Pixhawk Copter "Bad Accel Health"

Posted by ActionDrone on April 1, 2016 at 2:44am
Hello,
I’m going to setup the Pixhawk now, but I continue to receive errors of "Bad Accel Health" or “inconsistent compass”. I'm doing the setup indoor.
I already updated the Firmware (with Craft and Theory build for the telemetry cable, but the issue was the same before this flashing), and I also already re-calibrated the accels, but the issue is still present.
Do you have any idea about what could be the problem? Could it be a hardware problem?
Thanks.

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

Join diydrones

Email me when people reply –
Follow

Replies

  • ActionDrone April 11, 2016 at 3:02am

    Here some older screenshots. I'm going to take something newer today.

    Screenshot 2016-04-07 12.59.42.png

    Screenshot 2016-04-07 12.56.53.png

  • ActionDrone April 11, 2016 at 2:59am

    Screenshot 2016-04-07 23.52.04.png

    Screenshot 2016-04-07 23.47.48.png

    Screenshot 2016-04-07 23.46.24.png

  • ActionDrone April 11, 2016 at 2:56am

    Hey guys, an update since I'm still stuck... maybe because APM Planner for Mac cannot calibrate properly the sensors?

    Some details:

    • Pixhawk with APM:Copter V3.3.3
    • Macbook Pro 13" early 2015 with OS 10.11.3
    • APM Planner software 0b4ab402-2.0.20
    • external GPS uBlox Neo M8N connected to GPS and I2C ports (I2C expansion bay)

    Besides that I cannot connect the SIK V2 telemetry radios (they can see each other, they have solid green LED, they can download settings to APM Planner SW correctly, but if I choose the radio serial port at 56k BAUD rate, they connect to each other but do not show the connection to the Pixhawk) using the Telem1 Port...

    I still have several errors while trying to arm the copter, they are all related to the sensors.

    The errors shown in the info view page are:

    • FAILURE: Failed CMD: 241 (trying to calibrate the compass)
    • PreArm: EKF compass variance
    • BAD accel health
    • inconsistent compasses
    • accelerometers not healthy

    I attach some screenshots. As you can see in one of the screenshots, the map shows a moving copter, which obviously was not moving at all. I was indoor in that moment, but yesterday in outdoor the copter was still showing movements while it was on the ground.

    Attaching more screenshot in the following posts.

    Please help gurus, I'm stuck here and I must fly this bird. Thanks.

    Screenshot 2016-04-08 00.13.07.png

    Screenshot 2016-04-08 00.12.33.png

    Screenshot 2016-04-08 00.10.56.png

    https://storage.ning.com/topology/rest/1.0/file/get/3702223384?profile=original
  • ActionDrone April 7, 2016 at 1:03pm

    Update:

    I attach to this post the screenshot of the graphics. Looks like the IMUs have different values.

    Questions:

    1. do I have to setup in some way the external Ublox Neo8 GPS antenna?

    2. from the screenshot, do you think I may have HW issues?

    Thanks.

    Screenshot 2016-04-07 12.56.53.png

    Screenshot 2016-04-07 12.59.42.png

  • Olav Strehl April 1, 2016 at 2:08pm

    At first I admit that BAD_ACCEL_HELATH / INCONSITENT_ACCELORMETERS and the compass error are two different problems.

    I also encounter theBAD_ACCEL_HELATH / INCONSITENT_ACCELORMETERS problems only on cold days. Which seems to come from values from IMU2 differing to much from IMU1 (I looked at the log) And for me it seems to be the IMU2 issue as described within the linked by Tony K:

    http://diydrones.com/forum/topics/solution-proposal-for-pixhawk-imu...

    (Thanks for this interesting analysis!)

    If the error occurs, it helps for me to let the pixhawk warm up for view minutes and than to move it around all axes until the warning disappears.

  • Justin April 1, 2016 at 9:11am
    I setup 3 Pixhawk's indoor. All three were working great but eventually one started receiving the same message as yours. Turned out to be a hardware issue. The only real way to determine that is by a process of elimination and it sounds like you already did that.
    • Martin Lachance > Justin April 1, 2016 at 10:56am

      If it's broken....fix it.  If it ain't broke......I'll soon fix that too.  Not my original quote, but thought I'd share ( :

      Tony K, thanks for all your insightful suggestions.  I'm hoping I don't have to go back down that rabbit hole as I have found myself.....finally.....spending less time on the bench and more time in the air. 

