Please help-RTL crash w/ log & video

Log Analysis -- deprecated Copter Logs -- deprecated
1

I began flying like normal but from the begining something felt a little strange. When in Poshold it would initially drift to the right when trying to move forward then it would correct itself. It was a little windy so i figured that was the problem. I was using my phone for telemetry and around 10 mins I started to get frequent disconnection from the telemetry. I started feeling uncomfortable flying anymore so I directed it back towards me and hit RTL (where the video starts). RTL seemed great so I didn’t pay much mind to it. After descending to a few feet above landing area it started to quickly drift to right. I gave input but b/w orientation and panic I couldn’t save it. Auto Analysis of the log says compass failed/high mag field and vcc problem. I cant figure out when mag field was high or did my compass fail? did vcc create a problem? I burnt an ESC and motor i assume b/c in the trees it kept spinning and got too hot. My GPS mount bent and I noticed there is bare wire showing on the plug of the mag. but not sure if that happen after crash. I plugged in pixhawk to see if compass is bad still but everything seems fine. Not sure if I had a hardware failure or maybe short in the wire? I cant recreate the problem. Maybe something really created high magnetic field but i fly here all the time and never had a problem with mag field.

pic of gps/mag after crash

Video of Crash

If someone could check out my log with video and help figure what happen i would greatly appreciate it. I was very lucky but I’m afraid to fly again until i know for sure what happen.

LOG file

Duration 0:12:18
Vehicletype ArduCopter
Firmware Version V3.3.3
Firmware Hash acf2e10c
Hardware Type
Free Mem 0
Skipped Lines 0

Test: Autotune = NA -
Test: Balance/Twist = NA -
Test: Brownout = GOOD -
Test: Compass = FAIL - Large change in mag_field (50.20%)
Max mag field length (707.81) > recommended (550.00)

Test: Dupe Log Data = GOOD -
Test: Empty = GOOD -
Test: Event/Failsafe = GOOD -
Test: GPS = GOOD -
Test: IMU Mismatch = GOOD - (Mismatch: 0.29, WARN: 0.75, FAIL: 1.50)
Test: Parameters = GOOD -
Test: PM = NA -
Test: Pitch/Roll = NA -
Test: Thrust = NA -
Test: VCC = WARN - VCC min/max diff 0.348v, should be <0.3v

2

From my inexperienced glance at your logs I can see that your motors are maxed out trying to stabilise the Octo while your battery current max’s out and your battery voltage hits 12v.

It seems like your batteries just ran out of juice and couldn’t stabilise anymore.

This could have caused the EKF anomaly.
Have you done a Compass/Motor calibration on this rig?

1 Like
3

thanks for the reply…i have done comp/mot calibration and it was low percentage and within range. The Ekf failsafe didnt seem to engage until i was sideways into the trees. Im using Frsky telemetry and qgroundcontrol telemetry on my batteries and i was monitoring them closely and i was landing around 3.9 v each cell. How can I see maxed out motors and the volts dropping to 12 volts… I am using 2 10,000 mah batts for redundancy. After the crash one of my batteries went flying into the woods. One battery continued to spin all the motors until i disarmed them.

4

Looking at CURR/curr the amps are around 35A for the flight but suddenly max out at 180 during the crash.
Looking at CURR/volts at the same moment they dive to a low of 11v at max amps.
All this is preceded by a FS-EKF-INAV and EKF-Check: Bad Variance Detected
I can see no RTL messages in the log.
Looking at RCOU shows a large disturbance on output just prior to the FS-EKF

This is where things obviously went south.
You will need someone better than I to analyse what the EKF messages mean.
Interesting that the log records a Stabilise mode change when you initiated RTL with CH7.
Might be just a log reading error.

5

could there be any correlation b/w my crash and telemetry errors…rssi errors. I cant figure out either if vcc and high mag field spiked before or after crash…do i need telemetry log to see mag field?

6

Agreed, a drift like that tells me you ran out of power.