Hi All,
I've been having quite a difficult time with tuning my Tarot t960 with Pixhawk.
I've had the system flying fairly well and doing missions, but I can't get the rapid oscillation out of the pitch access, roll is fine. The CG is set perfectly and there is little to no vibration coming from the motors & props. I have the main flight pack (6s 22000mAh) strapped to the bottom and pushed slightly back to accommodate the brushless gimbal holding the Sony NEX5 in front of the battery. So I know there is more weight distributed on the forward and back portion of the hex than the roll so I've been trying to adjust my PIDs accordingly.
Here is my issue. I started out with the Roll and Pitch (stabilize) PIDs the same, Roll is rock solid. Than I've been adjusting the 'P' and the 'I' down on the Pitch to reduce the oscillation. As I was adjusting bit by bit the oscillation was slowly going away, until gone. Than I recharged the battery to test it, but after the charge the oscillations returned. I then tried to return the PIDs to where they were originally in order to try something new, but the PIDs will not raise.
I think it has something to do with the battery voltage.
Screen shot 1 is where I started, Screen 2shot is where it got good, Screen shot 3 is directly after the charge and the oscillations returned.
What on Earth am I doing wrong? Please help. (time sensitive)
I have 21 screen shots in total. But I could only load 3. So here's the first one and one's where it started to get stable before the charge.
Replies
What motors and props are you running?
Can you shift the gimbal or battery to get it balanced in the pitch direction? I think this is the thing that will help you most.
Try lowering the rate D for pitch and roll to 0.15 and see if it flies better? I would think that the PIDs for the heavier axis (pitch) would be slightly higher than those for roll. You could try running rate P 0.31 on pitch and 0.29 on roll perhaps if changing D first doesn't help.
Thank you so much.
I'm running 5008 340Kv motors swinging 17.5 inch carbon fiber slow fly props. 6 in total.
I did a full recalibration and auto tune in no wind.. It has helped and adjusted the PIDs. I would have thought the PIDs would have compensated for this, but I seem to be wrong on this assumption. I adjusted the gimbal so that the camera sits closer to the roll motor, just under the spot where the gimbal mounts to the frame and rebalanced perfectly to < 1 cubic millimeter. This works much much better now and the feed back between hex gimbal are gone. The Pixhawk achievies stability very differently from the DJI system. This was never an issue with the DJI system. That system seems to be able to recognize feed back, so this was never an issue. But everything seems to be much better now with some more tweaks.
Sorry I mean 0.015 for rate D.
Autotune?
Yes,
I did a full recalibration and auto tune last night on hour before sunset when there was no wind.. It has helped and adjusted the PIDs. I would have thought the PIDs would have compensated for this, but I seem to be wrong on this assumption. I've been able to isolate the oscillation to somewhere in the gimbal mount that is causing feedback. Thanks for for you interest in this thread.