Hello everyone,
Sadly, the electric VTOL airframe we finished building today crashed while hovering. The aircraft weighs about 14–15 kg (MTOW 25 kg), and we are using 22×7.4-inch multicopter propellers. I believe the root cause is that I experimentally raised Q_A_ANG_PIT_P and Q_A_ANG_RLL_P from the default 4.5 up to 9. I don’t have video, but in QLOITER mode, after I pushed the sticks forward to translate and then released them, the vehicle became highly unstable in pitch and roll.
What puzzles me is that a similar airframe with an engine installed—slightly heavier at around 19 kg—had no issues even with Q_A_ANG_PIT_P and Q_A_ANG_RLL_P increased to 8. Yet the airframe mentioned above became very unstable. I had assumed Q_A_ANG_PIT_P and Q_A_ANG_RLL_P would only affect RATE.RDes and RATE.PDes, but if you look at the log below, RATE.RDes is around 90 whereas RATE.R reaches about 170. I don’t have a clear understanding of this behavior.
You can review the attached log file. For reference, there are two flights recorded: in the first flight the P gains (Q_A_ANG_PIT_P, Q_A_ANG_RLL_P) were 4.5, and in the second flight they were 9.
Note: In the log you may see Q_A_ANG_PIT_P and Q_A_ANG_RLL_P recorded as 4.5. That’s because after the crash I reverted those parameters from 9 back to the default 4.5.