Prearm Issues with APM 3.3

Posted by Thomas Schmitt on April 20, 2016 at 12:27am

Hi,

since I use Copter 3.3 Firmware i have a strange behavior that i like to share here. Maybe the one or the other has the same effect or knows how to fix it.

Description: If I try to arm my Pixhawk I get different prearming check failures (EKF..., GPS Horizon...., Gyros not calibrated etc.). Some of them are not documented by now.

After some reboots the arming works and the copter flies without any visible issues! Also the logs are fine.

First i thought that this is due to the hardware or may the cold temperatures outside in Jan/Feb. But meanwhile i see this behavior also on a second Pixhawk from a different vendor and  (even in Germany) the temperature improves.

Is there someone who saw similar effects or may has a solution for this? I would not like to disable the checks because they were helpful in some cases.

Thanks Thomas

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

Join diydrones

Email me when people reply –
Follow

Replies

    • Tony Wall > looninho April 23, 2016 at 12:36am

      The auto-tune video I mentioned is here: https://youtu.be/qXUEZmchBSI

      The pixhawk wobble video (fixed for me by higher filter on multiple 250 quads) is here: https://youtu.be/sYRfmbrCFsM

      If you take those starter settings and do the manual pre-tune, the auto tune should work also in the current AC 3.3.3. Actually I found the manual part was not necessary as I ended up with the same 0.004 roll/yaw D rate anyway.

      Dennis appears to have one bad IMU in his video and disables it, but I wonder if he just needs to increase the filter or IF this is a just a firmware bug. When the EKF calculation between the two IMU values has a bug then then disabling one clearly works around that code.

      Also I noticed a difference in the behaviour of Mission Planner vs APM Planner 2 regarding the PID values. The current Mission Planner 1.3.37 version bumps-up some of the PID values if you edit then using the GUI because they are lower than the minimum range. I played around with APM Planner 2 a bit and it appears to be more aware of new settings/defaults. But need time to confirm this. Makes me wonder if the whole thing was just messed-up by the calibration routine of the planner. So that's another factor to test.

  • Thomas Schmitt April 20, 2016 at 6:08am

    Dear all,

    i opened an issue in the Arducopter 3.3 support forum Link. Hope that these guys have an idea or can fix the issue.

    Thanks all for replying so fast!

    Thomas

  • looninho April 20, 2016 at 2:06am

    Hi,

    same issues for my Pixhawk 2 FC from goodluckybuy.com. Also My GPS M8Nis very less sensible (can not fixe if i'm slightly in indoor) compare to my other M6N (from another vendor). I then open the case of the M8N and see that the patch antenna has some scratch. The internal compass of my Pixhawk has very big values compare to the external one.

    I suspect this is hardware issue. Can somebody knows how to check all sensors? Standard calibration doesn't reveal fault.

    Bests

    Looninho

  • ActionDrone April 20, 2016 at 1:00am
    I also have those issues, randomly. I'm having some doubts about the reliability of this FC that I never flown.
  • Tony Wall April 20, 2016 at 12:43am

    I can confirm this. Me and my friend both have two new PixFalcon quads we built and suffer the same problems. Both originally and after being completely rebuilt with freshly soldered connections, so I'm sure we are not suffering an installation problem. All devices are properly spaced away from each other and both accelerometer and compass calibration were completed several times.

    It's really random, on one day both drones were being flown side by side. The first drone armed immediately (after the GPS cold/warm start delay which is expected). The second done tried to arm a minute later but was stuck for over 20 minutes with various errors like accels not healthy/compass/GPS-horiz, multiple reboots didn't help. Next battery and the second drone armed first time but now the first drone suffered the same faults as the second. The evening continued that way, at complete random either APM arms first time sweet or just plays-up for 10-15 minutes.

    It's so frustrating we had to take a risk and disable pre-arm checks and GPS fence, just so we could fly without hassle! My friend is considering moving to CleanFlight because APM appears too unreliable to use. I find that sad because I'm behind the platform, but have to agree taken into account this behaviour.

    If find normally that the accel and compass errors are cleared by waiting a minute, moving the drone around, moving to a new spot of ground then retrying the arm sequence. However, the GPS horiz error is really stubborn, even when 12 satellites were detected and 3D lock APM was still refusing to arm with this error, which doesn't make sense if we have a 3D lock that should be good enough already! Even the map on ground station was accurate and not jumping around, so definitely no horizontal error in the GPS. I'm sure this is just a logic problem, perhaps the EKF or other complex part of the "health" calculation.

    Anyone else? Guess we need to upload logs. Or can somebody suggest the correct set of log parameters so we can analyse the logs ourselves?

This reply was deleted.