[NSRCA-discussion] Which servo?

[Bob Richards]

You are confusing two things: Postion, and speed.

Expo affects the position the servo moves to for a given stick movement. Lets say that without expo, the servo moves one degree for every degree the stick is moved. Move the stick 10 degrees and hold it there, the servo will move 10 degrees and stop. 20 degrees = 20 degrees, 30 = 30, etc.  When expo is programmed in, however, the relationship between stick position and servo position is no longer linear. Whe the stick is near centered, the servo will move less than one degree per degree of stick movement. But at the extremes of the stick movement, the servo will move more than 1 degree per degree of stick movement. It makes the plane less touchy when small stick movements are used, but gains the total throw back before the stick moves to the stops.

Servo speed is a different parameter altogether and has nothing to do with expo. It is simply how fast the servo will move to the commanded position. Generally speaking, the faster the servo moves, the better the "feel" will be. However, the fastest servos you can buy may not be the best for all applications. If you have a heavy control surface with lots of deflection, it is possible the fastest servo will oscillate since it does not damp very well. Some of the fastest servos you can buy are those designed for use as a gyro controlled tail rotor. Very little mass or load for the servo to fight, but the speed will help the tail "lock on". A servo that is excessively slow, however, can lead to pilot induced oscillation. The plane is not reacting fast enough, so the pilot moves the stick a little more. When the servo finally catches up, the movement is more than was needed, so the pilot moves the stick in the opposite direction, and the cycle continues. The
 first space shuttle landing with Fred Hayes at the control suffered from this phenomena, he was a little too far in front of the vehicle. The "fly-by-wire" system was a little slow.

As for whether 2.4 ghz improves the "feel", this has to do with how fast the commanded stick position shows up at the servo input. This is most often referred to as "latency". Whether a 2.4ghz system improves this parameter depends on the implementation. If you have a 2.4g module upgrade to an older radio, one that requires the transmitter to be in PPM mode to talk to the module, then you probably won't get any improvement in latency. On the other hand, if you have a radio that is what I call "native 2.4g" then the latency will probably be improved over PPM and most of the older PCM radios. And, the resolution will be also probably be better than most of the computer radios (2048 <> 1024).

So, how fast the servo moves to the postion you want is a function of latency and servo speed.

How far the servo moves is a function of end point adjustment, dual rate and expo.

Resolution is a function of the transmitter analog-digital conversion (And the analog-digital conversion in digital servos, and the resolution of any matchbox in the system.... etc.)

Now that I think of it, I don't remember seeing any specs on the resolution of digital servos.

I hope this helps.

Bob R

--- On Mon, 6/2/08, george w. kennie <geobet at gis.net> wrote:

> I think I'm still having some rationalizatiion issues.
> If I program my servos to move slower and train my fingers
> to speed up their movement at the extremes of stick travel
> it should produce an output somewhat similar to what expo
> gives me (not quite exactly). Not that I could, in all
> probability, learn to do that, but there's lots that I
> hear other guys say that they can feel that always seems to
> escape me. Like, I just converted to 2.4 and I hear guys
> talking how it's so much faster feeling and more
> connected. Well I can't feel any difference. It
> certainly works wonderfully, but so did 72 for me. After
> all, isn't it only converting the frequency of the
> output signal ? Wavelength propagation speed should be the
> same, shouldn't it?
> I still kinda feel that a two-tenth second, high power
> servo, with a one micro sec deadband is not going to feel
> any different to me than a one-tenth second with 10% more
> expo. I just ain't that smart !!!!!
> When I fly something that follows the input command
> instantly, all I do is over-correct and re-over-correct and
> twitch all over the sky and if it's too twitchy I
> ultimately bury it !  The only hope for me in that
> situation is to go to idle and let go of EVERYTHING !!!!
> 
> G.