Hey All,

I was wondering if someone could confirm a hardware failure from looking at this log for me?


My setup:

APM 2.5 with internal compass

3DR Quad kit (essentially this one: http://store.3drobotics.com/products/3dr-arducopter-quad-c-frame-kit-1 but I bought it with the uBlox LEA-6 GPS module before the compass was included)

3S-4000mah battery (fully charged and balanced before the flight)

10x4.7 props (balanced but some minor battle nicks on them)

850kv blue motors

Moon Gel pads (double layer) under the corners of the APM and APM secured to top plate with rubberbands

3DR Telemetry radios


I have done all of the calibrations and scored a 7% on my compassmot test! :-)  I do not recall my vibration test results but they were well within margins.


The day here was partially cloudy, about 55F (a nice summer day in Alaska!) with wind gusting up to about 15mph and about 10 steady coming from behind me as I faced the quad.


I had already done a successful flight previous to this one with no problems.  I was testing loiter and RTL on the first flight as well as just flying around practicing circles and nose-in and nose-out orientations.  I swapped batteries and allowed over 5 minutes between flights for the APM to warm up and settle out.


During this flight I was mostly again testing Loiter with a bit of RTL and some basic forward and back maneuvers along with side to side.  Some friends dropped by while I was flying so I basically parked the Quad in a Loiter and talked to them while still keeping and eye on the quad.


After a few minutes of talking and loitering we all just watched the quad sort of dip one arm and fall from the sky.  It wasn't a tumbling type of a fall, it was flat but sort of wobbly as it fell pretty much straight down.  Luckily it had been raining for the last 5 days so the field I was flying over was really soft and there wasn't any real damage except for the landing gear which cracked.  One arm/motor was buried in the soft dirt and were stopped but the other 3 were still spinning.  Moving the throttle had no effect and trying to disarm didn't either.  I thought about letting the battery run itself out but then decided to carefully reach under the quad and unplug the battery.  As I think about it now I was probably still in Loiter mode and simply switching to Stabilize probably would've allowed me to disarm normally but it didn't occur to me at the time.


Anyway, that was my flight.  I took a look at the log tutorial and to me it looks like when I graph the roll values and the pitch values that a mechanical error occurred but I am certainly no expert reading/graphing the log data.  If someone could confirm this I would be very grateful.




Views: 1502


Reply to This

Replies to This Discussion

Yes, that sounds like a mechanical problem: a motor, prop collet, or motor connection failed. 

Thanks Chris.

Prop was still firmly attached so I can rule out collet and prop probably. :-)

On impact the motor got packed up with some fairly sandy dirt and I haven't been able to get it all out so I don't know if the grinding I feel in the bearings occurred during the flight and caused the failure or if it's because of the silt.  Either way one motor is toast without some work.

I'll pop on a spare and get this back up in the air soon!


Thanks again!


An unceremonious landing! :)


Sorry about your quad Mike!

I didn't look at your logs yet but the description sounds similar to what i experienced with my hexa recently.

here is my post and theres a video of the actual motor stop. http://diydrones.com/forum/topics/need-some-help-with-these-logs

Does that look similar to what you experienced?


Just had a look at your Log.

Findings so far:

You had around 10 sats all the time, so not GPS glitch

Mode was in LOITER all the time, so not "mode change glitch"

Battery always above 10.5V

Up to the point Line 16400 everything seemed hunkey dory, then suddenly your output throttle went wild.
Unfortunately I have no idea why it did this.

Barometer VS WP Alt (Above) looks great as well, except that AFTER the quad started to fall, the WPAlt went as well. I would have expected it to stay.

I do think I have a possible idea how the problem manifests itself, but no idea what causes it. Will start a new thread to see if it is something we can replicate, and needs looking into. If I am right, it will explain quite a number of unsolved crashes reported lately, including mine.

Thanks for taking the time to review the logs Wessie; I appreciate it.


Not sure why the baro and wp alt would differ either.  I haven't done anything special to the case to further protect the baro from outside influences (wind, light, heat, etc.)  It's a pretty default build so not sure if the wobbling fall had anything to do with the graphed values or not?



So just to wrap this up a bit, I replaced the motor on the arm that had buried itself in the dirt with a spare I had on hand and fired up the quad and found that the esc was really the problem.  I replaced the esc and everything appears normal.


Once the quad was put back together I took a closer look at the solders on the esc and found that the ones on the Deans connecter (the one on the esc that plugs into the PDB) was pretty grey and looked suspect to me.


In any case, I'm back in the air with the new esc and now I'll take some time to inspect the motor.



Great to hear you got to the bottom if it. Happy flying!

Hi Mike and/or Wessie, I have been having what sounds like identical problems with both my APM quad and hex.  Did you ever resolve this beyond changing out the ESC?  I have built 2 new machines with different ESCs (different brands even) and still have the problem. ThrOut suddenly drops, then there is a major roll issue as the machine drops to the ground.  No issues with any of my fixed wings running APM though...  So frustrating!!



Reply to Discussion


© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service