Analysis crash and impossible disarm

ArduCopter Copter 4.1
1

Hello,

I just recovered a quadcopter that had a crash and I have to analyze what happened.

I think I found it, but I would like a second opinion.

Setup:

Quadcopter X

Cube Black

Firmware: 4.1.3

4S LIHV battery (max load 17.2 volts)

ESC 4 in 1 T Motor 55A

By analyzing the logs of the flight during which there was a crash (log 2022-08-10 17-03-43), we see that the battery failed. At first, the battery was only charged to 16.76 volts instead of 17.2 volts, and after several minutes of flight, the battery voltage collapsed, leading to the crash.

For having discussed with the user, he told me that the battery was swollen, confirming what I thought.

The drone showed little damage after the crash.

The user replaced the battery and when arming the drone (log 2022-08-12 10-37-36), the right rear motor (n°4) seems “out of sync” and the drone which was placed on the ground vibrated a lot. Unable to disarm the drone, he had to disconnect the battery. The motor does not appear to be mechanically damaged.

I picked up the drone for testing. I removed the propellers and put the drone into operation. Without propellers, the motors visually seemed to be working fine (Loiter flight mode), but after a few seconds I got a “Potential Thrust Loss” alarm, which seemed logical to me, since there was no propeller, but impossible to disarm the drone (with the throttle stick at the bottom left), I had to disconnect the battery.

Questions:

Could a battery failure have caused a synchronization fault on an ESC ?

Why can’t the drone be disarmed ?

Is there a way to know the number of flight hours of the drone with the logs ?

Thanks for your help !

https://drive.google.com/drive/folders/1kGa9aIfS5K0QwRCv9IMIXZ9LTLkyd1X3?usp=sharing

2

I would agree the battery voltage fell off a cliff. Even ESCs tried to continue working, but just couldnt with the low voltage.
Apart from the battery possibly not being fully charged, the real problem is the battery failsafe voltages are wrong. Having these set would have forced the aircraft to land some time before crashing. If nothing else, it would have already been descending and attempting a landing before the battery got so bad control was lost.
Set these for HV LIPO 4S
BATT_ARM_VOLT,14.70
BATT_CRT_VOLT,14.00
BATT_LOW_VOLT,14.40
MOT_BAT_VOLT_MAX,17.40
MOT_BAT_VOLT_MIN,13.20

and you can probably set
BATT_FS_LOW_ACT,2
if you are comfortable with an RTL instead of Land.

Ideally you would run Autotune again since those MOT_BAT params will affect thrust (scaling) a little bit.

Normally before first flights you would use the Initial Parameters section in MissionPlanner and put in your prop size and battery details.

Thrust loss during bench testing is expected.
We just use Stabilise mode, and you can even hold the aircraft frame and do pitch and roll to check which motors speed up and slow down.
Also it can be hard to disarm if Ardupilot cant detect a landing. Stabilise mode is good for that too. We also have an Emergency Stop switch on the transmitter for human safety and for the times it is better to fall a couple of meters without motors running than it is to hit the ground with everything still going for maximum (destruction)
RC13_OPTION,31

3

Hello,

Thank you for all your information, I will follow your advice.

Dorian

4

I wouldnt mind betting that the wrong battery params lead to the battery being over-discharged several times and now we are here…

5

I meant to say, check the STAT_FLTTIME parameter, it’s in seconds.
There’s other interesting STAT values too, I think they are read-only.

1 Like
6

Ok,

Thanks again, I’ll take a look.

7

Unfortunately stats are not read only.
I did a PR some time ago to make them read -only.