CRASH: Hexacopter / PX4 - Please, help me find the cause

Hey guys! On my first flight I had a crash. Would be very nice if someone could kindly help me find what happened. Thanks in advance!
I was flying in PosHold mode when the crash happened. I’m sending a link to the video showing the failure as well the log file.

Video: crash moment

Log file

  • Pixhawk PX4
  • GARTT ML 5010 300KV
  • Hobbywing XRotor 40A (running BLHeli_S fw)
  • 20x5.5 CF propellers

Log File C:\Users\Dolence\AppData\Local\Temp\tmpEC1F.tmp.log
Size (kb) 7479.5107421875
No of lines 86278
Duration 0:04:15
Vehicletype ArduCopter
Firmware Version V4.0.3
Firmware Hash ffd08628
Hardware Type
Free Mem 0
Skipped Lines 0
Test: Autotune = UNKNOWN - No ATUN log data
Test: Brownout = GOOD -
Test: Compass = FAIL - Large change in mag_field (236.08%)
Max mag field length (569.05) > recommended (550.00)

Test: Dupe Log Data = GOOD -
Test: Empty = GOOD -
Test: Event/Failsafe = FAIL - ERRs found: FS_THR CRASH
Test: GPS = GOOD -
Test: IMU Mismatch = GOOD - (Mismatch: 0.53, WARN: 0.75, FAIL: 1.50)
Test: Motor Balance = GOOD - Motor channel averages = [1172, 1155, 1168, 1158, 1140, 1186]
Average motor output = 1163
Difference between min and max motor averages = 46
Test: NaNs = FAIL - Found NaN in CTUN.TAlt
Found NaN in CTUN.DSAlt

Test: OpticalFlow = FAIL - FAIL: no optical flow data

Test: Parameters = FAIL - ‘MAG_ENABLE’ not found
Test: PM = GOOD -
Test: Pitch/Roll = UNKNOWN - ‘BarAlt’
Test: Thrust = GOOD -
Test: VCC = UNKNOWN - No CURR log data

Upload your bin file…

Hi! The bin file is linked on the topic. Let me know if you can’t download it

Ok I will check it…thanks

1 Like

Did you disarm the quad in mid air?

No. It started wobbling (I’ve linked a video from the crash too) so I tried to land it.

Looks like badly tuned. Did you auto tune it?

I’m on stock PIDs and it’s a very large copter. First flight, didn’t tuned yet.

Test: Compass = FAIL - Large change in mag_field (236.08%)

Could it be because I did the compass calibration inside my house? I have some steel frame walls.

Do a compass recalibration and first time try in loiter mode and notice any issues.

Then do auto tune.

I have flown such big hexa quads with default settings and worked for me.

1 Like

Ok, I will try it. Waiting for a new GPS/Compass antenna support to arrive. Luckly it’s the only thing broken.

From the logs i do not see the mag as the cause of your crash. The hexa in my opinion is simply not tuned enough to fly. Do not take off in loiter with those pids, it will crash again.
Possibly read all this thread:

Towards the end you’ll find a spreadsheet wich will help you with defaults.

2 Likes

Thank you very much, I’m reading carefully. I wil ltry this and report back.

Ok, take your time :slight_smile:

1 Like

There’s a physical yaw bias, your CW Motors 6, 3 and 1 are working harder, fighting against CCW motors. Slightly twisted motor mounts on round arms is the usual suspect.

Get well acquainted with that tuning guide that Corrado listed

Set these:
ACRO_YAW_P,2
ATC_ACCEL_P_MAX,52100
ATC_ACCEL_R_MAX,52100
ATC_ACCEL_Y_MAX,18000
ATC_RAT_PIT_FLTD,11.5
ATC_RAT_PIT_FLTE,0
ATC_RAT_PIT_FLTT,11.5
ATC_RAT_RLL_FLTD,11.5
ATC_RAT_RLL_FLTE,0
ATC_RAT_RLL_FLTT,11.5
ATC_RAT_YAW_FLTD,0
ATC_RAT_YAW_FLTE,2
ATC_RAT_YAW_FLTT,11.5
ATC_THR_MIX_MAN,0.1
MOT_THST_EXPO,0.75
FENCE_ENABLE,1
FENCE_TYPE,3
BATT_ARM_VOLT,22.1
BATT_CRT_VOLT,21
BATT_LOW_VOLT,21.6
MOT_BAT_VOLT_MAX,25.2
MOT_BAT_VOLT_MIN,19.8
PSC_ACCZ_I,0.26
PSC_ACCZ_P,0.13
ATC_RAT_RLL_P,0.10
ATC_RAT_RLL_I,0.10
ATC_RAT_RLL_D,0.005
ATC_RAT_PIT_P,0.10
ATC_RAT_PIT_I,0.10
ATC_RAT_PIT_D,0.005

The last few, ATC_RAT_PIT and ATC_RAT_RLL - you might need to increase the P and I terms a little to 0.12 or 0.15 in order to get a steady flight - test in Stabilize first, just a couple of meters high out of ground effect and prop wash.
If that’s working OK do some basic gentle test flights in AltHold and you might be able to move onto Autotune.

EDIT: if you copy/paste all those params into notepad then save as “Williams-initial-settings.param” you can apply that in MissionPlanner without looking for them all and mistyping some things.

I will run a test flight right now. Wish me luck!

1 Like

Did you test it? How did it go?

Yes, I did! Well, I had some windy day here and wasn’t confident enough to flight in stabilize for much time.
I raised ATC_RAT_PIT and ATC_RAT_RLL to 0.15 and tried alt_hold and pos_hold modes. Looking to DES_YAW x YAW they lines overlap perfectly, but the same isn’t true for ROLL and PITCH. I have a large DSLR gimbal attached to the frame. THis gimbal have some large blue silicone dampers. Would this be an issue?

Flight time was very very short. I had a battery failsafe (3.5V) and it crashed again when descending. Without all the gimbal and landing gear weight I had around 20-22 min flight time. Is there any way to get total time actually flying from TLOG? I think it will not be possible to achieve a successful auto tune on any axis within the actual flight time. It wasn’t even close to 10 min.

image

Hi, the recommended way to run AT is always make your vehicle as light as possible and especially without the gimbal and moving parts.
GL

1 Like

Really? I thought I had to do it as close as possible as it will be when done or it would be affected later. I don’t know why I thought this.

1 Like