EN CN
YAW axis sloppy with encoders
  • I am using GBM8108 motors with encoders for all axes. ROLL and PITCH are both performing very well, but when I shake the gimbal I am seeing a lot of balance errors from the YAW. I am running the system at 24v, with encoders on all three axes, with power at 255. So there is PLENTY of power to stabilize the YAW, but for some reason I am seeing this error. Any suggestions! Thank you!
  • are you balanced in YAW axis can you tilt the gimbal 45 degrees and yaw stay in place? 255 on power is too much if your motors are getting hot?
    vibrations can be removed in tuning lots of video explaining how to do that !
    http://www.simplebgc.org/pid-tuning.html

  • Yes YAW is balanced. Motors are not getting hot.
  • Could be that your yaw encoder is not linear, i.e. magnet is not at center, or encoder calibration has failed, could be also something else on the basic setup.

    And it could be just a mechanical problem, is everything moving freely?

    If full power goes to GBM8108 it should be getting pretty hot in couple of minutes, if not it could be also broken motor. I am not sure though what you mean " I am seeing a lot of balance errors from the YAW" I understand the power indicator shows full power?

  • Garug, thank you for your response. The encoder setup was purchased from AOPEN3434 with CNC machined housings for the encoder. How can I check if the encoder (AS5048) is linear?

    There is no mechanical problem, everything is balanced properly, and free moving.

    When I say "I am seeing a lot of balancing errors," I mean that the blue ball in the GUI interface is showing me a lot of shake on the YAW axis when disturbed. (I can also see this in the footage). The other axes (PITCH and ROLL) are looking great!

    Although each axis has the same motor and encoder (GBM8108) and full power, the pitch axis seems to have much more torque when manually moving the axis than the others. The roll has a little less torque than pitch, and yaw axis has a little less torque than roll. Any ideas?

    One experiment I did was to set the I value to 0 from the PID values for YAW, and it seemed to perform much better. What are your thoughts?

    Also, what can cause horizon drift with encoders? I thought encoders got rid of that problem?
  • When testing motor tongue, note to push always on the same distance from the motor center axis, If you do and the motors have the same power settings, there should not be any difference.

    It sounds like your yaw is not well tuned, and might be related to gimbal construction also, but without seeing the gimbal I am just guessing.

    Encoders do not help on horizon drift. good calibration and installation of IMU does.
  • sound like yaw in not tuned & balanced properly!
    I suspect the IMU are not fasten properly or in the wrong orientation Please post photos of your setup especially IMU ?
  • I will ask my questions again...

    How can I check if the encoder (AS5048) is linear?
    Can you use a flight controller to get rid of horizon drift with encoder? Or a Frame IMU?
  • May be Garug can answer your linear Question!
    IMU can cause drift especially if it moves orientation after calibration or mounted near EMF interference ! properly mounted IMU should be displayed in GUI when you tilt UP & DOWN should be reflected in the GUI same motion orientation direction.

    also when you move the YAW axis should be displayed in the GUI & rotate in the direction you turn it ! Also calibration issues manifest them self with horizon continuously rotating also the gimbal yaw axis will rotate is this happening ? usually the issue with large gimbals just increase the LPF setting to resolve yaw axis will rotating!
  • To verify the linearity see the encoder raw data on realtime data display.

    16,384 corresponds to 360 degrees, so if you change the angle 90 degrees, the raw value should change 16,384/4 etc.
  • AOPEN, where is low pass filter?
  • RC control section in the GUI