Dual Elevator Servo Question
 

I'm building a big Kaos from plans and have a servo question...

I'll be installing dual elevator servos in the aft fuselage without the benefit of a matchbox or synchronizer. A buddy informs me that neither are needed if I simply employ a "Y-harness" and use opposite arms on my servo outputs.

I really do not have the finances right now to buy more electronics so this solution is VERY appealing! Naturally, anything that appears TOO simply usually is, too simple. So, having said that...let's get some feedback from y'all...

Two high torque servos mounted in the aft fuselage; one with the pushrod connected to the "bottom" of the output arm and the other to the "top" on the other. MANUAL adjustments aside, what are the drawbacks? By the way, the two servos are exact matches.

I'm not going to be in any competitions. This is a fun-fly project. I won't be using a reversing harness, simply a Y-harness...

The reason for my question is that the normal 1/4 scale servo I would "normally" be using, can't be due to the large air cylinder mounted aft of the wing saddle for the Robarts pneumatic retracts.

So, what are your thoughts? Sensible or folly?

Pointmagu

RE: Dual Elevator Servo Question
 

The motion geometry you show will make one side move more than the other.
A reversing harness with both horns pointing in the same direction is your best choice.

RE: Dual Elevator Servo Question
 

To use two elevator servos properly, the two servo/linkages should be mirror images. This requires a Y harness and one servo reverser. The reverser is not expensive. You don't need servo "matchers" or any other special electronics.

Dr.1

RE: Dual Elevator Servo Question
 

Tall Paul and Dr1Driver are right on.

I studied this on my Wild Hare Giles and decided the stock setup (similar to yours with the servo and elevator horns pointing in opposite directions) could be improved so I reworked the elevator (and rudder) geometry to get it as close to perfect as possible geometrically. I use a Hitec servo programmer to set the servo centers and max travel but you can do the same with a computer radio.

The drawing was a working drawing for me - a bit cryptic but it represents looking at the right side of the fuselage. The right side of the drawing represents the elevator servo arm and linkage and the left side represents the elevator horn. Looks a bit strange with the offset angles but it results in mechanically (almost) perfect travel, and travel rate without endpoint adjustment. When I flew it it was noticeably smoother in pitch. Each setup is unique. Change anything and everything else changes.

RE: Dual Elevator Servo Question
 

Rather than buy a servo reverser you can rewire the servo to reverse the travel. That way both pushrods can be off the lower side of the wheel or the upper at your whim. This used to be the standard way of reversing travel before computer radios made it so easy with a switch on the Tx.

To reverse the travel of a servo you need to flip the two motor leads and the outer two leads on the feelback potentiometer. If you do that the servo will travel opposite to what it used to.

And this isn't really an aerodynamics problem but you raised so many issues I'm not really sure where to move it to.

RE: Dual Elevator Servo Question
 

Yes, you MIGHT be able to rewire the servo, provided the motor and pot are loose wired to the board. A lot of servos nowadays have these components hardwired, and reversing is almost impossible. Also, unless you have some experience with close-quarters electronic wiring, don't try it. Stuff is really jammed into a servo. Why do an "iffy" when you can buy a device that simply plugs in and does the same thing? In "the old days" we had to manually reverse servos, but not now. Take advantage of the technology that's available.


If you decide to do it manually; as stated, reverse the two motor wires and the two outside wires on the feedback pot. DO NOT reverse the incoming black and red power wires. Don't laugh, I've seen people try it that way. It fries the servo board real quick.

Dr.1

RE: Dual Elevator Servo Question
 

Well,

I posted my question here as I too, could not determine to which forum it was best suited.

Generally, folks with aerodynamic savvy also have lotza "other" knowledge to impart.

I rewired many a KPS-15 in my day. However, as mentioned today's technologies and circuits (combined with my failing eyesight) will most likely make taking up the soldering gun, unfeasible.

So, servo reversing it will be. Now, my next challenge...

Anyone know of a source for this item? I persued eBAY to no avail and am not sure of a brand name with which to do a Google search.

Seems like I'm asking alot, huh? Don't worry, I won't ask for anyone to do the actual installation :-)


Thanks.
PM

RE: Dual Elevator Servo Question
 

You are in big time Luck.

Delaware and Maryland still have loads of active modelers and clubs.

Do a search on the AMA site for clubs near you.

Contact each one till a person helps you.

Reach out to them and they will help.

