Hexrotor fell out of sky, inflight board freeze

Log Analysis -- deprecated Copter Logs -- deprecated
1

Hi!

My hexacopter just made a 20 meter drop. I’m still trying to figure out what exactly happened. It hat a GoPro running on it, and the recording confirms what I saw, all motors stopped, the copter starts to fall, the ESCs complain about the missing signal and the copter hits the ground. Apparently my APM board froze up in midair.

There are a lot of GPS issues during the last minute. My GPS module (Neo6) seems to be broken, it still has trouble to get a stable GPS lock. Could a failing GPS module cause the board to freeze?

There are some spikes in the power issues analysis (CURR/vcc). Could a drop in the voltage cause the board to freeze, even several seconds later? I’m not sure I can trust these readings. There are spikes in volt and curr along with the voltage drops, but I do not have an APM power module installed. I have a BEC connected to the output rail, with JP1 in place. Can the APM board measure curr an volt readings without a power module?

[attachment=0]Power_Issues_01.png[/attachment]

The board remained frozen and unresponsive with the ESCs complaining about the missing signal until I disconnected the battery. It seems to be fine now, but I think it’s time to move to a Pixhawk…

Best Regards,
Claus

2

Browsing a bit more in the log i found a raising number of I2CErrs, which seems to be to root cause for the following troubles, i.e. the GPS issues. And, what I find interesting, a raising NLon along with a MaxT of more than 100.000.

[attachment=0]IC2Errs-01.png[/attachment]
Are there subroutines that take more an more time in case of GPS or I2C problems? Can loop delays get to a point where the board freezes? What is INAVErr?

Best Regards,
Claus

3

[mf]3DR Hardware Support/APM 1.x and 2.x (2.0, 2.5, 2.6)[/mf]

[infocopter][/infocopter]