I've been test flying my multiplex cularis with a lot of weight, so it tends to stall around 25kph in level flight (I only have GPS speed for now, but I had this happen in multiple directions, compensating for wind. Arduplane does display airspeed, but I'm not too sure how accurate that calculated value is).
Anyway, I got a stall in level flight while doing about 40kph and not much nose up or severe angle of attack as far as I can tell, and I'm not sure what happened:
http://youtu.be/Oz1TPiY33rE?t=11m28s
Do you see something obviously wrong, or did I get bad data from the APM and my speed was way lower than indicated?
Other stalls can be see at the 09:00 and 13:00 mark. Those happened because I was too slow (15kph slower or so).
Flight info is here:
http://www.droneshare.com/mission/14336
The stall at 11m28 in the video is 20:26 for the APM timestamp
Also, I'm not too sure about how APM Plane computes the airspeed if I don't have an airspeed sensor (I don't), but it looked way wrong at 4:50 (14:20 APM time)
Any idea why or how I can help improve it (other than adding an airspeed sensor of course)
Thanks
PS: I have 3 videos (2 highres from other cameras), if you can recommend software that works on linux to merge all 3 videos in one frame, that'd be awesome :)
Replies
I thought about signal loss, but the plane stalls first, and then the signal drops due to the change of attitude and antenna orientation.
Also throttle was at 0% the entire time, so it's not a loss of power.
Now, I came to wonder if my APM sent an elevator down command, and looking at the logs, I can see the autopilot received and sent an elevator down command (data log shows this around 338 on the data samples, or offset 11:35mn)
Now I have to debug how and why that happened. I can see RC2 (RCIN) dropping the elevator, and I have no idea why. It looks like my L9R frsky receiver did that, and did so before losing signal. Weird...
I had a look at the logs some more and remrssi did not drop before the stall, so the controller didn't go into failsafe from what I can tell
Someone asked me to make sure that the receiver (frsky L9R, long range with RSSI port, but no telemetry sent back over wireless) didn't go into failsafe.
I checked the logs again, and they show a full 6 seconds before the uncontrolled elevator input on channel 2, and remrssi dropping due to the severe change in plane attitude.
I'm attaching a graph with the data at 2 zoom levels RemRSSI drops around 340300 on the 2nd graph while the elevator input happens around 337800, about 6 seconds apart. Also, I see that elevator went to full down.
I know I didn't send this from my controller, so I'm wondering if the L9R can pick up interference that it doesn't reject, or a bug in the receiver code dropped one of my channels to 0, in this case the elevator.
stall.jpg