Hello PX4 community,
I’m new to this gig, so I need a little help debugging issues with a custom quadcopter drone. I’ve tried reading through the px4 setup documentation, but I’m a little fuzzy on a lot of details such as assessing what are good levels of vibration, etc.
- Generic X Quadcopter Arrangement (http://www.helipal.com/storm-drone-4-pro-flying-platform-frame.html)
- FMU: mRo Pixracer R15
- Power supply: 3DR Power Module, APM version 4
- Motors/ESC: Storm 2216-800kv brushless motors + 30A ESC
- GPS: UBlox M8N GPS module
- Battery: 3S 5100 mAh capacity LiPo
- Manual Control: Sony PS4 Joystick/Game Controller
- Damping mount for flight controller: http://www.helipal.com/damping-mount-for-apm-pixhack-fc.html
- Other Setup: Connecting QGCS on my iPhone through RPi mavproxy bridge (using MAVProxy with joystick controller). RPi bridge connects to quadcopter drone through a 915MHz telemetry radio.
- Other details: I mounted the flight controller on the top of the quadcopter frame, as I couldn’t fit it in the lower bay with the damping mount.
Flight Log: https://review.px4.io/plot_app?log=b4a07af4-97cb-4484-95f7-d6c3b6e13e48
Region of Interest: The area of immediate concern regarding the crash is from 13:06 - 13:14. I only had about 8 seconds of flight time before the crash. I am definitely sloppy with the joystick with yaw, but I left the pitch/roll controls untouched until post-crash. After taking off, I experienced a flyaway upon which I took the throttle down so as not to lose it. The copter became severely unstable on the reduced throttle and plummeted to the ground. I appears that it is over-correcting to instability, but it is hard for me to figure out.
From the full graph, you may noticed that I was in Position flight mode for some time, and tried to up the thrust but the copter would not take off. I then transitioned to Manual mode and was able to successfully get lift but very unstable and a flyaway.
I don’t know what the exact problem is at the moment. My only guess is either bad PID tuning with the defaults params, or some kind of vibration issues. I definitely see some potential vibration issues when at idle thrust on the ground in Position mode, specifically centered around 39Hz and 47Hz. I also see other fainter lines that could be problematic, but another strong one at 76/77Hz. However, this is all with props spinning on ground, and a lot of the example graphs I see I’m guessing are showing vibration patterns in the air?
When you look at the Acceleration Power Spectral Density graph at the pointer where I take off and crash, the vibration dynamics are different from those in Position mode, even on the ground, and it looks like what I’m seeing in this graph could be the problem but I don’t have enough experience with this.
Any hints on what you think what is causing the problem? Why would Position mode not allow me to get enough thrust to take off, but Manual mode would? I was initially wanting to be in Position mode to prevent fly-aways, but now that I think about it, if the system is unstable, that wouldn’t really help.
Any hints/ideas would be useful. If it’s clearly a vibration problem, how can I dampen any more given that I mount my controller on a damping mount?