Right after receiving my new motor and mounting it on my plane, I went for a quick test flight at the local park which I've flown at before since it's closer than Rancho San Antonio. I thought I'd save a few minutes by going to the park since I just wanted to have a feel for the new motor.
Ok, I'll save you the details, there was wind (no big deal), I was too close to the street at end of the park and when I made my turn to land back close to where I was, the plane flew right into the sun. I got blinded and by the time I could have seen it, I tried to dodge a power/phone line and lost. The plane hit the line motor first, and the plane fell right onto the street outside the park.
I have to say I actually had a string of good luck there:
cars avoided running over my plane (thank you).
plane actually got fairly little damage, glue to the nose will fix it.
the motor shaft didn't get bent and the motor looks fine.
prop is broken, but that's why I buy packs of 10, no biggie.
the only reason why the plane wasn't flying 1H later was the plastic piece that is holding the motor is broken and I'll have to buy another one. It's more the time to go get one than the $3 it likely costs (update: very strangely, one of my aileron servos died, considering I have zero impact damage on the plane, I was very surprised when I went to the field after fixing the plane, and it failed preflight due to a broken servo. Bummer, now I have to wait a few more days for a replacement...).
lipo got compressed during the sudden stop but looked ok.
Plane damage isn't too bad
I had a look at the lipo and it looked like only the connector piece of it got compressed but the 3 sheets of lipo were otherwise unbent/unarmed. The lipo looked ok so I put it back in my car with the rest of the pieces and drove home.
A good 25mn after the crash, it self ignited on the desk I was working on. I was able to throw it on the ground and then away from close to my car where it landed while the 3 layers of lipo ignited one after another. Nothing actually got burnt despite an otherwise intense fire as the 3 mostly fully charged layers caught fire one after another (it shot pretty vivid flames 2 feet sideways). Sorry, I was too busy watching the fire to take a video :)
this is what's left of the lipo
Needless to say that on top of very limited damage to my plane, I'm very lucky that the lipo didn't self-ignite in my car's trunk, or later on when I wasn't watching it.
Moral of the story, a charged lipo that got any kind of reasonable shock is a ticking timebomb! Check Lipo Fire on youtube for more examples.
After adding a tail ribbon to my plane (to make it easier to see where is the front and back when it's far away from me), I went for a few more flights to catch some higher altitude HD video.
My first flight went mostly ok: (make sure to select HD in the resolution slider)
Unfortunately, on the second flight, the motor crapped out. First it started vibrating a lot, hence the shaky video, and when I removed the throtlle in flight when preparing for landing, I wasn't able to re-engage it. The video after the fact gives the 4 beep code from the ESC which means motor failure, shutting down. At that point, the minimag with all that weight feels like a 2 for 1 glider at best. I don't remember if I tried to flare, but it looks like I barely had enough lift glide energy left to make it to the edge of the hill (beyond that, it would have gone below my field of view)
Turns out I was able to restart the motor on the ground after a full shutdown, but it didn't restart in flight and you can hear the unhappy motor sounds :)
Just like motor, the video had a problem: after downloading it on my laptop, I screwed up while recompressing the first video and wiped the second video in the process. The second video was still on the sd card, but its cluster table had been freed since I had reused the card already.
In this case, photo recovery software usually only has to find the signature of the file by scanning blocks, read the file header and deduce how long the file is from that info since it's already lost in the cluster table, and then rely on the fact that all blocks are super likely to be contiguous, so you can just keep reading linearly until you have the whole file. This usually worked well, but Christophe GRENIER's very handy photorec wasn't succeeding there, and neither were other pieces of commercial software I tried: the restored file would not play (I restored the file by hand myself too by just copying the blocks and got the same result).
In the end, I got really good help on the mencoder-users list from Reimar Döffinger, whom I owe a debt of gratitude to: he looked at my sd card image and figured out that my GoPro HD Camera wrote the index-related part of the file header, after the data blocks. With Reimar's help on what offsets I should be looking at, I was able to get the first part of the header at the beginning, concatenate the end of the header that was written after the data blocks, and then add the actual data blocks. After almost 2 weeks of trying, his answer worked and I got my file back. If you are interested in more details, see this message from the mencoder-users list.
Here it is: (make sure to select HD in the resolution slider). Jump to 04:20 for the motor failure.
