corrupt video

[mailman35]

so it was bound to happen, i had just gotten the quad up in the air and took it out for a little flight it was flying good, until i got to the end of the field i was flying in. and all the sudden i ended up loosing control. but i was able to avoid the bigger trees, and other obstacles and ended up just snapping an arm in the soft ground. so i need to clean out the motor, as well as take it apart and look for any damage.

so my question is, i have a corrupt mp4 file from the onboard video looking through the file with a hex editor, im missing 32000 bytes of the header, ive tried copying them over from a known good file but have had no luck, im more intrested to see what exactly happened on the failed flight

 

[RTRyder]

so it was bound to happen, i had just gotten the quad up in the air and took it out for a little flight it was flying good, until i got to the end of the field i was flying in. and all the sudden i ended up loosing control. but i was able to avoid the bigger trees, and other obstacles and ended up just snapping an arm in the soft ground. so i need to clean out the motor, as well as take it apart and look for any damage.

so my question is, i have a corrupt mp4 file from the onboard video looking through the file with a hex editor, im missing 32000 bytes of the header, ive tried copying them over from a known good file but have had no luck, im more intrested to see what exactly happened on the failed flight

If it's a file from a GoPro I have a Perl script that will get it to the point that you can view whatever amount of video that is salvageable. I used it on the corrupted video from the crash of my FPV quad this past weekend and I got roughly half the video back, you will need to install a version of Perl to use the script. I've never tried it on any other MP4 beside what came out of a GoPro so I have no idea if it works, don't see why it wouldn't.

If you don't want to load Perl just to try this, figure out someplace to post the file so I can download it and I'll see if I can fix it.

Ken

 

[mailman35]

i'll take the .pl i can always toss the video over on to a vm and run it from there. im just more curious if something else happened that caused the loss of control.
and the video is from a gopro.

 

[mailman35]

actually i found that perl script it gets me almost to the point where everything started to go south

 

[RTRyder]

i'll take the .pl i can always toss the video over on to a vm and run it from there. im just more curious if something else happened that caused the loss of control.
and the video is from a gopro.

View attachment 4113

I couldn't upload it with .pl extension so I changed it to .txt, rename it and you should be good to go.

Usage is fix_header_v3a.pl <filename>.mp4 -reso (R1, R2, R3, whichever it was recorded in)

Don't be surprised if you don't have all the video, I lost the part I most wanted to see, the last 10 to 15 seconds and I still had to import it into an editor program to see what the script recovered, it wouldn't play in Windows media player even after the repair. Not ideal but as far as I know the only real chance you have to get anything usable from a corrupted MP4 unless you want to spend $$ on a recovery service.

Ken

 

[mailman35]

here is what i was able to recover, its the same the last 10-20sec is what was missed
i ended up from the end height i was at, to the ground over by the wires on the center of the last frame.
i was pointed nose in, and it just slid back to the right. so im wondering if it wasnt a bearing failure in that motor? where it just lost thrust and the quad didnt turn in to a brick just an uncontrolled decent.

http://youtu.be/RkZDuqmD2nQ

 

[RTRyder]

I'm coming to motor failure as the likely reason for my crash. I've been through all the data in flight log, nice feature of MK electronics, and can't see anything to indicate a problem with the signal from the TX or any part of the flight control system, still it just kept losing altitude and not responding to stick inputs even though the flight controller saw the commands from the TX and interpreted them correctly.

At the scene of the "landing" there was debris spread across a fairly large area, appears to have had some decent horizontal velocity when it hit. One of the things I found was a motor bell with prop and shaft intact and looking fairly pristine, I'm wondering now if the E clip came off the bottom of the shaft and the crash was a result of the bell pulling up and off the coils as it was spinning and the resulting loss of lift the cause of the sudden altitude loss. Of course it could also have happened after impact, no real way to know, but awfully strange that this one piece was separate from all of the other bits and pieces and looking like it had when the multi took off.

I'll have to go back through the motor data for the last few seconds of slight and see if that motor was drawing more or less current than the others, about the only data the controller logs that might provide a clue.

Ken

 

[mailman35]

yeah lack of logging data is where im at. if i could look at what the lipo was putting out id have a better idea. but all and all the crash could of been alot worse. esp with all the trees around where i was.
also the motor was sticking a bit when i would turn it.
just need to get the size of the bearings and then order some replacements from boca.