RE: Dual Elevator Servo Question
 

PointMagu,
I'd do ALMOST exactly what you've got drawn in your (decently done) picture. Your original idea is simple and costs nothing. It only has a little problem that's easily solved.

You're building a large model, right? And you're building it from plans, so nothing is locked in, right?

It's always good advice to have your horns and pushrods at right angles, right? And to have the servo arm and the pushrod at right angles, right? So do that for both servo's and you're good to go.

All you've got to do is move the servo on the right side of the fuselage down some and you could angle it a bit to point it's long axis toward the hingeline. Truth is, pointing them at the hingeline is a good idea all by itself. You've got the room to move where that servo is going to be placed, so use that room and you've got the problem whipped. At zero cost and no extra doodads.

BTW, "pointing" the servo's axis anywhichaway really isn't important at all. We've gotten in the habit of lining them up square to things. But that's really not anything like a requirement. Since we can place the servo arm in any orientation, the servo could be sideways, straight on, at a 45degree angle, or anything at all. The important thing is to place that little white drive spline where it will carry the servo arm perpendicular to the pushrod. And truth is, it's often better to have a very slight angle there too.... chuckle.... think I need to draw a picture......... But if you think about it..........

RE: Dual Elevator Servo Question
 

About the "geometry"...........

About the only thing that matters a lot is placing the servo output wheel. What really matters is that the pushrod makes almost a 90degree angle with the horn and with the servo arm. And truth is, you actually don't want a perfect 90degree angle always.

The problem is that you usually need to know ahead of time how far out the horn you're going to wind up with the pushrod connected, and how far out the servo arm for that pushrod connection. And most of us have gotten into the habit of discovering that after we've followed the instructions in the kit/plans/ARF manual.

When you're building your own design or building from plans, if you plan ahead, you can solve the servo location problem without pain. Take a look at what surface throw you want. Look at any other model you already have that's got that throw and you've got your pushrod connect points laid out for you already. Nothing to figure, just duplicate the connect points and you'll know where to place your servo so the output wheel is right where it needs to be to drive those connect points.

BTW, yes, the "up pointing" arm will give slightly different horn movement than the "down pointing" arm will give. But only if you have the pushrod angles at 90degrees. Set up the geometry so that all those 90degree angles aren't 90degrees but are very slightly less than 90degrees and you'll be good to go.

RE: Dual Elevator Servo Question
 

BTW, about those 90degree angles........... and your picture of the two servos on opposite sides of your model...........

The servo on the left side of the aircraft (the left picture) has a very acute angle between the pushrod and the servo arm. And it's got a somewhat obtuse angle between the pushrod and the horn. That's supposed to be bad, right? And the only way to fix it would be to either have a longer horn (not a good solution as it screws up the elevator throw) or to place the servo "into" the stabilizer so the servo wheel is "higher" up and moves the servo arm further up the side of the fuselage.

Well, if you already have a clue where the connect point is on the horn you can simply move that horn "back" so the connect point is placed to give a right angle from line drawn from the elevator axis to the connect point and back along the connecting rod. And you can set the servo arm onto a different spline on the servo wheel so the arm is square to the pushrod (the servo's center doesn't know where the arm is pointing and doesn't really care either, does it).

PS..... when you're building from your own plans, you can do almost anything you want..... elegant solutions are quite rewarding

RE: Dual Elevator Servo Question
 

darock,

Talk of geometry and angles makes my old head hurt!

The illustrations I posted were quick and dirty only to give a visual of what I was trying to explain in the text. No cutouts have been done in the fuselage...in fact, I haven't installed either horizontal or vertical stab as of yet.

Here's a shot of just how far (or not) along the Kaos truly is...

I'm not sure I grasp the concept of mounting the servo at an angle to the horizontal stab but your illustration is self explanatory as far as installation. What again, of the opposite side?

I plan on a pull-pull for the rudder and will be using Sullivan metal control horns on all moveable surfaces. My buddy called last night and said not to forget to join the two elevator halves together even though I'll be using independent servos on each(!?).

Thnks!
PM

RE: Dual Elevator Servo Question
 

graphically:

RE: Dual Elevator Servo Question
 

using any good modern servo--( the new inexpensive JR Sport 125 is simply incredibly good -and very inexpensive )--the best arrangement for a model up to say 10 lbs is done like the setup on old control line airplanes -- a pushrod going to a horn connecting both elevators
done properly with substantial ball ends - the unified movement of both elevators is always identical -irrespective of servo power etc..
This setup is not easy to do but is the best for perfectly matched movement - it duplicates the best full scale aerobatic craft setups .
All of these others we commonly use are simply compromises .
the pic shown in on a 15lb YAK using a CF pushrod and spherical link into a brazed horn which is bolted into each elevator .

