X8 - Crash analysis

ArduCopter Copter 4.5
1

Hello everyone,

Today, I experienced a crash with my X8 multicopter. The setup is as follows:

  • Weight: 67 kg
  • Flight Controller: Narin FC
  • Firmware: ArduCopter v4.5.7

During the flight, I noticed that the GPS reported 38 satellites, which is excellent in terms of signal quantity. However, I’m wondering if the GPS update rate might have been too high, potentially affecting the heading alignment. It’s possible the system hadn’t reached ideal heading accuracy before takeoff.

The takeoff was performed in LOITER mode, which relies heavily on a stable and accurate heading reference. I suspect that any deviation in heading data—possibly from the GPS—may have led the aircraft to tilt and lose stability during takeoff.

One thing that still puzzles me is the **difference between ATT.Des and ATT. actual attitude. I would appreciate your insights on what might have caused this divergence.

log file

Looking forward to hearing your thoughts.
Thanks in advance!

2

It was probably all that water in The Sea of Japan that did it - just kidding :slight_smile:

The trouble is the log only has a very small amount of data. It would be handy to see a log without a crash just to check on more. I suspect there is some more tuning to do and you may have some oscillations and poor attitude control happening when the copter does get airborne.

Large copters can easily tip over during takeoff if the GPS position does wander and landing gear snags on the ground.
In your case there may have been a wandering position, but the problem is not really with heading in particular or the GPS update rate.

These are ESSENTIAL for safety and ensuring GPS is working properly before you can arm:

BATT_FS_CRT_ACT,1
BATT_FS_LOW_ACT,2
FENCE_ENABLE,1

And these will help with a more positive takeoff:

PILOT_THR_BHV,7
PILOT_TKOFF_ALT,200
WP_NAVALT_MIN,2

You can disable these since they are not working as planned:

FFT_ENABLE,0
INS_HNTCH_ENABLE,0

You can probably reduce the thrust expo:
MOT_THST_EXPO,0.80

Set these to gather more data in the next flight.

INS_LOG_BAT_MASK,1
INS_LOG_BAT_OPT,4
LOG_BITMASK,180222
2 Likes
3

Thank you for your response, Shawn.

I’ve uploaded several additional .log files from previous flights for your review:

  1. Notch filter (Throttle-based) - Test 1
  2. Notch filter (Throttle-based) - Test 2
  3. Auto flight

During the flights, the flight controller was installed without a damper. Copter had significant vibrations, especially on the Z axis. You can see this clearly in the logs.

Please refer to the logs for more detailed information. I appreciate your time and support.

4

I suspect prop balance and maybe wobbly landing gear is a problem. Yaw is showing oscillations before takeoff. I think this copter will be helped by some more tuning, or running Autotune.
Set ATC_RAT_YAW_I,0.05
but upgrading to 4.6 gives some extra parameters to reduce PIDs during takeoff to avoid oscillations, but still have good/high PIDs during flight:

ATC_LAND_P_MULT,0.8
ATC_LAND_R_MULT,0.8
ATC_LAND_Y_MULT,0.8

Otherwise the log is not much use to me regarding the harmonic notch filter.
You can PM a .bin log to me if you are worried about your location being public.

5

Or, Sun spots… lol

6

well - there has been sun spots, so you are probably correct!

7

Thank you for your support. Since the damage isn’t too severe, I’ll repair the hardware and set it up according to your suggestions. After flying again, I’ll share the bin files with you. Thanks again Shawn!