[Crash][ArduCopter 4.6.2]DesRoll and DesPitch Error when Arming.

ArduCopter
1

I hope this message finds you well.

I am currently testing a new copter using ArduCopter firmware version 4.6.2. During the first flight, everything functioned normally. However, on the second flight, the copter unexpectedly rolled over during takeoff.

After reviewing the flight log, I noticed that both DesRoll and DesPitch values were not normal immediately after arming while the copter was still on the ground.

Key Observations:

First flight was stable.
On second flight, the copter flipped on takeoff.
In the log, DesRoll/DesPitch are offset right after arming.
No hardware changes were made between flights.

I have already checked motor order and propeller orientation, and they are correct. I’m currently reviewing EKF and sensor calibration but would appreciate any insight or suggestions you may have.

I’ve attached the log file from the second flight for your reference:

Please let me know if there’s anything else I should check or reconfigure.

Thank you for your support.

2

Attache a link to a log.

4

Sorry i forgot link log file : https://drive.google.com/file/d/17-7tPE0QqDyVnIoFGXLDw_KJA8PWopAb/view?usp=sharing

5

Fellow Enthusiast still learning here. Some ideas:

Your GPS1(blue) and GPS2(green) differ. Seems there is a scale difference between the two. Maybe worth investigating. Set GPS_AUTO_SWITCH to default value 1

Not all Prearm checks where enabled. Why? Maybe something was overlooked before arm?

Was the Copter moved during Initialization? (LOG_DISARMED is disabled so we can’t check)
That could result in wrong EKF Init and flipping.

The compass also shows errors maybe magnetic interference?
COMPASS_AUTO_ROT is disabled

Hope it helps!

1 Like
6

Set:

ARMING_CHECK,1
ATC_RAT_PIT_I,0.129
ATC_RAT_RLL_I,0.1359
INS_HNTCH_FM_RAT,0.7
INS_HNTCH_HMNC,1

Arming checks should not be disabled unless there’s some known fault that’s safe to ignore (before you get a chance to fix it) otherwise work through any arming checks and resolve the issues.
The Rate I values should usually be the same as the P values, Arducopter is not the same as Betaflight.
Minimise any harmonic notch filter interference until there’s a good clean flight.

Also perform the mandatory Accel calibration if possible, or at least do the “Calibrate Level” option - just to be sure the hardware is OK.

EDIT: to be clear, I believe the crazy high I term values are the cause of the issue

Need to lower the D terms too - I missed them the first time around. I saw the over-active D term in @Joshua_Kordani 's graph.


ATC_RAT_PIT_D,0.0065
ATC_RAT_RLL_D,0.0068
1 Like
7

I’d like to see the log of the working flight because I have no idea how it got off the ground the first time.

newplot(13)
newplot(13)1244Ă—833 242 KB

9

This Log working Flight: https://drive.google.com/file/d/14Dom8B56ycOGxdp-6mO4z3GwD1YG75Qo/view?usp=sharing