RTH sideways - compass calibration?

[kubais]

I am facing problem that when I initiate RTH, MP is not facing directly to home but always some 30-40 degrees sideways.

I tried several compass calibrations in proper environment and it does not seem to get better. When I try to manually correct heading angle during RTH it immediately goes back to original orientation.

 

[rydfree]

When you point the Mavic at a known landmark is the red arrow on the map pointing correctly at the same landmark ?

Has it always shown this behavior ?

 

[sonof40]

Does RTH circumvent flying modes? If it doesn't (I assume it does) then maybe you are in Home Lock or Course Lock. ?

 

[kubais]

I

When you point the Mavic at a known landmark is the red arrow on the map pointing correctly at the same landmark ?

Has it always shown this behavior ?

I have never watched arrow direction on the map, will try it out tomorrow and report back.

 

[kubais]

I have yet to

Does RTH circumvent flying modes? If it doesn't (I assume it does) then maybe you are in Home Lock or Course Lock. ?

I am quite positive RTH circumvent flying modes and I have yet to even try Course lock and other modes. I am simply just flying on basic GPS mode, no "intelligent modes" activated.

 

[kubais]

Still trying to figure out MP flying sideways during RTH.

Could it be somehow related, that my remote controller maxes out on left stick at extreme bottom left position at 86% (max yaw/min altitude - Mode 2)? Tried sticks calibration several times and no changes there. Seems like its hardware problem to me on controller. But that is probably separate problem and has nothing to do with RTH orientation.

 

[dwallersv]

Well, the obvious first question to ask: If you leave it be, does it return home and land at the Home Point?

If it does, go out and reproduce the problem, but let it come home and land. Then grab the logs and post them here so the gurus can take a look. I'm sure someone else will chime in with details of how to download the logs, if you don't know how.

 

[kubais]

Yes, it always returns it its master, only problem is MP not facing home directly which can result in extreme case into collision since sensors won´t detect obstacle.

Other thing that might be connected is also weird sound MP makes when coming home in RTH mode which I suspect is due to not flying straight, but sideways - some motors working more than others, which is something I never hear when flying manually nearby.

 

[scooot]

I am new to this so my proposal may not be possible.To eliminate the controller could you not try a flight in wifi mode just using your phone.A short test flight should verify if the controller is the issue or not.
Personally I don't think I would want to try it myself as I'm not confident enough to trust wifi but it was just a thought.

 

[kubais]

Good idea, never even tried wifi for obvious reasons

 

[dwallersv]

The motor sound definitely may be related to something wrong, however it's not because of some differential loading due to it flying turned at an angle.

Quads are fully symmetrical insofar as flight characteristics are concerned. Any direction can be the "front" -- the only thing that makes the front the front in our case is that's where the camera points, and some sensors. As far as flying goes, all directions are the same mechanically.

 

[kubais]

That's what I thought. But weird motor sound only appears during RTH. Apart from that quad flies and sounds absolutely normally.
I am not new to quads, built plenty on my own and I heard something simular only when accelerometer was totally off or rather when I flew with barometer on and quad was jumping up and down but that is not the case here since it flies absolutely fine.

The sound is kind of pulsating...