      One last question.  I have a second (Pixhawk based) quad, she's about 2 yrs old now, that has been rock solid for me running V3.2.1.  Would you update it to 3.3.3?

    • Tony K > Martin Lachance April 1, 2016 at 11:06am

      You see 3.3.3 might expose some vibration issues that 3.2.1 does not. If that quad has very good vibration isolation i would say upgrade if you are not sure id say dont do it.  So it all depends on what size quad it is and how good your vibration damping is, otherwise you should not have any issues. 3.3.3 is more sensitive to vibration issues from what i understand. See here 

      https://github.com/ArduPilot/ardupilot/blob/master/ArduCopter/Relea...

      Copter 3.3-rc1 11-Apr-2015
      Changes from 3.2.1
      1) Only support fast CPUs boards (Pixhawk, VRBrain, etc) and drop support for APM1, APM2 (sorry!)
      2) AutoTune for yaw
      3) Smooth throttle curve which should reduce wobbles during fast climbs and descents
      4) ch7/ch8 aux switches expanded to ch9 ~ ch12 (see CH9_OPT ~ CH12_OPT params)
      5) PX4Flow support in Loiter mode (still somewhat experimental)
      6) Safety features:
      a) EKF on by default replacing DCM/InertialNav which should improve robustness
      b) increased accelerometer range from 8G to 16G to reduce chance of climb due to high vibrations (requires accel calibration)
      7) Landing features:
      a) improved landing on slopes
      b) retractable landing gear (see LGR_ parameters)
      8) Camera Gimbal features:
      a) SToRM32 gimbal support (using MAVLink)
      b) AlexMos gimbal support (using AlexMos serial interface)
      c) do-mount-control commands supported in missions (allows controlling gimbal angle in missions)
      9) Battery related features:
      a) PID scaling for battery voltage (disabled by default, see MOT_THST_BAT_ parameters)
      b) smart battery support
      10) Other:
      a) support do-set-home command (allows return-to-me and locked home position once GCS enhancements are completed)
      b) performance improvements for Pixhawk reduce CPU load from 80% to 15%
      c) firmware string name changed from ArduCopter to APM:Copter
      ArduPilot/ardupilot
      ArduPlane, ArduCopter, ArduRover source. Contribute to ArduPilot/ardupilot development by creating an account on GitHub.
  • Martin Lachance April 1, 2016 at 6:31am

    There are many things I'm going to assume you have checked out.  Such as keeping metal objects (and your cell phone) away from compass and the Pixhawk, along with location and orientation of the GPS/compass puck, double check you have the GPS/compass connectors in the correct ports, etc, etc.    Are you getting "error compass variance....bad AHRS" along with the "bad accel health" and/or "inconsistent compass"??  If so, there's plenty to read up on in DIY Dones and Github.com.  But, before going down those rabbit holes......let's try "kiss" first.....as I wish I had before purchasing 3 Pixhawks and spending a month's worth of time.  What I ended up doing is repeating those 2 calibrations, accel and compass calibration ignoring all the audibles.  On Mission Planner screen do you get "calibration successful"?  Repeat those two calibrations until the audibles go away.  I eliminated all unnecessary items and started with the very basics connected to the Pixhawk on a block of wood.  Just the arming switch, buzzer and compass/GPS puck and provided power through the USB.  But even then, I was getting the error audibles.....on three brand new Pixhawks!!!  Ignore the audibles and push your way through the calibrations.  Granted, I would get the "radio calibration" warnings until I connected the receiver.  Then I repeated the compass and accel calibrations.....4 pr 5 times until things got "quiet"......no more screaming (from me) and no more, "error compass variance.....bad AHRS".  I have follwed through with the rest of the "build" and calibrations, and have flown her without issues using telemetry radios with Droidplanner and Tower apps (which have audible warnings) on a ground station to "monitor" systems.    So far, so good.  Hope this helps. 

    • ActionDrone > Martin Lachance April 1, 2016 at 6:35am

      Thanks Martin,

      I will double check again everything, despite I'm sure the wiring is OK. The strange thing is that the Mission Planner reports the calibration is successful, the system can continue to be setup (I'm doing the setup of Taranis, lights strobes, Connex, failsafe lading gear, simple mode switch etc.) so let's see...

This reply was deleted.