wheels

Other than actual control throw distance what is the advantage of long servo horns? It seems to me that no matter what the horn lengths on the servo an surface are, equal surface travel means equal mechanical advantage/disadvantage. So can anyone tell me something I'm not seeing here? If not I'll be using the short aluminum servo horns that came with my hitecs and keeping thesurface horns short enough to get the travel I want.

DMcQuinn

My Feedback: (13)

On large planes, the control horns on the surface may be quite far from the C/L of the movable surface (rudder, etc.) because of the thickness of the surface itself. So if you use a "standard" servo arm, you may not be able to get enough throw on the surface even if the control horn on the surface is short.

Dr1Driver

A short servo arm and a long control surface horn will give you more power, but less throw. The opposite will give you more throw but less power. Arms and horns of equal length will give you the rated torque and the same throw as the servo.

Dr.1

bodywerks

My Feedback: (4)

You are correct. A 1" servo arm on a 1" control surface arm (distance from the hinge axis to the pushrod pivot point) is the same as a 1.5 to 1.5 setup, in terms of mechanical advantage. However, I have found that the longer the two arms get, the more play there is in the system. Most of this is due to higher tortional loads on the servo arm ( you can watch it twist a little), and increased flex in the control hor if not designed properly.
Use the shortest control horn you can while still keeping the pivot point directly above the hingeline and being able to get the desired throws without the control horn hitting the stab (this is one of the reasons longer control arms are necessary), and use the shortest servo arm you can use to get the desired throws with MAX ATV and MAX Dual rates. It still typically takes a 1:1 mechanical advantage for 3D throws.

DO NOT USE THE SUPPLIED BLUE ALUMINUM ARMS THAT COME WITH HITEC DIGITALS!!! They fit too loosely on the spline shaft and will only get worse over time. Go with SWB arms or even the included karbonite armes befor you use those blue POS's!

JoeAirPort

My Feedback: (41)

I agree with everything and will try to contribute a little more here. I would use the shortest servo arm that will give you the throws you need. The horn should always be the same length or longer for good mechanical advantage. For example, with a 1 inch servo arm, use a 1.25 inch control horn height (from hinge to pin). That will give you 1.00/1.25 x 60 = 48 degrees of surface travel. That's plenty for just about any plane. A 1 inch arm and a 1.5 inch horn will give you 1/1.50 x 60 = 40 degrees. If you always use close to the full 60 degrees of your servo travel you can get some very good setups. The reason why I use bigger servo arms on my bigger gas planes is so the pushrod clears the wing/stab etc. Also someone mentioned that the surfaces can get pretty thick on giant planes. That forces you into needing a long control horn and a long servo arm.

And yes those Hitech blue arms are a little soft. SWB is my pick these days.

mglavin

My Feedback: (31)

"Mechanical Advantage" is not viable or realized until such time the linkage ratio becomes offset. There is NO such thing as mechanical advantage with 1:1 setup, a 1.5"-1.5" arm setup is the same as 1"-1" and so on; as long as the arms are equal in length the ratio is 1:1 and mechanical advantage is NOT present. However once the linkage ratio is splayed in favor of a longer lever arm "Mechanical Advantage" comes into play....

There is NO distinct advantage to longer arms, they are simply a necessity as noted previously on GS models.

bodywerks

My Feedback: (4)

Agreed. Mechanical advantage is a loose term used to describe such geometries, and everyone seems to use it. I guess you could start coining a term like "mechanical situation"!
Joeairport, I found that the math you speak of only works out if both the control horn and servo arm are swinging on the same plain, like a typical pull-pull rudder setup. But on common aileron setups, however, where the servo arm swings on a perpendicular plain, some of that effective throw is lost to the swinging control rod. For example, on my QQ Yak elevators, I am only getting around 45 degrees with a 10-times measured 1:1 mechanical situation.

