My HL octo is flying great with APM2.6 (w/ 3DRuBlox w compass) and AC3.1.5, but I have challenges in loiter. I can take off and enable loiter. Then the octo will stand perfectly steady, for lets say 5 minutes. Then I land, disarm, arm, and take off again and the octo flys away when I try to enable loiter again. 

I have enclosed a photo of the 3DR uBlox GPS w/compass installation in relation to the batteries. Could it be that the batteries are blocking for good GPS reception? First good loiter after calibration had 8 satelites and hdop of 1.7 - 1.6. Next loiter, which was bad, I had 7 satelites and hdop of 2.2-2.3

I have now enabled compassmot and inteference at half throttle is 10% which should be acceptable. I am not sure if this affects the hdop though. 

Does anyone have experience with these issues?

Best regards,


2014-09-09 11.33.30.jpg

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

Join diydrones

Email me when people reply –


  • Today I discovered a rattling sound in my 3DR gps/mag unit. I opened and found the battery loose. The terminals have come loose from what seems to be a poor soldering job by 3DR, possibly cold soldering. This is alarming to me. It could be part of the solution to my initial issue. I have another unit that I will inspect before installation.

    • It looks like the tabs on the battery failed, rather than the solder.

      I think 3DR knows about this issue, but in the meantime I would suggest gluing the batteries down.

    • I agree. I checked again and it looks like the tabs just broke. I will secure the batteries with glue as suggested. Thanks?

  • My quad stays super stable in just 3-4 sats available !!! in loiter mode maybe the arrangement of satellites matter not the numbers to get a 3D lock

  • Thank you for advice,

    I will lift the gps/compass with the mast. The gimbal itself can be a source as I have a huge geared GBM8108 ble motor on the pan axis. Depending on the pan position this could be almost directly below the compass. I thought the carbon booms would shield for the three phase motor wires though. 

    I ordered the Pixhawk now but I am not sure if I will upgrade now in fall season as I will not interupt flying sice the APM is working so well. I have no experience with Pixhawk.

  • I'm guessing there's an issue with serious compass noise because the compass is on top of the 3-phase wires to the motor. You also have a massive Z offset - must be something magnetic nearby.

    Lifting that compass up a few centimeters and recalibrating would probably make this all go away. Place it further out on the arm if you can, as well. Tightly twisting or braiding the wires to the motor will help as well.

    I'd buy these and place them on that same mounting platform, then move it as far out as your props will allow. If you go this route, I wouldn't bother with braiding or twisting.




    I'd also pick up a PIXHAWK for a vehicle like that.

  • Just a thought. Could it be that static electricity is building up and causing interference with the compass? If so, the carbon structure should be grounded.

  • I have looked into the tlogs as well and I cannot find consistency between the three last flights. First flight mag field seems ok, second flight (one short landing in this flight) there are some mag field noise present on the first GO, then it vanishes. Third flight is a mess. The difference between the flights is the time from power up to ARMING. First flight it was sitting long time on ground to ensure good GPS coverage. Should I let it sit on ground longer for each flight?

    I also included the tlog file for the last flight. Any thoughts? 3701828476?profile=original


    2014-09-10 19-45-56.tlog

  • I did three flights tonight. I recalibrated the compass. First two flights, each 9 minutes, loiter worked excellent. Third flight I got toilet bowling when entering loiter. Going back to stabilize and then loiter again the octo just wanted to fly away. At landing I checked hdop of 1.4 and 10 satellites. I included the dataflash log file for the third flight.

    2014-09-10 19-36-02.log

  • Developer


         7 satellites and hdop over 2.0 could definitely lead to bad loiter performance.  Generally 9 satellites and hdop of less than 2.0 is good.

         A flyaway in Loiter could be because of the GPS or it could be a compass issue.  It could be the compass-orientation or compass-external parameters not being set properly.  It could also be interference from the motors.

         Best to include a dataflash log file.

This reply was deleted.