establishing proper linkage ratios and servo requirements

[Mithrandir]

the resolution is what it is..... for the servo....

In all of the large 3D planes I have had, I never had enuff excess servo power that I could give up leverage on the flippers for increased slew rate.... (And I have used multiple 300 oz*in'ish servos)

Don is simply pointing out that he prefers the increased slew rate... faster response of the increased ratio between the servo and the flipper... (100'ish % verses 140'ish %)... at least that is what I got....

I also got... get the geometry equal and proper up front... instead of Programming-subtrimming-ATV'ing bandaids...

It seems to me that absolute resolution is sacrificed using less total servo throw, (for a given elev, ail or rudder throw)... but like many things... it is a "Feel" thing....

One thing I want to test, is the slew rate of the servo vs load..... there is likely an optimum that can be achieved if we know how much a servo is slowed down by the varying loads....

Anyone have any of this data?

[DKjens]

	Quote: Originally Posted by don szczur
	He was like 150 percent on one elevator half and something like 120 percent or less on the other half for up deflection, then vice-versa for the down deflection. The result was one servo trailing the other. Don

If one servo's ATV is -150%/+120%, the other -120%/+150%, control surface throws the same and center position the same with linkage perpendicular to servo arm, one servo will not have to travel 150% while the other 120%. The first servo's subtrim is likely adjusted to around -15%, the other to +15%, so from center position to each extreme, each servo has to travel approximately the same percentage. I'm in no way saying this is a good setup, I'm just saying that I find the quoted statement wrong.

The fact that one surface/servo was trailing the other can very well have been a function in the radio or the programming.

[don szczur]

Thanks for all the comments. DK, you are correct for end to end time but I was referencing just up, then just down (from neutral) where one servo moved 30 percent more than the other when you give up elevator from neutral, so whatever his programming he had in his Tx, it took 30 percent more time to get to the same deflection. Servo speeds are fast so you would not notice it unless you were moving the stick really fast like in a rolling harrier or a pop up or push under like a waterfall. He may have had some varied dual rates as well in there. Again this setup was pretty out of the ordinary but it got me thinking.

Well, perhaps I'll show what I have as a baseline for your reference. Here is my ATV (i.e., endpoints) setups for my Radiowave Extra 300 (set up for both sequence flying and for 3D): Aileron (right wing) 101 left 101 right; Flap (left wing) 100 left, 100 right. Elevator (right side) 120 up, 116 down; (left side) 120 up, 116 down, Rudder 145 left, 150 right. I use the ATV at near max on the rudder because I use the standard Hangar 9 control arms for the rudder servos, and can't get the rudder dontrol horn (practically) closer to the control surface. Now for dual rate I use 70 percent low ailerons for sequence and 100 percent for 3D. Elevator rates, I use 25 percent for sequence and 100 percent dual rate for 3D. Rudder I use 63 percent for sequence rate, 85 percent for spins , and 100 percent for 3D. The expo for rudder is 56, 66, and 85 percent respectively for each rate. Elevator has similar expo spreads between rates. This gives the same feel around neutral but more throw on the edges of the stick during 3D.

[don szczur]

Regarding life of the servo, not sure ATV impacts that- primarily wear is in the neutral point. From my experiences, other than changing out the four 8611s for three 8711s on the rudder, I've been using the same on the Extra since 2006 and will continue to do so- they are great servos.

[mmcconville]

The statement that greater mechanical advantange (i.e. higher ATV values) is easier on the servo is correct. The load on the servo will be less, thus wear is less.
Other issue with lower ATV/lessened mechanical advantag other than reduced resolution and torque is that the centering of the servo is more critical. Meaning, a given roataion angle of the servo results in a larger control surface deflection.
I do understand the reasoning behind running lower being for faster transit time.