JoeAirPort

My Feedback: (41)

bobywerks, I hear ya. I made those statements with ideal setups in mind. But the examples still point out the important concept. i.e. it's still always a good idea try to keep those servo arms short and the control horns long. And of course max out the end points as far as possible.

Red B.

mglavin wrote: This is not entirely correct. When using long servo arms and control horns, the slop in the linkage itself (e.g between horns and links) plays a smaller role than if shorter horns are used.

/Red B.

CrashGaalaas

My Feedback: (8)

I have to agree 100% with Red here, there is less slop in the contols when using the longer control arms on both the servo and the control surface. In addition, the force on the control rods is reduced in proportion to the length of the arms. All things being equal, I use the furthest out holes available on both the servo arm and the control horn.

bodywerks

My Feedback: (4)

And I would have to disagree, somewhat. I can see the logic behind less slop in the control surface if you use a longer control horn and there is even the slightest play between the link and the horn. But a good, new setup, like Dubro ball links between a double-truss horn, should have no play at all, so the point is kinda moot. And, if there is any gear slop or backlash in the servo itself, it will be more apparent with a longer servo arm.
Back in the day, when servos weren't all that great (I'm taling about around the early-mid 90's), the TOC guys were actually switching out their servo arms between precision and freestyle - the shortest possible arm for precision and the longest possible arm for freestyle. They did this to increase the holding power of the servo and to eliminate some of the effects of servo gear slop.

JoeAirPort

My Feedback: (41)

That's what I was thinking too. The long arms/horns reduce the effect of linkage slop but they increase the effect of the servo backlash. With good hardware and titanium gears maybe both are a moot point.

mglavin

My Feedback: (31)

Can you quantify your thoughts with examples of how the "the force on the control rods is reduced in proportion to the length of the arms"?

CrashGaalaas

My Feedback: (8)

I use Futaba S-3004 standard servos, rated at 44 oz-inches of torque. (Ok guys, I'm using 4.8 volt batteries).

If the servo arm is exactly 1 inch long from the center of the servo to the control rod hole and the servo is indeed applying its fully rated load of 44 oz-inches, then the force on the control rod is 44 ounces when the control rod is perpendicular to the servo arm.

If the servo arm is exactly 1/2 inch long from the center of the servo to the control rod hole and the servo is indeed applying its fully rated load of 44 oz-inches, then the force on the control rod is 88 ounces when the control rod is perpendicular to the servo arm.

If the servo arm is exactly 2 inches long from the center of the servo to the control rod hole and the servo is indeed applying its fully rated load of 44 oz-inches, then the force on the control rod is 22 ounces when the control rod is perpendicular to the servo arm.


Keep in mind that as you go to a longer arm on the servo, you will probably be going to an equally longer arm on the control surface. There again the same rules apply:
I am talking about torque applied from the control rod to the elevator
44 ounces at 1 inch on the elevator control horn give 44 ounce-inches of torque
22 ounces at 2 inches on the elevator control horn give 44 ounce-inches of torque
88 ounces at 1/2 inch on the elevator control horn give 44 ounce-inches of torque

Same amount of torque but with much different force on the control rods. (Ever seen those control rods flex?)

JoeAirPort

My Feedback: (41)

Excellent job Crash. I'd say you backed up your words very well there.

One thing I realized the other day was that with a given setup (1 inch arm, 1 inch horn), the force on the push rod increases as you go from 0 to 60 degrees. Again this is an ideal setup, servo axis of rotation and hinge line are parallel. Also this is assuming the required torque stays the same through out the throw. The force has to increase since the distance from the pushrod to the axis of rotatation decreases. Now if the required torque increases as it usually does at higher throws, the force gets even higher. Lots of things going on in these little planes.

mglavin

My Feedback: (31)

I concur, F=T/D. However the Forces on the control rods are either Compression and or Tension and they are not constants either side of center. Force is realized at the pivot point of the servo arm and is as noted a factor of the servo arm length and the servos “Torque� specification. Servo specifications are garnered with 1� arms, hence the term oz-in.

