I see that these gizmos are pretty common out there for RC / robotics apps and even come with a handful of actuators to play with. I have a nice switchable stick build that might get one of these bolted on to see how well it can handle it. I'm a little concerned about the torque and longevity, but it's definitely worth a try (now that I understand what these are) There's also the "what if" scenario of something getting stuck. They don't give up until they are in position, and I'm not sure what happens if they are physically prevented from getting there for a length of time. Maybe that's what the gear rebuild kits are for .
That's actually the same concern I have, and I'm not sure what's the response. The good thing is that you really cannot force the plate to turn (and thus, do any force to the servo) using the joystick, the only way to block it is to keep it with force into a corner of the plate while the servo tries to turn the plate. In any case, this kind of servo costs 7USD on my country, and I think the price is half that or so in the US, not much to worry about.
If you work something similar, please let us know your experience.
An upgrade I made to the circuit is I took the relay out and used the transistor "switch" to enable directly the second resistance path for the alternate position.
Now I have to check if ledcontrol can have the resolution needed to drive the servo (signal) in itself with no electronics involved in the middle, that would be great, but I would need microseconds resolution and I'm not sure if that's possible, even changing the code.
Thanks all for the feedback.