Hi, last thursday I did the maiden flight of my Believer on Arduplane 3.9.6 with the Cube. It took off (auto takeoff) and 2 or 3 seconds later switched to safe mode, and obviously it crashed as servos and motor deactivated. BRD_SAFETY_MASK was set to 0 (my mistake) so I couldn’t take the control with the RC.
My question: is there any wrong parameters setting that could have caused this behavior or it was strictly a hardware failure (the button located on the GPS activated itself)?
do you have a log? This has happen a couple of times in the past and is why BRD_SAFETYOPTION param was added, you can disable the safety switch with the aircraft is armed
nothing obvious in the log, although you compass is very unhappy, try re-calibrating it and moving it away from high current stuff.
I’m not sure known hardware failure is fair. Sometimes they trigger without being pressed, due to bad wiring, or getting wet and stuff. I think there have only ever been a handful of reported cases tho.
Hi, I looked at your log and I think the problem is a glitchy safety switch or wiring and your BRD_SAFETYOP[TION is set to 7, but the parameter range is 0 to 3 and it is a bit mask. 7 in bits is 0111 and 3 is 0011. If its only using two bits then 7 would be taken as a 3? Not sure about that, but 7 is not an option. For what ever reason all of your RC outputs were going intermittently to zero or no pulses at all. This is what caused the crash. It was happening for the many minutes your were sitting with the FC logging. It stopped, cause you pushed it before take off. Did a nice auto take off and then the no pulse glitches comes back and all motors stop moving. You switched to manual but were still going in to safety mode so it was no use. There must not be a message sent for the safety switch, but I can’t think of anything else that could stop all RC out pulses. Look at the RC outs near the end of the log. Glitches start at the same time as thing looses control.