Stopping a gimbal from wobbling

[SuperSharpShooter]

I have a regular style 3 axis nex5 sized gimbal underneath my hexa, but in anything other than hovering, it's getting a lot of wobble through the yaw motor interface which I can't seem to get rid of. The manufacturer is telling me that on his Naza he drops the gains till it stops doing it. I'd like to avoid buying DJI if possible.

The yaw motor is hard mounted to a carbon plate which is attached to the rails with rubber grommets.

I've just tried lowering my tilt rate D to 26000 and angle P to 210 and D to 7000 to see if that'd fix it, but it didn't much. I'm wondering if any of you have a strategy for dealing with this sort of wobble? Which PIDs should I be changing? Should the Tilt angle PIDS be changed together, or would you keep the D up while lowering the P?

I spent hours last night checking and replacing motors that weren't running smoothly, and the props have been balanced and checked to be running smooth thinking that would fix the issue, but it didn't.

Thanks!

[hogster]

Hi Glenn,

I'm afraid I don't have any specific suggestions for solving your problem, but I will be interested to hear your findings as I've got one of GrouseHouse's gimbals on the way as I type! I went for the lightweight 2-axis version with the optional 3rd yaw axis (my platform already has undercarriage legs).

When you say you see wobbling in the yaw axis do you mean the motor is actually rotating back and forth, or there is some other mechanical slop in the system?

Also presumably the PID values you're talking about are those for the AlexMos controller? Or are you not using one?

Good luck finding the solution to your problem and sorry I can't be of more help.

David

[kinderkram]

Yah, more info, pleeze.

Which controller, firmware, settings, motors?

What does "wobble" mean? Shaking, swinging, vibrating?

[SuperSharpShooter]

It's a viacopter 8 bit alexmos board I'm running with 2.4b7 fw. Powered with 12v reg, and the yaw motor is an ipower 5208 with large hollow shaft and slipring. Running 4008 motors on pitch & roll.

I've pulled the yaw motor apart to make sure the grub screws are holding the two halves of the motor together solidly, but there's a tiny amount of play I can feel but can't get rid of, and on the bench if I tap an arm of the copter it'll bounce around and flex slightly through the yaw motor.

The problem is the whole gimbal seems to swing at the yaw motor interface (you can see the gap between the two halves of the gimbal motor increase and decrease very slightly side to side), and once it starts, it makes the AQ try and compensate, and it feeds upon itself, and I end up with shakey video. In perfect calm conditions it doesn't do it, but it's pretty much never calm here, so I'm looking to tune this to be smooth when actually moving around in a real sort of situation.

I think the gimbal PIDs are good, so I've just been working with the AQ PIDs to see if I can maybe decrease the aggressiveness of the stabilising correction to avoid this feedback oscillation, without decreasing my flight handling characteristics.

Most other 3 axis gimbals just mount to the yaw motor, so I'm just wondering if anyone's been there and found a solution? My previous rctimer 3 axis gimbal had the same sort of induced wobble on this hexa, so this is why I'm looking at the AQ PIDs for now.

[hogster]

Hi Glenn,

If the motor itself is moving around / flexing then maybe that's your problem? Maybe it's worth trying to fashion a second bearing support for the yaw motor to remove that slop?

Otherwise with that slop present, I don't see how you'll ever get rid of the shakes ...?

My 2p

David

[SuperSharpShooter]

I don't know if it's any more slop than any other brushless motor with a reasonable mass attached to it would experience? I thought most other 3 axis gimbals are set up this way and just hung from the yaw motor?

[Max]

Yea I don't see how you'll get rid of mechanical slop by adjusting any software settings. IMHO a good gimbal design does not hang the whole assembly on a motor, regardless of how popular such a shortcut design seems to be. Don't know about yours in particular, but with some of these setups the whole thing is held together by the little circlip on the motor shaft...

-Max

[SuperSharpShooter]

I know what you mean, but I've had this same sort of wobbling motion on gimbals that were mounted on rubber dampers, where the whole gimbal would shake in the same way with the FC PIDs too high. I'm just wanting to know if anyone's had this before, and the strategy for tuning this sort of shake out on AQ.

[Max]

Well, maybe someone will have some tuning tips. But again I think the rubber damper issue is the same problem (a form of mechanical slop). Properly tuned dampers only absorb whatever vibration they need to, but don't let the whole assembly wobble around. The suspension needs to match the weight very precisely, or it is likely to cause more problems than it solves. Sort of like mounting the flight controller on rubber vs. hard standoffs (the former usually being a bad idea). Most gimbal assemblies are sold/designed with no regard for this. The mounts where you can change the number/type of dampeners can help.

3.36¢,
-Max

[Kisssys]

Hard mount the camera to the airframe first and get out the vibrations and tune PID's for as smooth as you can get it. Nothing shows how smooth a ship is better than video. Preferably with a lense with a FOV less than 100 degrees or so it won't hide as much vibration.

Then mount the camera to the gimbal but freeze the camera in position and then tune the mounts until you're at least as good as step one but preferably you have eliminated any visible vibration. Look at the video sitting on the ground without the motors running and just after takeoff. If you have a perceptible difference in clarity of the grass you still have vibration.

Now turn on your gimbal and tune your gimbal PID's. I spent a whole day trying fix a non existent vibration problem because the gimbal would oscillate when off the ground. I finally went to step one and the video was great, it wasn't vibration causing my problem.

Max hit the nail on the head about the dampers, I've gone through many stages of dampers. You might get out vibration but induce a side to side swing especially if camera is not directly below the dampers which of course isn't always possible especially with a wide angle lens and your trying to keep booms out of view.

2.00
Steve

My experience shows the gimbal only fixes my poor flying technique, not a poor ship with wobbles and vibration.