APM 3.3.2 DIY H-frame copter in Auto or RTL suddenly veers off course

SW: APM Copter 3.3.2
HW: DIY quadcopter (H frame), pixhawk clone, power module clone, 915 kHZ telemetry clone, ext. compass + GPS (the HK clone combo pack), 30A SimonK Afro ESC, 2.4 GHz RC, 2x 4S battery
Flight video (different time, only to give an idea): https://www.instagram.com/p/BAxvqBiDwRB/
Conditions: Soccer field, some trees on one side, not much wind, no power lines.

Log: https://www.dropbox.com/s/gnxe1kt7upb8xsi/17.BIN?dl=0

Hi,

this is my first post. Sorry for using clones from China - but I hope that someone can identify what's wrong with my setup anyway.

Problem: Copter in Auto or RTL suddenly veers off course (and I have to save it manually). I'm attaching a log file where this happens (Auto -> Stabilize) around line 282.000. The second occurrence is later, where it goes from RLT to Stabilize. For the first event I've looked at the topics listed in http://copter.ardupilot.com/wiki/common-diagnosing-problems-using-logs but could not find much. The DesPitch vs. Pitch, DesRoll vs. Roll and DesYaw vs. Yaw seem to show the course deviation but I can't identify a reason:
- EKF4.SMX/Y/Z are all normal (below 0.1) at that time - so not a vibration issue
- MAG.MagX/Z show bumps which may well be caused by the course change and not the other way around
- GPS.HDop is at 0.83 until after the switch to Stabilize
- CURR.Vcc/Volt are normal as well (no brown-out)
- All motors, ESCs and batteries are cold to the touch right after emergency landing.
- In both cases the event occurs while the copter is supposed to fly a straight line.

Any ideas?? Thanks in advance!

Cheers
Till


----

Log File ...tmp805D.tmp.log
Size (kb) 27399.052734375
No of lines 331078
Duration 0:19:43
Vehicletype ArduCopter
Firmware Version V3.3.2
Firmware Hash 7f16e4d6
Hardware Type
Free Mem 0
Skipped Lines 0

Test: Autotune = NA -
Test: Balance/Twist = NA -
Test: Brownout = GOOD -
Test: Compass = GOOD - mag_field interference within limits (19.45%)

Test: Dupe Log Data = GOOD -
Test: Empty = GOOD -
Test: Event/Failsafe = GOOD -
Test: GPS = GOOD -
Test: IMU Mismatch = GOOD - (Mismatch: 0.54, WARN: 0.75, FAIL: 1.50)
Test: Parameters = GOOD -
Test: PM = NA -
Test: Pitch/Roll = NA -
Test: Thrust = NA -
Test: VCC = GOOD -

Views: 418

Reply to This

Replies to This Discussion

No ideas at all? Help me out here guys! To which version of APM would you recommend falling back? My usecase: mapping missions with Tower on Android. 

I discovered that AccY of IMU2 is not working properly, it's stuck at 0! I checked earlier logs and there it's fine. Perhaps that's why VibeY is twice as high as VibeX? But anyway they're both way better than what's considered good enough (15 m/s^2 = good, 45-60 = critical, mine < 7).

I've plotted the flight path in AUTO in Google Earth, the course deviation is clearly visible:

Ok, I'll share my own analysis details and hope that someone can tell me where else I should look. Again: I'm following http://copter.ardupilot.com/wiki/common-diagnosing-problems-using-logs.

DesPitch / Pitch - reacting sort of slow?

DesRoll / Roll - reacting sort of slow?

DesYaw / Yaw - WTF!

EKF4.SMX/Y/Z - GOOD

VIBE.VibeX/Y/Z - GOOD

MAG.MagX/Y/Z - inconclusive

GPS.HDop - GOOD

CURR.Vcc - GOOD

CURR.Volt - GOOD

Reply to Discussion

RSS

Groups

Season Two of the Trust Time Trial (T3) Contest 
A list of all T3 contests is here. The current round, the Vertical Horizontal one, is here

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service