Unexpected plunge from 40m

My disasters always seem to happen when I say, just one more flight... I should know better by now.

Anyway, I'm putting it down to unexpected equipment failure, unless the log below (see line 21100 onwards) and the insight of this user group can tell me otherwise. 

Visually it appeared as thought the quad flipped upside down, and then plunged to the ground.  The aircrash investigation indicated that the props were all rock solid tight.  On initial takeoff at the start of this log (presumably), I did notice the front motor being a little sluggish to start up when I first powered up (all others spinning, but it was not), but on a second go no issue noticed, so we chose to proceed...

Any assistance otherwise appreciated, still figuring out if the thing is trimmed right, my vibrations appear within tolerances (and I haven't even received my delivery of moongel yet) and otherwise having a fine time playing with this wonderful stuff.  Congratulations on the people who have worked so hard to make this work so very well.  oh, and hello from New Zealand.

2013-07-28 14-24 4.log

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • Thank you all for the helpful comments and observations below.  Addressing some comments individually:

    Euan:  My frame is a jdrones Quad.  It fell into softish earth/grass covered hill side.  Props all smashed as you would expect, cover frame over the APM took most of the impact, assume my gps unit (on top) but sandwiched in perspex cover is ok, it seems to be, but proof will be in the pudding when I fire it up again.  APM fine, and all the bits underneath are fine, two arms are bent out of shape, motors appear to be fine.

    Randy: really useful log analysis as always, thank you.  Sonar I had thought was behaving fine (I assumed, perhaps wrongly that readings over a certain height just came out zero).  I have had some very nice stable loiters at low levels that I also assumed were being controlled by the sonar.  I will turn it off when I next fly and see what difference that makes.  CLI test of it has come up with accurate readings (at least when the sonar was on the table).

    Forrest:  happy to help you with your moongel experiments.  May be a few weeks before I am back on my feet though:  making a new internet order for replacements tonight or tomorrow.  The a new esc was not going to be one of those items, it is now.

    John:  Thanks to this experience and all of the helpful comments, thankfully I will not fall for this twice.  Once is enough.  When I was building the thing I had a twitchy esc for a while, but I figured that out to be a bad connector solder job.  Not sure it was the same esc.

    Thanks again to all of you.

  • This is a bad esc. I have fallen for this twice now but have learned the hard way (broken gopro) NEVER take off if the escs do not power up correctly at the same time. Many times a reboot seems like it fixes it but it will fall out of the sky within minutes. Just replace the esc or risk death.
  • MR60

    Julian - Sorry to hear about the event.  The same thing happened to my brother ... a bearing failure that sent his copter into the side of a cactus ... ouch.  So we built a worksheet to track relative motor usage to see if we could predict motor failure.  It's tough to do because it may happen so fast.

    Noticed that you are expecting a delivery of moon gel.  Are you interested in helping the forum on something we are trying to figure out in regards to vibration?  It would mean doing a controlled flight before and after you add the moon gel.  Interested?  I'll understand if not ... after all, it is work and we love to play.

  • Developer

    Sorry about your crash.

    I wonder if it could be caused by a low battery or perhaps the ESCs low voltage protection kicking in?  It spikes down to about 10.25 volts a bit before the troubles happen although by the time things go really far wrong it's back up to 10.8V.

    3692787228?profile=originalIt looks to me like around 21100 in the logs the copter starts to fall despite the throttle going to full.

    3692787141?profile=originalNow looking at that graph the throttle is basically pinging from top to bottom which isn't great but I believe it's caused by the wild rolling and pitching that happens as it starts to descend very quickly.

    3692787272?profile=originalFrom looking at the desired roll and pitch vs the actual roll and pitch it also looks like a mechanical failure.  You can see at the same time (21100) the roll just suddenly goes to -110 degrees while the desired roll is +25 degrees in the other direction.  That difference is usually a sign of a mechanical problem because if it was the firmware messing up it would be commanding a crazy angle.

    By the way, I see the sonar is enabled but it's not providing a good altitude (all zero) so you should probably disable it.

    ..and logging is convenient to have on for analysis but best to turn off the heavy messages (like IMU) once you're done with them 'cuz they can negatively impact CPU although this isn't a factor in this crash.

  • 40m! Sorry to hear it - did much survive from that height?

    Sounds like a bearing or ESC was on the way out.
This reply was deleted.