EN CN
Horizon drifting in direction of heli travel
-
I'm really loving this 32bit version. So many incredible features. The second IMU adds much more stability as well.
My only problem I can't figure out is why my roll is leaning in the direction the copter is traveling. I did the full 6 point calibration on both sensors and both sensors are lined up with the motor axis's. I've adjusted gyro thrust in ten increments from default 100 to 140. There hasn't visually been a difference in my tests. One thing I've noticed is when I do the auto PID adjustment, the P, and D are sky high. I used to use 8 to 10 at most for P. With the auto PID, its going all the way up to 150 plus on P and D. Roll I was 0.46. I tried lowering that to 0.25 and same results. I've tried disconnecting the frame IMU and still same exact symptom.
Here is a video. I'm getting to the point where I may be of some help to others. I just signed up on the forum and hope to pay it forward when I see something I know about.
-
How do the DJI zenmuse gimbals not have this? My 8bit didn't really seem to do it, as much.
BTW, thanks. I hear you are the guru of gimbals. At least I know the affect isn't just me.
With Gyro Thrust, the higher the number, the more its taking the gyro into account instead of acceleratometer right? How high can the Gyro Thrust go? There doesn't seem to be a bracket. With the single rotor heli hitting 45 to 50mpg, I'd like to rely more on the gyro.
The affect can be see with my just hand holding and running back and forth. -
Gyros are not affected by the accelerations (much) , so yes, more gyro trust should help, but at some point the gyro drift starts to play a role.
BTW, it is possible to control the gyro trust with RC, see the adjustable parameter.
Going fast, will require a lot of the gimbal construction. The smaller and more solid, the better.
-
What is the highest gyro thrust the software will allow. Understanding real physical spinning gyros I know what you mean. I could switch to a higher gyro thrust before starting a fast pass, then switch back so that the accelerometer can find where down is again properly.
I've gone 45mph so far with no problem. I just get the drift. -
Nothing wrong on high PID values if the gimbal works good and usually high values are indication of good and solid gimbal. If the gimbal vibrates after auto PID tuning, lowering the values some 10 to 20% usually solves the issue.
It also depends of motors, gimbal size and power settings. Maybe one thing to ensure is that Power is high enough, but if the gimbal works good, it is high enough. -
It's pretty solid. It works way better than the 8 bit did. I can make extremely fast movements and the response is so fast I never see drift. With the 8 bit if I moved it fast, it didn't react fast enough.
I'm using WASP cages with 8017 motors. 25mm tubes.
My last problem is this jello from the pitch axis. I'm waiting for my Bluetooth adapter. On the 8bit I put the imu on the gimbal mount with motors off and used the alexmos to tweak vibrations out. Once I was go on the airframe then I moved the imu back for gimbal operation and adjusted PIds -
Vibration is probably because vibration dampers. If the gimbal starts easily ringing, when knocking it, then the PID could be too high and contribute to the vibration. In this case you could lower PID and/or use filters.
Adaptive PID tuning can be useful, but I would first tune the gimbal so that it operates on normal conditions well without it.
Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!