[JoeAirPort]

Yep a lower surface speed is the trade off to using the high numers on the ATV's like 150/150. But I still try to max mine out. For 3D throws I need all the help I can get with resolution and I also need the added mechanical advantage for higher torque at the surface. Speed might be more of a precision thing. But I agree if you are too close to the 150/150, the servo may not ever get that far if mixes, sub trim, or trim is used. I see that happen on my throttle all the time and have to go to a different hole on the linkage. One other advantage of using the max (or close) ATVs is that your trim increments at the surface are smaller. That's good for fine tuning (same thing Mike is saying in previous post).

[Mithrandir]

	Quote: Originally Posted by JoeAirPort
	Yep a lower surface speed is the trade off to using the high numers on the ATV's like 150/150. But I still try to max mine out. For 3D throws I need all the help I can get with resolution and I also need the added mechanical advantage for higher torque at the surface. Speed might be more of a precision thing. But I agree if you are too close to the 150/150, the servo may not ever get that far if mixes, sub trim, or trim is used. I see that happen on my throttle all the time and have to go to a different hole on the linkage. One other advantage of using the max (or close) ATVs is that your trim increments at the surface are smaller. That's good for fine tuning (same thing Mike is saying in previous post).

The surface might be lower under NO LOAD!!!
But... under a high aerodynamic load, the surface might be faster with a higher ATV.... without Load vs Speed data.... we don't know the exact numbers.... but taken to the extreme....

A servo using 70% ATV may not be able to move a control surface at all, under an extreme aero-load......but shorten the servo arm and go to 140% ATV.. and now it can move the serface....

We need Load vs Speed data!!!

I am not familiar with other systems, but FUTABA has the ability to stipulate 155% ATV for the total travel of that control, but 140% of the travel is the limit for that specific function....

(Did that make sense??)

[JoeAirPort]

That's a very good point Mith. I never considered that under load it might just be faster due to the higher mechanical advantage.

[don szczur]

This is a really good discussion, and appreciate the points. Brings back some memories of a past life when I used to do that kind of stuff on big UAVs... control surface servo power use (current and voltage), as a function of airspeed, control surface deflection and aircraft g-loading. I did not measure control surface deflection speed though. These tests were done to predict servo/actuator life and reliability.

[Mithrandir]

	Quote: Originally Posted by don szczur
	This is a really good discussion, and appreciate the points. Brings back some memories of a past life when I used to do that kind of stuff on big UAVs... control surface servo power use (current and voltage), as a function of airspeed, control surface deflection and aircraft g-loading. I did not measure control surface deflection speed though. These tests were done to predict servo/actuator life and reliability.

did you find the power consumed by the actuators was linear with speed??? or maybe speed squared???

[don szczur]

I can't remember what the power consumption was under load, but I do remember the servos were worm gear driven and were not a weak link in the system.

Cheers,
Don

[Bill Clark]

its not gonna help right now cause im still a few months away fom my big bird going up but i found out that i can incorporate a load cell into my Eagle tree data logger as well as track servo movement so i intend to intall this in the linkage. i'll pass along the info once i get it

[Mithrandir]

	Quote: Originally Posted by don szczur
	I can't remember what the power consumption was under load, but I do remember the servos were worm gear driven and were not a weak link in the system. Cheers, Don

Just curious... what drone company were you involved with???

[Oilsands]

So for setting up the mechanical linkages for IMAC flying, you want to go with the control horn as close to the surface as possible and the longest servo arm length possible to achieve the max throw you want. Then use the Tx to fine tune. Or do I have this backwards?

[Mithrandir]

	Quote: Originally Posted by Oilsands
	So for setting up the mechanical linkages for IMAC flying, you want to go with the control horn as close to the surface as possible and the longest servo arm length possible to achieve the max throw you want. Then use the Tx to fine tune. Or do I have this backwards?

IMHO, longest horns possible and as high on the ATV you can go with 10%'ish margin on each end of the travel for centering/mixing blah blah lah..... and the shortest control horns on the servo as possible that can still achieve the needed throw.....

Though Don has stated he prefers a zippy'er response so he backs off the ATV and runs a little longer servo arm:control horn ratio....


I'm just sayin'

lol