After waiting for the weather to clear up, I managed to get the quad out for an altitude test. To make a long story short, everything was going quite well up to a couple hundred meters or so. At that point I figured it might be time to descend so as to not kill the battery entirely. I used 'Land' mode, since I'm sure I would have messed something up descending manually. Everything was working properly until about 100m when the poor thing decided to spin excitingly out of control. I think I managed to recover at the last second, avoiding a complete free fall.
I'm not terribly savvy with analyzing logs, so I'm hoping someone can help me here. What went wrong? Propellers were intact until the landing, only one blade broke on impact. I've attached the log file, as well as a video of the curious event. Is it possible that the battery was dead by that time, and was unable to provide enough current (until the last second)?
Edit: I'll admit to being a terrible human being and allowing the 3S I was using to get down to 3.2v per cell. The question still stands: would that still have been enough to descend with, or did that contribute to copter craziness?
I'm running ArduCopter 2.6 still on APM1 (haven't flown much since early summer). Any help would be well appreciated!
I can't help with the log analysis but I can shed some light on LiPo use.
The operational voltage of a LiPo is 4.2 to 3 volts. 3 volts is regards as discharged but not in danger.
2.5 volts is regards as the point at which you will kill a LiPo, but I have managed to regain some cells from near zero in better quality packs, but I wouldn't count on it.
Most ESC's are set to auto cut off at 3 volts, some some can be reduced to as low as 2.8, but this is under load voltage, which after 3 volts is reached will happen pretty quickly.
If the no load voltage after a short rest is 3.2 volts, then maybe you did hit the 3 volt cutoff (unless you programmed the ESC to think they were using NiMh, which is a good idea) and the cells have recovered back some voltage under no load (which they do).
So, no, at 3.2 volts, I don't think your a terrible person ;-)
Mike, everything about that clicks. I should have considered that, so thank you very much for pointing it out! I'll check the settings on my ESCs and see what they're set at but your explanation makes perfect sense and I'm fairly confident it's exactly what happened, at least to one motor if not more.
Though I still think my reckless disregard for in-flight voltage is inexcusable, I quite appreciate your confidence in the quality of my character :) Thanks a ton for your help!