Why the cross-shaped servo arms?

[Haoyang Wang]

One of the mysteries in life is the cross-shaped servo arm. Every servo comes with it, but why would you need 4 arms per servo? Once in a while it is the only kind I have at hand, then I would have to cut off the other 3 arms I don't use.

And KST actually has one with 6 arms!

KST Plastic 25 Spline Servo Arm Set

This is the replacement servo arm set for the KST 515 servos. Fits KST DS515/715/525/725/1509/2509/3509MG and BLS815/805X/915/905X/159/259

alofthobbies.com

 

[Haoyang Wang]

Ah, I remember that once I attached the cross under the mount of a little camera, to rotate the camera directly with the servo. OK, so the cross can form a stable base.

 

[Konrad]

One of the mysteries in life is the cross-shaped servo arm. Every servo comes with it, but why would you need 4 arms per servo? Once in a while it is the only kind I have at hand, then I would have to cut off the other 3 arms I don't use.

And KST actually has one with 6 arms!

KST Plastic 25 Spline Servo Arm Set

This is the replacement servo arm set for the KST 515 servos. Fits KST DS515/715/525/725/1509/2509/3509MG and BLS815/805X/915/905X/159/259

alofthobbies.com

They and the odd spline count are there to allow you to rig the pushrods 90° (tangent) to the arm's arc when centered. This is to avoid unwanted differential throw.

On the flip side it allows you to use differential throws to correct control surface horn/spoon misalignments with the surface hinge.

Or as I like to use them to match the servo output with the control surface output. For example most of my gliders have a something like a 2:1 aileron differential. With the spline and multi arms I can find the servo arm position that allows this while maintaining the servos full resolution and torque in both surface directions. Use of the radios programing to set this differential has a lot of limitations when looked at this through the lens of servo performance. Yes, I do use the radio's differential to fine tune this mechanicaly derived differential.

By the way I try to use differential for the elevator. You might have seen in one of my glider threads where I said it took me 4 hours to set up 2 servos. Some of this was matching the servo motion to the surface motion.

Now when flipping the servos between sides like the flaps and ailerons this flexibility can be a bit of a pain with the single servo arm. As the odd spline count messes up arm's placement when flipping it 180° to the other side of the servo.

After you find the correct servo arm to spline relationship you are expected to trim off the other arms. This helps with space concerns and also minimizes the other servo leads getting tangled in the arms and pull the lead out of the RX.

As to servo arm stability the round one is much stiffer and stable than any of the cruciform style arms. So if using the servo in high torque applications try to use the round servo output wheel.

 

[rlocke]

it aids pull-pull which i really like.

 

[Haoyang Wang]

Ah, since 25 and 4 are relative prime, with the 4 arms you can have 25 X 4 = 100 different angles.

25 and 6 are also relative prime, so the 6-armed version gives you even more possibilities.

Now when flipping the servos between sides like the flaps and ailerons this flexibility can be a bit of a pain with the single servo arm. As the odd spline count messes up arm's placement when flipping it 180° to the other side of the servo.

Different servos can also have the splines aligned differently. It’s frustrating to align the arms of 2 servos.

Thank you for the clear explanations.

By the way, do you calibrate your transmitters to have all stick positions centered at 1500?

 

[Haoyang Wang]

it aids pull-pull which i really like.

So that you get to pick which pair of arms (out of 4 or 6) to use? That’s nifty.

 

[Konrad]

Yep, 1500 the starting point.

In the old days with the square servo output the servo feedback pot needed to be aligned at the factory with the output drive. Now with the splines and multiple arms, you the end user have to do this, if you feel you need that kind of mechanical precision. This saves the servo manufacture the manpower to perform this alignment. As most models are set up, where this isn't noticeable, there is good reason not to spend the manpower doing this. The Pull-Pull and multi servo per surface are good examples where this is needed.

 

[Konrad]

By the way here is how Futaba tried to explain this relationship.

 

[TyFlies]

Thanks for the explanation. It was a mystery to me too. Make sense.

 

[Wayne]

And here I was thinking it was for old school aileron differential.

I usually grab the single arm units and do a little dance if they line up the way I need them too.

 

[Konrad]

And here I was thinking it was for old school aileron differential.

I usually grab the single arm units and do a little dance if they line up the way I need them too.

It is for that.

I set the servo arm for the mid throw position. When bringing the surface back into profile it will look like what you have been doing all a long. Understanding what is really going on, you are freed from the frustration that was that single arm approach.

 

[Wayne]

I'm guessing this came about as a result of the old square drive flanges. Those may have been a bit limiting.

 

[Konrad]

I'm guessing this came about as a result of the old square drive flanges. Those may have been a bit limiting.

Yes, as the servo torque was increasing the servo arms were starting to split at the hub (corners). But I suspect the real reason was so that the servo OEM didn’t have to index the feedback pot to the output shaft.