Again I concur, T= FxD. However “Torque� is not applied from the control rod to the elevator. “Force� is introduced to the control arm of the elevator as gleaned from the servo arm. Torque is realized at the elevator because of the relationship of the control arm to the hinge line or pivot point. Since Torque is a twisting force it requires a pivot point. 44ozs Force introduced to a 1� control arm from hinge center-line nets 44oz-in Torque at the elevators hinge line.

FWIW: the servo “Torque� rating is a constant No matter what length arm is utilized.

Your original comment "the force on the control rods is reduced in proportion to the length of the arms" in simple terms is accurate, but when you reduce arm length(s) you increase the FORCEs introduced into the control rod, while increasing arm length(s) decreases Forces introduced into the control rod.

There are many other variables involved that cause and affect the output numbers. Such as but not limited to, vertical offset between the servo arm axis in relation to the control arm axis, horizontal offset between the servo and control arm axis, diminished or increased surface deflection and more.

“All things being equal, I use the furthest out holes available on both the servo arm and the control horn�

The results you’ll realize with said implementation will be contrary to your statement. All things cannot or will not be equal without physically relocating variables as noted above. Additionally the farther you migrate from shorter servo arms the more disparaging the difference becomes with regard to compression and tension at the control rod/pushrod.

CrashGaalaas

My Feedback: (8)

Huh?

bodywerks

My Feedback: (4)

I switched from 5945's to 5955's on my QQ Yak, and lost 1/4" of play in both directions at the counterbalance with the same geometry setup, so you have a point there...

CrashGaalaas

My Feedback: (8)

Can someone explain what "servo backlash" is?

JoeAirPort

My Feedback: (41)

It's just slop in the servo due to wear on the gears. If you wiggle the servo arm back and forth you can feel it (if it's worn).

CrashGaalaas

My Feedback: (8)

Yes, I can see where longer arms would magnify the slop of any backlash in a servo.
But that backlash is limited to a certain number of degrees of rotation of the servo.

A control rod on a longer arm will travel further through the same degrees of backlash than will a control rod on a shorter arm.
1 inch arm has 1/4 inch slop
1/2 inch arm would have 1/8 inch slop

But remember that if you change the length of the servo arm you will also change the length of the control horn.

In the end you still end up with the same amount of servo backlash (degrees of servo slop) reflected in the control surface. And if you shortened the arms, you have increased the load on the control rods.

CrashGaalaas

My Feedback: (8)

And just so I can thoroughly confuse everyone, here is what I was thinking as I had trouble falling asleep last night:

The servo is not providing a constant level of torque. The figures we quote are merely the max torque available. ie my Futaba s-3004 has 44 ounce-inches at 4.8 volts. But what the servo really does is use whatever torque it need to hold a position matching that coming from the transmitter.

The real load on the servos is produced by air pressure acting on a control surface (aileron, rudder, elevator, etc.) during flight. This load is a torque around the axis of the hinges and gets transferred as a force through the control horn and control rod to the servo. The servo applies exactly enough torque to hold the servo in the correct position.

I don't have even a wild a** guess as to what the loads are like during a real flight, but can imagine they are very dynamic.

JoeAirPort

My Feedback: (41)

Crash, that all makes sense to me. You can only assume a certain torque load generated by the control surface and work your way back to the servo (through the linkage). It can't exceed the servo's rated max torque.

bodywerks

My Feedback: (4)

the 5945's were new servos, so it wasn't wear, just sloppy gear mesh. some of this slop is also a result of the loose deadband of the servo. If I had a programmer, I could take some of it out, but my servos would draw more current at idle from trying harder to maintain their center.

bodywerks

My Feedback: (4)

This is one reason why theTOC guys use to have to put shorter servo arms on all their servos for precision flight - less backlash and slop, more force to the control surface, and better centering, but less throws.