VTOL crash when flying in AUTO mode

Hello,

today I was testing a delivery flight plan with the arduplane 4.4.0-dev firmware.

As you can see in the log, the target speed changes while flying, and the target speed set is 20 m/s.

As you can see, while the target speed is changing the plane is trying to increase and decrease the velocity, reaching maximums of 28 m/s, entering in a oscillation that ends with a crash because the altitude decreased a lot in no much time (5 seconds aprox).

log: https://1drv.ms/u/s!AiZKla5p79iGhYpxKnwh3rWqe-nZ5w?e=5NPdND

Just mention that the aircraft crashed 2 WP before the PAYLOAD_PLACE waypoint.

Does anybody know what exactly happened? Why the target speed is not constant at 20 m/s?

Thank you so much.

Hi Eduard,

do you have a *.bin logfile directly recorded by the flight controller from the SD card instead of the *.tlog ?

Regards Rolf

Yes!

Here you have: https://1drv.ms/u/s!AiZKla5p79iGhYp1NJ4LSGvGDzXdeA?e=GqEbON

Thank you so much.

The target speed is not 20 m/s, but almost 29 m/s, because (THROTTLE_NUDGE=1) the throttle is not in the middle at 1500 µs, but at 1750 µs.


I cannot say why it takes so long for the TECS target speed to approach the CTUN target speed. One of the developers should know more.

The increasing altitude oscillation in phase with the speed changes could perhaps be favoured by poorly PID values. Just one question: Have you done autotune and adjusted the TECS parameters ?
Another side note: You started with the preflight check switched off. This is not a good idea.

But I have not yet found out for sure why the plane crashed.

Rolf

I think you should ask in the forum about the error due to SUBSYS 30 with ECODE 1.

The documentation doesn’t describe it, I can’t give you more help.

It seems that the SUBSYS 30 error is a type of “internal error”, in a word something generic and hardware related.

I also see that when it occurs you have a significant voltage drop at 15.33V (2.55V per cell, assuming 6P).

Were you using LIPO or LIION?

Both the internal error and the voltage drop only occur with or after the impact. They could therefore also be the result of the crash.

Rolf

Hello @Lano

thank you so much for your response. I see this voltage drop and I think that is after the crash. So the problem I think that comes from another source of error. Hpwever, we will check the battery status. We are using a 16000mAh 6S LiPo battery.

Thanks again!

Regards

Hello @Rolf

thank you for you analysis. We checked after the crash that the THROTTLE_NUDGE was abled. This could be the source of error because the pilot always put the stick up just in case to pass to Manual mode to reduce the response time, but in this case maybe this could afect on the crash.

Also, as you mentioned, Iswitched on the preflight checks.

About the PIDs and TECS, were taken from another VTOL platform similar that we have that works well… maybe we need to do a new autotune and readjust the TECS.

About the Target airspeed at almost 29 m/s, the only source of error is the stick of the trasmitter right? Because the it is set at 20 m/s:

image

Thanks to your suggestions,

Regards