Crash Analysis

Log Analysis -- deprecated Copter Logs -- deprecated
1

Hi,

Could someone please look at my dataflash logs? My first impression was that the crash was caused by an ESC/motor failure of the pair off channel 1 due to Arducopter maxing out channel 1. But after bench testing all motors/ESCs it appears that they all still work fine. This leaves me at a loss as to the possible cause of the crash.

TIA

2

Here are a couple of screen shots of the graphed tlogs.

[attachment=0]Screen Shot 2014-10-02 at 9.51.59 AM.png[/attachment]

[attachment=1]Screen Shot 2014-10-02 at 9.54.43 AM.png[/attachment]

3

It looks like it hit a WP at line 12123 and changed directions. It’s at this moment that the RCOut jump up, and RCOUT1 maxes out. The current draw jumps also. It looks like motor one couldn’t keep up with the needed power and the quad pitched and rolled uncontrollably forward to the left.

Probably stuff you already know, but that’s my assessment. Bad #1 Esc or motor.

I would love to hear others input.

Rich

4

Thanks Rich,

When I consider the motor failure I keep getting hung up on the fact that the pitch and roll desired/actual were tracking so well. I would expect if there was a mechanical failure one or both would not track well at all. Do you have any thoughts regarding this?

Thanks,
Carl

5

That’s a good point. I’m not sure why The desired Roll and Pitch are so high. You can see in NTUN the Desired Posx and PosY are quickly diverging from the actual Posx and Posy. So it is not tracking with the direction is should be going.
Maybe the quad over shot and tried to correct, but over corrected. The quad was giving a steep desired pitch and roll, but it could not keep up with the power required and lost altitude.

Now I’m just guessing. :confused: With as many crashes as me and my friend have, I am getting better and dissecting these log files, but most of the time I can find no clear cause. Doesn’t mean there isn’t on, I just can’t find it.

Sorry,
Rich

6

Definitely had yaw issues. I think something went very wrong with the compass.

I’m not certain what it is. Try doing compassmot and recalibrating offsets. Ensure that compass orientation is correct.

7

Here’s a graph of the integration of the gyros in the earth-frame Z axis against the observed yaw. Basically, your compass was correct while you were pointed south but not otherwise. To me, this suggests orientation issues or crazily large offsets.

8

You can help me out by posting a tlog or binary log in which you:

  • hold the copter level, do one full rotation about the vertical axis
  • hold the copter nose-down, do one full rotation about the vertical axis
  • hold the copter nose-up, do one full rotation about the vertical axis
  • hold the copter left-side-down, do one full rotation about the vertical axis
  • hold the copter right-side-down, do one full rotation about the vertical axis
  • hold the copter upside down, do one full rotation about the vertical axis