BAD DCM Heading error with AC 3.2

I have a F450 quad with APM 2.6 AC3.2. Recently I'm getting BAD DCM Heading error and the copter goes into LAND mode. This is happening very frequently. I have mounted my GPS Compass on a mast. I didn't face this problem earlier. Compassmot showed mag interference of 3%. So what is causing this? Pls help me in identifying problem.

Bad Heading.jpg

2014-12-24 12-00-29.log

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

Join diydrones

Email me when people reply –

Replies

          • Did another run after recalibrating ( offsets -68, -31, 43 ). The wiring seems to be OK. Still getting frequent glitches. The attached log shows several Err: EKF_Check-2 followed by Err: EFK_Check-0 

            Is the only option to change the h/w ? ( I have a GPS+Mag external unit )

            2015-04-19 18-57-18.log

            https://storage.ning.com/topology/rest/1.0/file/get/3702838762?profile=original
            • If I had to guess I would say you have the same problem I have - which is compass periodically going unhealty. Problem is in your setup there is no backup so the FC is forced to use the unhealthy compass. But compass logs will tell a lot I think.

              • Andy - This is what I am going to do:

                1. Have already raised the GPS+Compass unit a few cms up.

                2. Basis Cala's suggestion, demag the bolts if required

                3. Double check orientation

                4. Rerun calibration and check for any lags

                5. Recheck calibration with external compass and align North to North

                6. Run compassmot again

                7. Enable dataflash logs to capture compass

                8. Run a flight test

                9. Set MPU Filter to 20Hz and do another test ( one at a time to follow ceteris paribus principle )

                10. Report back ( while I will do rest of the things during the week, flight test will be weekend :-( unfortunately )

                BTW, are you still having the same problems?

                • #2 - Bolts are OK. Not magnetised.

                  #3 - Orientation is fine - double checked.

                  #4 - Did compass calibration again. Did not notice any lag. Offsets -60, -34, 43 ( previous -63, -31, 43)

                  #5 - North (on DCM) to North on my phone compass matches

                  #6 - Ran compassmot again. Results attached. The spike is when I bring the throttle down sharp.

                  Rest when I do a flight test tomorrow.

                  I also notice that the ErrYaw builds up slowly and then EKF error is thrown and ErrYaw comes down and the whole thing repeats. I wonder why this is building up.. 

                  @Andy - I am not able to reply to your posts. The reply link does not appear. Thx

                  CompassMot Calibration.png

                  ErrYaw Log 18.png

                  • I borrowed a GPS+DCM and I am still getting a DCM Bad Heading and EKF_CHECK=2 errors!!

                    Needless to say I did a compass recalibration and compassmot again.

                    Do I have a bad APM? or is it a software / config issue?

                    The GPS/DCM is raised up 10cm

                  • Post above, took her in the air. Pretty OK with AltHold. Switched to Land and it started to descend nicely and then suddenly tilted, banked and began to drift until I took control in stabilize. Log with MAG on attached  (23.log ). ErrYaw has come down but hit a high soon after Land.

                    @Andy - I am printing a 3D GPS/MAG "pole". Will do another test after raising the mast further.

                    Compassmot also attached from previous post/calibration before the flight. Spike you see is when I bring the throttle down fast.

                    2015-04-25 14-05-23.log

                    CompassMot Calibration.png

                  • @Cala/Andy - I am not able to reply to your posts So the thread will go a bit haywire.

                    I just did a static test i.e. quad is held secure ( not in flight ) and I ran the throttle up with props down and shifted for a short while. The DCM line stays rock solid on MP (video attached). ErrYaw is literally zero ( log attached with MAG on). 

                    I will take her out for a flight test and see what is going on and post the logs etc.

                    @Andy - Let me do one test and then see if putting the GPS+MAG on a pole will help. I will have to 3D print a "pole".

                    DCM Static Test.mp4

                    2015-04-25 09-22-09 1.bin.log

                  • ErrYaw builds up for me when I lose my primary compass and it fails over to the pixhawk compass which has a lot of interference. My guess is that this is the issue, which is why it would be nice to see compass values for in flight. Caveat - I have seen other people with similar log profiles where interference is not at work, it does make me wonder whether there is some software issue at play. That said when my external compass works, my ErrYaw is very low.

                    Can you put the compass/GPS on a pole say? Mine is about 15cm above the ESCs.

                  • If, with extreme carefull you acelerate the copter on the floor, better putting your props facing down like compassmot and you observe on first MPlanner the copter and compass line when you rise your trottle what happens?

                • I just realized that I was not getting this issue in AC 3.1.5 but I had not flown that much so difficult to say with certainty.

                  I did find a discussion on DIY drones here 

                  Could this be a 3.2.1 problem?

This reply was deleted.

Activity