EN CN
Simple BGC 2.2b2 Roll twitching jello on gopro gimbal FY680?
  • Hey guys,

    Still pretty new to tuning gimbals specially with BGC just after a little help pointing me in the right direction.

    I'm getting awkward twitches or jellow in which i think is my Roll and maybe a little in the pitch? but i'm not sure if it is because the gimbal is getting to much vibration from the hexcopter? or if my PID settings are way off?.

    I have balanced my Props and have balanced my motors as good as i could.
    my concern is that my frame being a fy680 is a foldable frame which perhaps might be the cause of all my vibrations?

    Anyway ill post all my details bellow and if anyone could give me some advice to maybe tuning my PID setting a littler better but i have been playing around with them for a few weeks now and can't seam to get them much better. I also have some sample footage i can upload if that helps anyone?

    Hexcopter
    Frame: fy680
    Motors: sunnysky v3508-16 700kv
    props: 1245 props
    flight board: Naza v2
    Gimbal: http://www.ebay.com.au/itm/New-Brushless-Gopro3-Camera-Mount-Gimbal-with-Motor-Controller-ctx-/200935710021?pt=AU_Toys_Hobbies_Radio_Controlled_Vehicles&hash=item2ec8b39d45&_uhb=1
    gimbal Firmware: BGC 2.2b2

    PID Settings:

    P I D Power
    Roll 20 .06 8 80
    Pitch 14 .11 6 80

  • Are you getting getting awkward twitches if you move the frame by hand (motors not running) ?

    Compared motors running video to motors not running is the video blurred? If not, then vibration level should be quite ok. If it is blurred, you need to improve vibration dampening. (and/or balancing of the propellers/motors)

    Are all axels moving totally freely? i.e. bearings ok and cables not limiting the normal movement.

    Is motor pole count correct in GUI?

    I would suggest updating to 2.3 FW or maybe even 2.4, (I am using 2.4 but many report easier tuning with 2.3)

    Are gimbal motors getting hot, if not I would try more power at least for roll and maybe higher D.
  • Yes i did get some twitches when moving it by hand but only when moving to copter forward and backward i noticed the camera got some fast shakes on the pitch axis? the roll seam fairly good.

    Axels are moving as free as possible the only cable going to the moving gimbal is a Fpv lead to the gopro camera and the cables for the sensor board.

    I believe my motor have 14 poles so yep set that in GUI.

    ok might try updating to 2.3

    motors aren't getting hot at all but i do start getting more twitches when i turn the power up. will try increasing D.


    i did cut together a quick example of the gimbal in the air might help a little to determine if my shakes are gimbal or vibration related.

    Thanks again for the advice!

  • By looking the video I think the main problem is vibration dampening. By looking the gimbal: the vibration dampers should be compressed, not pulled like they are on the Ebay picture. (look for example Tarot or DJI gimbals)

    Also; are the camera and IMU firmly installed on that gimbal? Any loose play will result vibrations
  • That was my original thoughts also so had originally modified the gimbals to the vibration dampers where compress but did not see much results, perhaps more vibration dampeners would this help?

    Camera is fairly secure with hardly any movement and IMU is stuck on using double sided tape?
  • OK so managed to change the dampeners back around so they were compressing and up the D settings to 11 and also took the gopro of medium view and spot on much better results from the video.

    will however go for another fly tomorrow when there is a bit more light and check out the footage again and let you know how i go.

    Thanks heaps for you help mate! really appreciate the advice!
  • You are welcome.

    About the vibration dampening still; I would rather try softer dampers. I do not know the ones you are using, but it looks on video that softer could be possibly better (but maybe this is also because they where not compressed). I like the Tarot dampers for this size of gimbal.

    (Softer dampers move the vibrations to lower frequencies.)
  • Mate you were spot on with your recommendation with the softer dampers. i bought the tarot clips and top dampening plate of the t-2d so i could use the clips on my gimbal to attach to my tarot FY680 frame. Anyway yestertday when i swapped over my gimbal i also swapped over to the tarot dampers and they are much soft and seam to being doing the job like you said!

    Will try fly today when there's a bit more sun to see if there's anymore vibrations or tuning i can adjust.
  • Went for a quick fly and here is a snippet of the footage much better then it originally was but still getting what i think is some tinny bits of jello though the video.

    it was also a bit windier today do you think its vibration again or from the gimbal?


  • the wind shakes the multi copter a lot and these are lower frequency movements that the dampening can take, so they will go all the way to the gimbal. Also the gusts are probably affecting the camera directly too. This is a difficult situation for any gimbal. I think PID tuning could possibly make it a bit better, but I would first try what you now have on not so windy conditions. I think the result will be good.

    this is my GoPro gimbal (heavy find at 1:30)



    This is a very low profile gimbal as is the multi copter, so it takes the wind relatively well. What I still like to have on this is the 3rd axis, to get rid of the Yaw movements.
  • Wow, Garug, that was really stable, even with such a high wind...

    I hope I can get rid of my extreme-situation vibrations on roll for my gimbal soon... :D
  • Thanks

    The flight controller (Naza) had much harder time with the wind than the gimbal ;-)

    On the gimbal, the mass evenly distributed around the vibration dampening on every direction, that helps a lot and also there is actual dampening not only the rubber 'springs'
  • Garug are you running the board FC pinout to your Naza?
  • No, I have been thinking to do that and to get also the controls to work, but have been waiting for support for SBUS, 32 bit boards should have that, will test on next week.