RE: Dual Elevator Servo Question
 

BTW, nothing in the rules require you to place the servo's "straight" with the horizontal stab or with anything.

Matter of fact, all those diagrams with the servos showing different movements etc etc..... All of them have the servos oriented with "their tops showing". Nothing in the rules say you gotta do that either. You can place each servo in a hatch on it's side, like some aileron servo's are oriented, so the servo arm is actually at a right angle to the horn. You really should use a ball connection with that, but hey, no rules say you're not supposed to, right?

And TallPaul's pictures show a couple of problems with mounting servo's and that's good stuff to know, BUT......... For example, the middle picture shows how the top pointing arm will give a different movement than the bottom pointing arm. No problem.... don't line the two servos up like that. Nothing in the rules say you gotta have some kind of geometric symmetry to how the two servos are screwed into the airplane, right? Hey, who appreciates where they're in the airplane when it's flying anyway?

In that middle picture, just move one of them up a little, and the other one down a little so their pushrod connection points are CLOSE to the same relative location and you're good to go. And you're not going to need anything extra to drive them.

Hey TallPaul, I'm not arguing that your pictures are wrong because they're perfectly correct. (edit.... Oops... see post below) And they show servo positioning that should be avoided. But like in the middle picture, if you position one servo slightly higher and the other slightly lower, the error is reduced to insignificance. There is going to be a very small difference of motion because one servo arm's arc is downward while the other servo arm's arc is upward, but that too can be solved. For a pattern plane, it'd probably matter more, but for a weekend flyer, it'd be for the flyer to decide if he actually sees any results in how the model flies.

The only fly in the ointment, is that whenever you do have paired servos with their arcs working opposite to the direction the horns arc, you're going to need to know ahead of time where your connect points are going to be so you can line the pushrods up to give relatively close to equal throws. And you're then slightly limited with how much changing you can slap on those connections because when the pushrod angles change, the horns will start to see slightly different movements.

You know how "the rules everybody knows" say that you're supposed to have a right angle between the servo arm and the pushrod? Well, make that angle slightly less than 90degrees when the servo arm is centered, and you're going to cancel out a bunch of the "error" you get from having paired servos that're on a Y-harness. Take TallPaul's third picture and move one of those servos just a bit toward the other one and you'll be good to go.

RE: Dual Elevator Servo Question
 

TallPaul,
Good diagrams. They show some good detail. But there is an error in the 3rd one. (edit... you know, I'd bet that isn't really what you meant, that it's just where you painted the arrows.)
If both servos move their arms 45 degrees, the pushrods will actually move to the right exactly the same distance. They have to if the pushrods are in the same holes in the arms and they both move 45degrees. But it looks like you're showing the black arm displacement as shorter to the right than the red arms displacement. Where the displacement is lost doesn't come at the servo arms, which is what it looks like the picture is showing with the arrows at the servo arms. It comes from the arc of the lower servo moving away from the direct line the pushrod is making to the horn connect point and from the horn's arc moving away as well. The upper servo's arc is complementary with the arc the horn is making so there is little or no lost displacement. The problem with the way those two servos are located is simply that they're located without considering that one servo's arc is complementary to the horn's arc and the other servo's arc is away from the horn's arc.

But there are no rules that say we gotta place the two servos exactly as in picture two or in picture three. Place one a little higher than shown in the diagrams and the other lower than pictured, and you got a pair of servos that give virtually equal movement.

You're right that most pattern guys wouldn't like the solution, but the goal as stated threw that out. And just a little, easily done experimentation will give virtually equal movement.

RE: Dual Elevator Servo Question
 

Tall Paul has it right. If you work it out mathematically or geometrically very carefully you will see why. I suspected something was wrong with my setup and I took careful measurements and worked it out in TurboCAD and sure enough; unequal elevator deflection. That's why I changed it. If the servo arm points up and the elevator horn points down, you will have unequal elevator travel proportional to the amount of deflection of the servo arm and the elevator horn, even if the arms and pushrod are square, and the servo rotates the same number of degrees both directions. Same thing is true if the servo arm is perpendicular to say, an aileron horn, but to a lesser extent.

