MR60

Arducopter 3.2.1 Autotune crash

Hi,

During a quadcopter autotune which started very good, I had passed the roll tuning phase and the quad was in the middle of the pitch tuning phase. I recentered the quad a few times. Until I had apparently the bad idea of correcting slightly the yaw position and the quad then started to get out of control : yawing into itself with no apparent control. I switched the autotune switch back to low to get manual control which I did and tried to fight the crazy yaw behaviour for a couple of minutes. As I could not get the quad back in a stable flying state, i made it land as softly as I could but broke one of the motor's mouting plate.

I do not understand what happened especially as I turned the autotune switch back to the low position, it should have flown back with its original PID parameters but yaw was uncontrollable for some reasons.

If Leonard or another expert could have a look at my log to determine if this was an autotune algorithm problem or another issue.

The quad config is the following:

-4S

-16x5,4 Tmotor props

-3508 - 380Kv Tmotors

-Pixhawk with 3.2.1

-GPS, telemetry

2015-07-15 14-59-20.bin

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

Join diydrones

Email me when people reply –

Replies

  • MR60

    Update on this issue:

    I rebuilt this quad with the same parts from scratch. I modified the electronics plate and the pixhawk damper/mount : instead of just using the four pieces of 3DR foam, I used the Omnimac mount. I then redid an autotune and it passed the procedure succesfully.

    My conclusion to this point is that my previous autotune failure was due to either or a mix of these both items:

    -Pixhawk was mounted on these 3DR foam pads which are way too soft and do not hold pixhawk in place well enough when autotune does its rapid pitch/roll moves

    -One motor mount was maybe starting to get loose, as this could explain the strong unstability I experienced all of the sudden. Although I cannot know for sure if the broken mount was the cause of the crash but I would bet that it was.

  • Is it possible you use an opto ESCs?

    • And ESC firmware please

  • Seems that there are quite a few people who have been experiencing issues with AC 3.2.1 and auto tuning...?

    I recenty did an auto tune myself with my asymetric quad, after some re-organisation of the Esc layout and general rationalisation of wiring. No other layout changes, but recalibrated compass and accelerometer, and transmitter.

    What was initially a nice stable platform, with virtually predictable results with 3.1.x, I landed up with a craft with very strange response in Stabilised and Altitude hold , eventually having a controlled collision with the ground and fortunately very little damage. Symtoms were virtually no response to throttle and erratic behaviour...

    I've now reverted to previous saved parameters which will hopefully restore the status quo! Haven't had a chance to test it yet (winter here)

    Is there an issue with the latest firmware?

    Some information from those who know would be most welcomed...

    Best DP

  • Thank's for share your issue, I have to re-tune my modified cuad and, sometimes I had to correct yaw perhaps for bad fixed the ext compass to solve yet but I'm not going to do it during autotune for your advertise, you avoid another crash, many thank's.

This reply was deleted.