Crash log review - Copter spinning out of control

Hi everyone, I hope you can help me with this one =/

I build an aluminum frame X octocopter with aprox. 30 kg with aprox 40% power to weigh ratio with a 3º degree inward tilt on all motors, here is the basic layout:

(All motors are evenly spaced from center, despite what my paint skills says)

Here is the components overview:

  • 8 x P60X tmotor 170kV @ 12S
  • 22 inches t motor foldable props
  • Pixhawk 6x
  • 2 x external gps
  • 2 x 22ah 6s 30c batteries

After carefully bench testing the frame at full throttle and long runs at flyable thrust, I was confident to test the drone on the field.
I got a solid compass calibration on all compasses and gave 1 and 2 priorities to external compasses. I did basic attitude and yaw tests, and all seemed well.

In alt hold I took off to about 5m with rock solid stability and no errors from event messages but then I saw that the drone started spinning counterclockwise with increasing speed and was out of control in a couple of seconds.
I tried to stop the spinning with yaw inputs, but the spinning got worse. In a last attempt to stop the spinning I changed FM to RTL but with no effects on spinning. However, the drone became even more instable with increasing oscillations in roll.

When the oscillations in roll went well past 45 degrees, I disarmed the drone to not harm anything or anybody.

Luckily, I “only” managed to damage the frame and props. I looked at the log and didn’t see anything out of the ordinary. Either something extremely basic went wrong or it was my lack of experience with metal frame copters.

Somehow, I managed to break the SD card upon removing it from FC (geez). Later I’ll try to recover the parameters file.

Can you guys help me find the cause of this spinning issue?

See how half of the motors (1,2,7,8) go to very low output and half (3,4,5,6 CCW) go to high output:

This usually indicates a motor mount or frame twist and CCW motors are trying to compensate for CW motors being twisted.
It also could happen if the spin directions are wrong.

As soon as motor outputs needed to be varied for attitude control, yaw control is lost, attitude control is very poor because half the motors are already at their minimum, and the other half are driven to near their maximum, leaving no fine control whatsoever.

image

In MissionPlanner motor test do “All in sequence” and the motors should operate in this order:
1,3,8,4,2… and so on.
Carefully check whether they are correctly doing clockwise or count-clockwise.

Aluminium frames have a habit of flexing and enhancing vibrations.
The frame can be “flat” with no need of any upswept angle on the arms or motors → that doesnt really help stability or anything else.

EDIT:
fixed motor order typo

Thanks for the review xfacta !

Regarding motor numbering and direction they were correct.

What I can theorize was that during compass calibration I misaligned one of the ccw motors by a few degrees. Could this be the cause of the crash?

From our test we did not found any considerable flex or twisting motion in the frame.

When I was reviewing the log I was expecting to see an increasing yaw error but it looked norma and definitely not increasing.

Here is the file the parameters file:
parameters.txt (28.3 KB)

Furthermore, no alarms popped until crashed.