RE: Dual Elevator Servo Question
 

The difference between the sides is due to the direction the servo arms move.
One side moves up toward the surface, the other side moves down away from it. The difference in the arm's end positions means the surface horns move different amounts since the pushrod length is the same, as is the horn length.
Both arms have to be oriented in the same direction for both surfaces to move the same with equal servo motion.

RE: Dual Elevator Servo Question
 

Darock has the right idea. You just need to vertically stagger the servo positions such that the pushrods leave the wheels and meet the control horns all at 90 degrees so that you avoid any odd differential travel effects. Darock is quite right about the ability to angle the servos to make stuff line up but this sketch may help you see how this can work. as long as the wheels and pushrods all work at right angles the bodies of the servos can point any which way. But this arrangement may be more what you're looking for.

RE: Dual Elevator Servo Question
 

BINGO!

THANK YOU! This is what I was looking for ,(even tho I didn't know what I was looking for, lol)

Measuring across the rear fuselage to get the output connect points of the servos parallel will be simple and the solution is ELEGANT...no geometry, angles or math...Ya-HOO!!!!!!!!!!!!!!

See? I KNEW this was the right forum for my dilemma. Thanks guys, y'all are simply Mah-velous

Pointmagu
Cumberland, MD

RE: Dual Elevator Servo Question
 

You will end up with different elevator deflections between left and right with this setup, unless you have assymmetric servo endpoints. Model it with construction circles and rotate the elevator and horn correspondingly in your CAD software and you will see. The orientation of the servo arm relative to the elevator horn matters. I never gave this much thought until recently when I took a very close look at it. I was surprised at the amount of assymetry when trying to get 47 degrees of elevator deflection.

And even if you program different endpoints with your radio to get the same up and down elevator deflection, you will end up with different surface deflection speeds, since one servo has to rotate farther to achieve the same elevator deflection. Furthermore, they will only match exactly when fully deflected--at intermediate travel points you will find if you measure carefully, the elevators will not match. I noticed the difference immediately when I changed setups--it was like a different airplane.

Admittedly this is being somewhat nitpicky, but the fact that I noticed the difference immediately (and I mean from takeoff onward) on the first flight after the change told me it was well worth the effort. Again, the difference becomes much more noticeable at greater deflections. If you are not going to have more than about 15 or 20 degrees of elevator deflection, you might not notice the effect, but at 3D deflections approaching 50 degrees, the difference will be easily noticeable. During hard pulls or pushes, your plane may roll toward the elevator whose servo had to rotate farther, ignoring the prop yawing/rolling effects.

RE: Dual Elevator Servo Question
 

Yup, that's almost what we've been saying all along. You want assymmetric servo enpoints, I want very slightly "assymmetric" orientation. Both solutions will work within the originally stated goals.

The diagrams are correct. Well, except for one typographical error. So don't place the servos where they are in the diagram. Place them so the movement is virtually equal. Simple. Elegant. Lighter without the extra mixer/balancer/crosser-upper.

BTW, put the servo's on their sides and TAA DAA!!! the disparate displacement (also known in engineering circles as "DD" ) is no longer (or shorter)... ok, the difference in movements goes away.

BTW, in theory, theory and practice are the same. In practice, they are not.

RE: Dual Elevator Servo Question
 

Assymetric servo endpoints cause the rate and intermediate travel problems I mentioned.

If the linkage geometries are mirrored across the fuselage longitudinal axis, then left and right travel and rate will be the same. No problem.

Maybe we are not talking about the same thing. I'm saying having one arm pointing up and the other down will cause you to to have assymentric servo endpoints to get equal elevator travel at max deflection, and you will have the rate and intermediate travel problems I mentioned, even if the pushrods and arms are square with each other. This is not ideal, but may be acceptable depending on mission.


And what do you mean by this?

It appears to contradict itself, since both parts are theoretical, yet unproven.

RE: Dual Elevator Servo Question
 

HEY POINTMAGU......... Good on 'ya mate for offering up the puzzle.

BTW, you want an easier solution? There is one. chuckle..... And it doesn't require any additional cost or time. [8D]

RE: Dual Elevator Servo Question
 

unless both geometry setups are identical - you will not get identical throws and centers
maybe "good enough for you"
the problem lies in the TX-used to rebalance the errors in geometry- If center is changed -- end point changes - if trim offset is changed - then throws are also offset
- you can get close -if that's good enough for you
these setups can suffer as much from internal servo error , as from geometry error -