Hello, I am setting up a large quadcopter.
I’m posting because I need help with something.
The detailed specs are as follows
FC : Pixhawk 6x pro
Motor : XU8 100KV
Prop : 28“(diameter) - 10.1”(pitch)
Frame : 3D printed, Wheelbase 1220mm
TOW : 9.5kg
First of all, the flight log is at the following link.
And here’s where I need your advice.
1. Setting up the Notch Filter
Since ESC telemetry is not available at this time, I want to use an FFT-based Notch Filter. I first tried flying with the FFT enabled, but it seems that the FFT is not tracking the peak noise again. Below is a graph of FTN1.PKAvg.
Below is the FFT result from the IMU log.
Since the FFT doesn’t seem to be tracking the peak noise properly, I’d like to make the following adjustments and would like your feedback.
FFT_MINHZ = 25 //Will moving the notch to a lower frequency have any effect on control latency?
FFT_OPTIONS = 1 //Is it okay to apply the Post Filter Chain FFT in conditions where I am only using FFT based notches without any other notches?
FFT_WINDOW_SIZE = 128 //Is the performance of the STM32H753 processor enough, or would it be better to go higher to capture low frequencies?
INS_HNTCH_BW = 10 //The noise of a large copter is messy, so this bandwidth doesn’t seem like enough, but I’m afraid to increase it because the bandwidth of the notch is proportional to the control delay.
INS_HNTCH_OPTS = 2 //Will this option improve peak noise tracking of notches?
2. PID Parameters
As shown in the plot above, I found significant overshoot in the roll and pitch axes. I’m considering a delay in the rate controller as the cause of this issue. Looking at the logs, it looks like there is a delay of around 100ms in the rate controller, as shown below.
Is this a mechanical characteristic of the large copter? Or is it something that needs to be solved?
3. High Vibration Issue
Below is a vibration plot.
Below are the gyro and accelerometer plots.
The copter experiences low vibrations of 5G or less when hovering, but higher vibrations when decelerating or experiencing a headwind. Will these vibrations have a dangerous effect on the autopilot?
The flight controller is already mounted on silicon dampers, so I’m not sure how to further reduce the vibration here.
4. Low Disk Loading
The above copter has 28” quad rotors and a 10kg takeoff weight, which I think is a bit on the low side for disk loading.
Some people have warned that low disk loading can be catastrophic in wind and other external disturbances. In my opinion, a disk loading of 6 kg/m^2 doesn’t seem that low, but… I’m curious what you think.
Also, can the throttle mix parameters (`ATC_THR_MIX_) solve this problem or is there another solution (besides increasing weight or reducing disk area)?
Thank you for reading this long post.
I’m looking forward to your help.
