.

Just a suggestion.

da Rock

Wellllllllllllllllllll

About your horns......

A couple of suggestions:
It is always worthwhile to have the servo arm connection closer to the servo and the connection on the horn on the surface farther from the hingeline. The idea is to give the servo the mechanical advantage. It's also worthwhile to give the surface less mechanical advantage. The way you have your connections, the servo has the least advantage to move a horn that is going to have even more advantage resisting that servo.

I would suggest that the leverages that're present will contribute to the problem you're having. They will also help the aileron flutter beat the servo's gears to death.

da Rock

It looks like you've made excellent choices of hardware. The aileron horns look especially good. They'd be even better if they were longer, but what the heck.

You ask about using metal servo arms. No real advantage since that application really doesn't need that much strength.

Actually, yes, but maybe in a surprising way. Every aerodynamic structure will have a specific flutter speed, even a bowling ball. The "airfoil" that resulted from the deflection could have been in "flutter speed range" at the speed you saw the flutter, whereas a "straight" orientation could have been outside that speed.

BTW. Every surface has a flutter speed. But sloppy linkages introduce new and wider flutter speed ranges for those surfaces.

flaphoto

in answer to your question, all the servos are brand new. If I understand you correctly, I should move the linkage on the servo arm inward so the servo arm is effectively shorter? I don't think there is much I can do with my horn on the control surface as I cut them down from their original length....I could always get new ones. Someone I just spoke to on the phone suggested I might be seeing so flexing of the servo mounting as the wing is of the molded variety. I had read that before, but it seems very stiff and has some sore of mesh prebuilt in it to which I epoxied a servo mount frame. I think I could try to remove the 2-3 millimeter of down aileron and flap I was flying around with. Maybe if the wing controls were more neutral the moderate dives that were resulting in flutter would not occur.

Tall Paul

The basic hardware looks to be up to the task...
Get rid of the dual clevises.
You only need adjustment at one end.
The connection to the surface. What is that? How securely is it installed?
I've had something like that flutter the wing so badly the slip-on nose cone on the glider came off in flight!
You mention it's a moldie.. the whole servo as mentioned could be moving with the skin.
Pack some stuff around the servo.

flaphoto

Tall Paul,
As to getting rid of the dual clevises, can I get away with just soldering one end so it can't rotate? If not, how do I get rid of a clevis...
The connection to the control surface was provided by ZB from Esprit Models. I don't know the proper terminology (combination of being a complete fixed wing novice and no instructions) but ZB supplied something that looked like an aluminium cylinder (almost like a big bicycle spoke nipple). It was too long to install as one piece in the control surface, so I cut in half and used half for the aileron, half for the flap. If you look closely at the photo, you can see one side (aileron) has a flange and the other had none. They were installed by dremeling out a hole and then I made the inside of the hole even wider than the hole in the surface. I filled the hole with JB weld and inserted the fitting. They are all very solid with no play.

Regarding some 'stuff' around the servo...that is another term I am not familier with...some sort of epoxy? Would a very thin piece of ply accompish the same thing?

Could the servos be under powered and not holding? The UBEC I got puts out 5.03 volts, but the hitec digital servos are rated up to 6 v.

Tall Paul

I use a Z-bend or a plain 90 degree bend with a wheel keeper, at the servo end on most of my planes.
The yellow is my Stick-It V.. and the other my Somethin' Extra.
Neither flutter, and the pushrods can't rotate.
With your surface connections JBWelded in place, that should be good enough.
The "stuff" around the servo would be balsa.
Or if nothing else works, pull the servos, mount them on a large piece of plywood, and then reinstall in the wings.

da Rock

Yes, but that's not even half of it.

You always want the connection to the servo arm "shorter" and the connection to the surface's horn "longer". And you always want your computer radio to have 100% throw on every servo, no more, no less. That's the most efficient, effective situation for the servo. If the connection on the servo arm is as close in as possible, you'll get whatever power the servo can deliver, and it'll deliver it with the most accuracy. If you have that close connection, the next problem is simply to set the connection on the surface's horn so that full movement of your TX stick gives the recommended surface deflection.

With your existing components you will probably have to swap out that aileron horn to be able to shorten the servo arm connection because that short aileron horn is going TO MOVE THAT AILERON a lot being that short. And if you shorten the servo arm connection, the pushrod movement might not be long enough. There is no rule that says you have to use only the stuff that came in the kit/ARF box.

As you've set up the connections, you're using a fairly weak servo and then putting it at every disadvantage.

If it was my airplane, I'd replace the aileron horns with longer ones. What do they cost? a couple of dollars? Then I'd reconnect the pushrods on the servo arm closer to the servo. Then find out where to make the connection on the aileron horn to get adequate deflection.

da Rock

You don't have to get rid of dual clevises. I always use clevises on both ends on purpose. The force transfer between a horn and clevis is about as perfect and efficient as it gets. I either solder or epoxy the fixed one. Both solder and epoxy work great. And clevises work excellently well.

da Rock

Actually, the suggestion is basically that if the servo isn't securely mounted, it's sensible to do something to secure them better.

