vmsguy

Ok...

I look at airplanes and their specifications, and their recommended throws.. And I see.. for example.. "Aileron: up 1/2 inch, down 3/8 inch."

How does one go about doing that?

I mean.. if one were to use one servo and torque rods, It'd be uneven one side would go up 1/2" while the other side goes down 1/2"

And it'd be the same if you had individual servos connected via a Y-cable.

I know one can play around with end-point adjustment on the radio.. and that'd work on the elevator.. but not ailerons.

A little help please....

exeter_acres

My Feedback: (2)

easiest and most common....

each aileron has its own servo (or more than one on bigger planes) and each servo goes to its own channel.....

can be done mechanically too ie: how you install the control rods to the servo arms, etc. etc., but this is about the easiest way.

MinnFlyer

My Feedback: (4)

Like this

exeter_acres

My Feedback: (2)

Thanks Minn... I knew you would have those!

Perfect

vmsguy

Thanks Mike...

I had a suspicion the offseting of the servo arm would result in such behavior..

but I just couldn't picture it in my head. I'm usually pretty good at these kind of things.

And I didn't want build a test aileron setup.. just to figure it out.


But I do have a question on your picrtures...

On the second picture.. should the grey (aileron neutral) be perpendicular to the aileron? and then blue / green be moved accordingly?

KitBuilder

I learned with the grey more nuetral (vertical) ... then this will translate into lessdifferential since it's just relying on offset at the servo end. Offset at both ends provides more differential... or the capability to increase differential if needed. You cannot adjust installed torque rods but you can adjust servo connection so better to have the angle in the torque rod as well during building.

Lnewqban

While Mike responds:

If the grey horn (aileron neutral) is perpendicular to the center line of the aileron, there is no differential there.

As shown in neutral position, the same angular deflection (number of degrees of rotation) of the servo, will induce more angular deflection of the aileron in the up direction than in the down direction.
Kind of the same effect that the first schematic shows, but in reverse.

Note that this configuration applies to a low wing airplane.
For a high wing plane, the deviation from the perpendicular position for neutral should be on the opposite side.

This differential can be used alone at the aileron horn, or combined with the differential at the servo (in order to multiply the effect).

ARUP

I have jig-built torque rods (TR) with offset. I have a low wing monoplane as an example. The servo is on top of wing and TR horns exit bottom of wing. The TR rod offset for desired differential is at the aileron end of TR. When the servo end of TR is vertical the aileron end horns are leaning fwd. Now putting differential throw on servo adds or negates this depending on which way servo horn set up. Diff increased if pivots for pushrod to TR ahead of servo horn pivot point (where the screw holds the horn onto servo). Good luck

MinnFlyer

My Feedback: (4)

Juan (LNEWQBAN) hit it on the head. I guess I was a little unclear on that.

Picture #1 shows one way of getting differential. Picture #2 shows another way. If you do both, you will get even more than either #1 or #2 by itself.

He was also correct in saying that Picture #2 is for a low wing. His diagram would be correct for a high wing.

vmsguy

Does anybody want to take a stab at the Math?

Ok.. So.. I understand all of Mike's pictures...

Now.... In practice.. There's a lot of variables.

1. Depth (front-to-back) of the aileron. Deeper ailerons would need less degree of change to create the amount of movement at the trailing edge desired.
2. Height of control horn where control rod connects.
3. Distance control horn is shifted front to back. (if at all)
4. Angle of servo horn away from perpendicular
5. Distance from servo horn pivot point control rod connects.
6. Distance between the servo arm and the control horn. An extremely short distance would change the geometry drastically.
7. Is the servo mounted flat or perpendicular to wing surface. Flat mount, some of the servo arm arc would be lost in side-to-side movement, whereas perpendicular mount it would not.

And Mike, before you say it.. I'll say it myself.. I'm overthinking this....

But a person could see that these (and probably some others I haven't thought of) all play a part in the complex geometry of the mechanics involved.

The best way to get the desired results would be to try different combinations and go from there.

But I also have to wonder.. does it really make a difference? Say the instructions require 1/2" up and 3/8" down.. it's a bleepin' 1/8" difference. If I get 3/32" or 5/32".. is it going to be a big deal? What if I'm further off, and manager 5/8" up and 3/8" down.. Then what? Am I really going to see difference as it flies?

And what would that difference be? I'm guessing rolls wouldn't be as axial, they'd be more cork-screwy.

jaka

Hi!
On some planes Yes! On others ...No!
Simple as that!

It's all in the wing profile!

da Rock

With most older plans, that "bleepin' 1/8 difference" suggested by the designer is the amount of difference he found to work for him.

As for the rolls not being axial, the reason a plane needs differential is exactly that, the rolls weren't axial with equal deflection. The sucker exihibited what's called adverse yaw when the ailerons deflected equally up and down.

With cambered airfoils, the aileron moving down creates a uniformly increasing lift with associated drag increase. However, the upward moving aileron actually reduces drag as it starts to move and doesn't decrease lift uniformly at all. Differential helps reduce or erase the problems equal deflection produces with certain airfoils.

Augie11

Adverse yaw effects more than just rolls. It effects turns, particularly on certain airfoils with high winged aircraft. As an example, initiating a left aileron turn results in a left bank and turn but the nose of the plane will fight to point to the right giving a 'funny looking' attitude to the plane as it turns. Differential aileron (more up than down) helps eliminate this. Coordinated rudder will also tend to minimize this (rudder in the same direction as the ailerons).

Lnewqban

Thank you, professor; your schematics make complicated things look crystal clear.[sm=thumbs_up.gif]

Remember that the bleeping 1/8" difference times half span is added area (1/8" x 24" = 3 sq.in.) to the projected area of one half wing that the airstream "sees" and lifts more, while a similar area (3 sq.in.) is subtracted from the projected area of the other half wing that the airstream "sees" and lifts less.

Those little modifications to both airfoils are enough to change the coefficient of lift of each half wing and to induce a roll (which does not have much antagonist force).
Note that the differential, when is good, only works perfectly for certain AOA (cruise AOA normally).

For high AOA’s, while flying slow, the differential may be less effective, since the relative angles of aileron deflection also change.
Again, the airstream “sees” more down than up deflection, and again the half wing with higher camber drags behind and induces the adverse yaw mentioned above.

Like everything else in any airplane, this is just a less than exact compromise, which works most of the times.

MinnFlyer

My Feedback: (4)

You're over thinking this

It doesn't have to be perfect. Just get some differential in there and see how it flies

Radical Departure

My Feedback: (13)

Ditto...

bubba153

Mike, excellent illustration on obtaining mechanical differential throws.

IMO, this is basically an application of what some have termed Ackerman differential. With known angles and lengths for the servo arms (and control horns if other than 90 degrees) the differential can actually be calculated. IIRC it's a cosine function. While the actual math isn't shown, http://www.qmfc.org/school/ackerman.htm illustrates this in a slightly different manner (in terms of pull-pull systems), but it's still applicable here.