Motors stopped/drone crash during landing

PX4 Autopilot
1

Hello everyone,

Log file: https://review.px4.io/plot_app?log=28e7ddb8-6129-4383-b915-df71ebb646f6

Our drone was flying the mission perfectly, but during landing it went down from 30m to 10m relative altitude and the motors stopped all of a sudden causing the drone to crash. I can see nothing suspicious in the log files though. Any help would be highly appreciated!

In sincerely
David Jablonski

2

Hey dayjaby.

It looks like a power outage? The log cuts while the drone is still ~10m from the ground while doing an apparently controlled descent. The moment it cuts the log looks fine

3

There was no hardfault logged?

4

@RicardoM17 Hey, thanks for your reply. I considered this option as well. We have an odroid xu4 companion computer, which had no outage, but it’s not on the same power line as the pixhawk. Is it possible to see whether this happened?

@jkflying In this case NuttX would create a log file on the SD card, right? I have only those files on the SD:

/mnt/sdcardfile
├── bootlog.txt
├── dataman
├── log
│   └── 2019-09-03
│       ├── 12_35_41.ulg
│       └── 12_48_27.ulg
└── logdata.txt
5

Since there is no hardfault, I would guess a power issue, or something that caused a lockup of the CPU/OS.

Could you give us a list of the electronics you have attached to the flight controller, exact hardware versions, and what ports everything is/was attached to?

6

@dayjaby in the log file there’s nothing suspicious, because there is no data of the crash itself!
At least, seem like the drone stopped loggind before the landing/crash.
This can relate to power issues or some hard faults.

Battery drain from 13400mAh to 20800mAh, 7400mAh in total (what is the capacity of your battery?), and battery went from 80% to 70% only, so in my opinion can be a sensing problem, drone sense a lot of battery left and continue flight regularly, when in fact is out of juice and so it crashes badly.

This is my opinion only based on the .log, hope it helps in some way!
Keep me updated! :slight_smile:

7

@simosilva That would be quite a lot for just one minute flight. It looks more like 2080mAh, or? At least if I convert the .ulg file to .csv the last entry for “discharged_mah” is 2079.388.

But yes, there was one issue: We had a 22000mAh battery inside, but the parameter was set to 44000. But if only 2080mAh got drained from this, it shouldn’t be much of an issue or am I mistaken?

8

Yes, if the amount of battery discharge was 2080mAh it might not be an issue at all.

You agree with me that there is no real log of the crash itself right? at least the autopilot have to detect a drop in the log or something like that.

for example here in my log (https://logs.px4.io/3d?log=731d240d-e17e-44f3-af9a-df2f43f89fd7) you can clearly see at 2.14 the raw acceleration of the impact with the ground and also the rapid decreasing of the z position.
(in my log, don’t look at the gps signal that drops once in a while, is a problem that occurs every time and it’s not solved yet, so i cannot fly in mission mode :sleepy:)