It's really simple to see if you have that problem. Grab the aileron and hold it so it can't deflect. Move your aileron stick on the TX. This will not only show you if the servo is solid, but it'll also let you feel if there is any slop and where it is. You can feel or see if the clevises are loose on the rod, in the horns, or if the servo gears are not tight or whatever...........

If all the connections and gears are tight, and the servo flexes, check out why. Are they taped down? tape giving? Or is what they're attached to flexing? If so......... stiffen the structure around the servo..... etc etc

edit..... oh wait, I just looked at your pictures again and noticed the servos are screwed down. whatever.........

da Rock

BTW, we should have suggest to you that you do the hands-on test first thing. It's what we would have done first thing.... chuckle.....

And how strong are those servos?

Not a good idea.
The structure has some weakness and has only hinted what that can do. It has shown you there is a problem. There is a problem. You've been given a second chance. If you're trying simply to save yourself some work, it's not going to be much work at all. And shouldn't cost much either. Depending on what you find out with that hands-on test.

flaphoto

Thanks for taking your time to give me all those suggestions. I definately will be putting them to work. I'll also be passing on the favor of helping a new guy, this time with something I know...helicopters.
It was fantasitic today doing what little flying I got in before the flutter. Started out the day flying a four stroke aerobatic something or other. It amazes me how willing fellows are handing over their radios to a guy that just told them he'd never flown a fixed wing before. I did a few nice passes and flew a couple of loops and rolls. Great stuff.
The first flights on the glider were more nerve racking. I had a tendency to work it into a stall when not powered by the motor...newbie error. The roll rate was much slower than the other powered planes, guess I really knew it would be. Anyway, nice to get my first flights in with my first fixed wing. These gliders are really sweet in the air. Can't wait to get out again and learn some more.

Tall Paul

The best method to prevent flutter is to maximize the moment arms of the pushrod at the servo end, and the surface end.
The closer the pushrod is to the hinge line at the surface, the less resistance there is to anything that might excite flutter.
The further out on the servo arm the pushrod can be, the more precision you have in surface movement, and the resistance to flutter increases.

flaphoto

Great explanation! I definitely have it wrong on my control horn. I'm off to the hobby shop in the morning to get longer arm. Thankyou.

da Rock

A good technique to follow that will always start you out on the right foot when setting up your control hookups.............

Make the connection in the surface's horn in the hole farthest from the hingeline and the connection at the servo in the closest hole to the servo. As long as you're using a horn on the movable surface that's appropriate to the size airplane, then using the farthest hole will give you a good starting point.

Then, crank up the radio and check the surface throws to see if they're what the designer recommends. They will almost never be too much. If they're spot on, you're good to go. If they're not enough, then move the connection on the servo arm out as your first choice. You want to leave the connection out at the horn in the outermost hole until you have no other choice. (If they did happen to be too much, your only solution is to replace the surface horn with a longer one.)

Always have your initial assembly start with the connection to the surface's horn with the connecting rod in the outermost hole.

Keep in mind that computer radios can change the amount a servo rotates, but you don't want to change that from 100% unless you have no other options. Whenever you change the servo throws in your radio TX, you're either reducing the accuracy of control or reducing the servo's power. Neither of those are worth doing just to save you from swapping out a dollar's worth of the wrong hardware.

Mastertech

My Feedback: (31)

This is not correct. The closer you get to the servo center the more power and resolution you have.

Tall Paul

I'm sure you're... not correct.
THe servo has x amount of torque available.
Moving the pushrod out from center increases the moment arm, which increases the push available to the surface.
Force times distance.

Mastertech

My Feedback: (31)

I submit this for your consideration.

http://www.horizonhobby.com/Articles...ArticleID=1192

Tall Paul

2000 words.

drela

You've got it backwards.

torque = force x distance
force = torque/distance

Doubling the length of both horns will not change the servo to surface motion, but it will halve the pushrod force.
The effect of clevis slop will also be halved.

In contrast, doubling both horns will not change the servo gear slop or servo resolution. To improve these, increase the servo throw as much as possible (to +/-45 degrees or more), by making the surface/servo horn length ratio as large as possible while still maintaining adequate surface travel. In the original photo, the surface horn looks too short, and the servo horn looks too long in comparison.

Mastertech

My Feedback: (31)

Obviously they don't understand levers or mechanical advantage. Think of tightening a bolt. The the longer wrench you use the less torque you have to apply to move the bolt, Same thing with a servo the closer you are to the rotational center the higher the torque the servo can handle to hold the arm still, the farther out the harder it is to hold the arm still, it's simple. That's why we use a 1 to 2 or so mechanical advantage when we set up our airplanes. Out on the servo arm gets you more throw and less torque and less resolution. In on the flight control surface gets you more throw and less mechanical advantage.

flaphoto

I think I've got it. (if only I had paid attention in that vector dynamics class I took about 20 years ago). It appears to me, after reading several very informative replys to my question, that I have set myself up in the worst way possible; very short control horn on the control surface and relatively long servo arm.
I already have my order in for replacment, longer control arm for the control surface and when those come in I will be moving the clevis on the servo inboard toward the drive output of the servo. The lugwrench analogy was very insightful as were the drawings.
I'll post the results as soon as I can try out the recommendations