Hi All,

This afternoon I flew a photo mission with my quad in AUTO mode ( take off and landing included in the mission). I have previously flown similar missions without a glitch .

Take-off and photo mission went according to plan.

However once the quad returned to launch (altitude 40m), and before starting the landing stage :

  • It proceeded to circle in a spiral path , moving further away
  • I switched control to RTL,
  • It  briefly jerked but did not proceed to RTL
  • After another spiral moving further away , it crashed on some low trees

Checking the logs, GPS satellite link was never lost...

I have very little experience checking the logs but I really want to understand what went wrong.

Here attached is the flight log:


Any help explaining the cause of this incident or interpreting the flight log data is greatly appreciated.

  • 3DR Quad- C
  • APM 2.6
  • Arducopter 3.0.1


2013-10-15 17-37 32.log

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

Join diydrones

Email me when people reply –


  • Developer
    Swirling is a good indicator of magnetic interference. It may only happen in one particular orientation, or heading. Changing the yaw the interference can subside. Using commpasmot & compass calibration procedures (or redoing them can help) see this post for a discussion i had with randy when i first flew 3.0 http://diydrones.com/forum/topics/arduplane-3-0-0-peculiarities

    What make of battery do you have. You say it was 10.8 (4.0). Voltage sag is normal, it will always read higher under no load. I always fly with telemetry, and the kind electronic lady always keeps me posted of vital stats, don't need to take my eyes of the craft.

    A good piece of advice is that, if auto mode is not working, RTL will not either. Best options is to try LAND. You can reposition the landing target as it descends. OR switch to stabilize and just use throttle to bring the craft down to land. You will drift, but you will probably get down before hitting anything. The secret is not to fight with pitch and roll. It will fly level with no input.

    I haven't look at logs, i'll do that later and give you some more info.
  • I wanted to add a few comments on the thread after reading all the posts...  some might be repetitive... 

    Regarding the battery reading 10.8V after flight.. that is not reliable as I am sure you have read above in some other posts.. I have alarm at 10.8 volts and when I hear the buzzer, I start to land but immediately putting it in Stabilize Mode and taking control if I am in a mission. When I land and read the voltage it jumps back up to 11.2V so that really doesn't mean anything. I also have Voice Module for the radio that constantly reads out the Voltage so that you can hear it instead of looking down at the Telemetry (if you have that).  Also, I used to have the Battery Failsafe Enabled.. and quickly turned that off as there were too many( I mean like 2) unexpected landings as it seemed that the RTL function for that Failsafe didn't quite engage.  Knowing what your flight times are is a good thing to understand so when you set your Timers on your Radio for that you can easily anticipate how long to fly and what kinds of missions//flying you should do.  Start to chart out your battery usage in Excel and quickly see the results.

    Hope this helps....

  • You were just about to enter the autoland phase, so had radio contact. One thing I've done when auto goes bad is switch to stab and take over from there.

    The only problem is, just switching to stab isn't enough, you have to switch to stab, back to say, loiter, and then back to stab.

    You probably could have landed it manually. Besides, in manual you can control the battery load a "little better".

  • I had my copter go bonkers while just flying around. No power module (voltage and current measuring should have been disabled and all failsafes and fence disabled.

    Was climbing in alt-hold when it suddenly went into RTL (some failsafe according to telemetry) and begun behaving erratically. Initially I thought it was because of faulty GPS but the track seems to be fairly coherent with the video. The final part of your story sounded very similar to my experience, all with the spiraling out while in RTL and descending into trees.

    Here's the video. Log attached.

    2013-08-28 19-20 41.log

  • I must say ive been seeing a lot of these false fail safe crashes lately. Im starting to get really weary with the APM. Someone please convince me that this is a great flight controller because im so close to switching back to the naza. I hate not knowing what to fly on my QAV. APM for flexibility, endless options and features or naza for reliability and complete trust!

    I just dont know what to do. I cant sleep at night!

  • Dropping into the voltage sag curve or is never a good idea, and no amount of tweaks will give you stable flight in there. Remember - your "load" voltage could be 1 volt or more lower than what you read at idle. If you read 10.8 at idle, you could easily have been at 9.5-9.8 in flight - below 3.5V per cell.

    Choose some good cell, set your alarm for 3.5V per cell, and land *immediately* when the alarm goes off.

  • I've noted my quad starts to sputter @ 10.5 but I've been able to land quickly.
    Recommendations on a cap value and source??
  • As per what is seen in the logs,

    #1. You triggered a battery FAILSAFE

    Sub Systems / Error Codes

    Battery failsafe


    Sub system 6/ ECode 1: battery voltage dropped below LOW_VOLT or total battery capacity used exceeded BATT_CAPACITY

    #2: It triggered an RTL twice. Once at the very beginning and the other at the end when you were expecting it to. When RTL was triggered, something went wrong (as is seen by the IMU data).

    #3: Something certainly went wrong leading to the crash. You can see this happening at certain point, but I believe something to do with battery voltage. Do you have battery voltage failsafe ON?

    Read this link:


    Also, see Randy's video on how to interpret logs. A nice 20 minute video.

  • Low battery?  My copter begins to list to one side and fly away when the battery gets to a lower level. 

    --- Just a guess. I'm still newb.

This reply was deleted.