Autostart: What happend in flight?

Hi

Today, I had no luck with the plane. And I do not understand, what happend. I tried autostart. Several days before, everything was really fine. This morning I had some problems with the altitude. The Wing flew to low, I had to recover several times in manual mode. When I look to the telemetry-log, I can see that the plane had enough speed to manuver, and what I see, is that the autopilot had no desire to climb faster. But why?
This video was from the morning:http://www.youtube.com/watch?v=S0w-2xEqyVs
You can see, the plane can roll fast…
But when you look the video from this afternoon (http://youtu.be/3_vP6MXxDqo), the rudder dont move a lot. As you can see then in the telemetry log, he had the minimum speed after 3 seconds, 25m/s after 5 seconds. Why is he not climbing to 30m? Better: Why it takes another 10s to reach 30m altitude.
Enclosed you find the telemetry log, the mission-file and the params.

And there is something other that is strange. The plane moves the wrong way in the log. Physically, I flow nearly 180° the other way (heading between WP 2 and 3). Ok, there was “Bad Compass” Msg. But the GPS Signal was ok. Why does MP move the plane this way? The Crashpoint is at :46.889535 / 7.382732. This is the first time I saw something like this. The HW was not changed.

Can someone told me what went wrong?

Link to the attachements: https://drive.google.com/file/d/0BxypCZKXidnzSktEQTNkNUVBdWs/view?usp=sharing

Why it crashed? Well, when switching to manual mode, throttle-cut was still active. Recovered that was no more altitude left to fly… Pilot-Error.

please post .bin or .log from the flight controller.

Find them here: drive.google.com/file/d/0BxypCZ … sp=sharing

Hopefully I found the correct. If not, tell me - or how can I check them? Some of them have no time. I mounted the SD-Card directly to the PC. Is this working right when used in the pixhawk?

the log you attached is from 2 Nov.
If you waited for GPS fix, the easiest thing is just to look at the file time/date.
If not - - you can open a log, and get GPS Week / TimeMS fields, - remove the last three digits to convert Time to seconds, and calculate using labsat.co.uk/index.php/en/gp … calculator

The date is correct, time must be after 14:30 (Switzerland). I flew also in the mornig befor 12:00.

the log does not match your description, seems to climb ok.

  • but ends before landing. (bad SD card usually causes that unless your pixhawk lost power.)

Yes, it ended befor landing cause I crashed…

Yes, it climbed, but it takes almost 300m to do that. The plane can climb 30m much faster. Why isn’t he climbed faster - lets say within 100m as days before?

By the way, how have you made this? If found an interesting tool called APM Flight Log Analysis. But I get only an Error trying analyse a log like "Log must be created by an Ardu Vehicle firmware. I get this message also, when I use MP do download a Log or when I use PX4 .Bin to .Log in MP.

Found another that gives almost your graph: FlightPlot
Thats great!

The graph is made by “APM Planner 2.0” a very neat and quick tool.
Your Elevator C2_TRIM is 1500
The actual logged input during RCIN.C2 when you don’t use the function, is 1584.
This make the AP behave as if you were manually correcting it.

Remember that you cannot use trims like on an standard RC plane, AP must also know what PWM are produced when sticks are centered, otherwise it have no way of knowing when you do manual corrections.

Damn thing - your absolulty right. I used trim this morning, because the payload changed and after level out I have not changed the trim mechanically…

Thank you very much!