Why magfit's output result is out of parameter range?

I tested with ORIENTATION set to Yaw90Roll180.

It has 11 satellites, and hdop seems good.

When I try magfit, it backed to YAW180 and still out of range.

If uncheck OFFSET, still out of range

@Paul_Paku From above link, I suspect interference. I used to put battery up not below the Mark4 frame. Maybe the power line is closer to the compass.

It looks like you have a 3D printer. So install GPS/Compass on longer, tower like holder.
It’s a good idea to keep angles at 45/90 deg. Calibration can only select position based on 45 deg changes.

I agree, I do not get how the automatic position selector works.It’s always Yaw180 in my case, what is wrong for sure.

MAGFit interface is strange so let me explain.
New users may find it useful, but I’m still uncertain about some fields.

“User sensor:” Asks if you want to export given compass data to output file"
“Orientation” What is expected step for position estimation. (but what Check means?)

“Existing” Asks if you want to plot current offset/calibration from your upload.

Following options here:

My understanding is that you may select the last option or retain first one, provided the results shown below are not significantly different.

It seems first should be grayed as always selected.

From WebTools/MAGFit at main · ArduPilot/WebTools · GitHub, I just reviewed my physics a little bit, below fomula is NOT from code, just theory.

Please correct me if it’s wrong.

@Paul_Paku If so, then it might be the item selected doesn’t affect the sensor much.

I have did changed aligment from YAW180(4) to Yaw90Roll180(10) and do the calibration. I found the configration is automatically switched back to YAW180(4).

And it’s strange that the first time fly after calibration, it can lock Home then the home is flicking: Why home icon is flicking, not locked?

Really have no idea now.

Same here, just wonder if it’s a bug or unimportant information as our compasses can’t be install the same way.

@xfacta As we discussed in other thread, you have mentioned “If the compass is close to the main wiring then a current monitor is essential or the compass will never work properly.”

Would please help to check if this above issue is kind of what you are saying: main wiring, current issue?

Yes, I mean the battery wiring, from the battery and to the ESCs - if the compass is anywhere near that there will be problems.
The compass will need some distance, and preferably a current sensor too.
A typical mast places the compass (and GNSS unit) about 10cm higher - which is good for both.

If that photo further back in this thread is your compass, then I’d expect it to never work properly and just give problems.

Thanks for checking and confirmation. It might be the root cause. I’ll modify the settings(mount) and get back to you all.

EDIT: It’s my drone :frowning:

If you use a 4in1 ESC I think all of them have a current sensor, you just need to make sure it’s connected to the FC and configured correctly.
It’s a very handy for diagnostics and compass calibration.

2 Likes

@xfacta @Paul_Paku @Allister @amilcarlucas

Thank you all. It seems the same as before. It’s the main power line too close to the GPS when I move the battery to the bottom of the copter.

Happy now! :grinning:



1 Like