That is something I was pranking with in my last flight and noticed. But knowing that it's the device that decides the orientation doesn't really give us a fix. It just tells us where the issue is. Maybe some pretty smart forum member here can use this knowledge to come up with a fix or maybe DJI will fix it in their next update.

Wouldn't standing next to a truck mess with the compass in the phone?

I've read somewhere on this forum, a post from a DJI representative, where they advised calibrating the phone compass with regard to this issue.

I'm guessing that the display of relative direction relies on the GPS coordinates of the phone and the drone, as well as the compass reading from the phone.

If the phone's compass is incorrect, then the display will be incorrect.

I haven't played with my phone's compass very much, but when I have used it, it was quite accurate.  I wonder if the DJI Radio Controller itself might contain some ferrous metals and require us to do a compass calibration of the phone while the phone is attached to the RC to tweak that error out.

And, of course, be sure we're well away from any sources of magnetic influence, such as a steel vehicle, building, etc.

You need to calibrate your phone . Don’t forget cold weather will also effect your phone. Also check small triangle on map lower left hand corner .

I have to add this:  I just played with the compass in my phone, without the drone or its RC.

The compass is accurate when I hold the phone in the usual vertical orientation (portrait mode).

And when I hold it in landscape mode, it is accurate with the right side of the phone away from me, but reads about 90 degrees off when held with the left side away from me!  (It shows north where east really is).

To mount my phone in the DJI remote, I have to place it with its USB connector to the left, and that places it with its left side away from me, which is the orientation that yields the error.

So I think I'd need to calibrate the phone compass with it mounted in the RC to have any use of this feature.  And then I'd need to recalibrate it again with it held "normally" to have the compass work under normal circumstances.

hallmark007 Posted at 12-12 11:48
You need to calibrate your phone . Don’t forget cold weather will also effect your phone. Also check small triangle on map lower left hand corner .

There is nothing wrong with the phone. it shows the right north. The problem is that the reference point is the top of the phone so when the phone is in  landscape mode, assuming you turn your device counterclockwise, the top part of the phone will point to the west. The DJI app don't take this into account so basically it thinks that you turn your remote 90 degrees to the left.

the1shark Posted at 12-12 12:06
There is nothing wrong with the phone. it shows the right north. The problem is that the reference point is the top of the phone so when the phone is in  landscape mode, assuming you turn your device counterclockwise, the top part of the phone will point to the west. The DJI app don't take this into account so basically it thinks that you turn your remote 90 degrees to the left.

All I know is I had similar problem, it was corrected by correctly calibrating my phone .

the1shark Posted at 12-12 12:06
There is nothing wrong with the phone. it shows the right north. The problem is that the reference point is the top of the phone so when the phone is in  landscape mode, assuming you turn your device counterclockwise, the top part of the phone will point to the west. The DJI app don't take this into account so basically it thinks that you turn your remote 90 degrees to the left.

I think it is more subtle and complex than that.

My phone reorients its compass display to always point north as I rotate the phone with the phone held flat with respect to the earth (phone plane parallel to the surface of the earth).

However, when I hold the phone with its plane at an angle of around 45 degrees from the plane of the earth's surface, that's when I see the error with the "left side up" AKA "left side away from me" orientation.

And we usually hold the phone at about a 45 degree angle when using the RC.  The phone would be hard to see if we held it flat the way we do when using the compass.  We tend to hold the phone the same way we hold a real compass when hiking, etc, when using it as a compass.

hallmark007 Posted at 12-12 12:11
All I know is I had similar problem, it was corrected by correctly calibrating my phone .

Ok I gave up. Can you or someone else please guide me how to calibrate an Iphone compass ?

Well, do it with the phone in the remote control, and holding the same way you would hold the RC when flying.

That's what I'm saying.

I don't know Apple, so I can't advise how to cal it's compass.

Also, do it out away from any ferrous metals, cars, or other magnetic structures.

That's interesting... weather is bad at the momment and I havent had a chance to test but I will as soon as I can.
I wonder if performing a calibration of the compass, while the phone is attached to the controller would have any effect in the behaviour...

InspektorGadjet Posted at 12-12 12:19
That's interesting... weather is bad at the momment and I havent had a chance to test but I will as soon as I can.
I wonder if performing a calibration of the compass, while the phone is attached to the controller would have any effect in the behaviour...

Hopefully, it will help.

I can't try any of this until later, when I can get out into an open area away from buildings and such.  But I'll make some tests then.

hallmark007 Posted at 12-12 11:48
You need to calibrate your phone . Don’t forget cold weather will also effect your phone. Also check small triangle on map lower left hand corner .

Hi all. Earlier I commented that the same thing happened to me, and in my town it is not so cold nor there is snow, as much we can be in 6 degrees. Also comment that the phone is NOT THE PROBLEM, because to me the compass works perfectly with the Phantom 3 and DJI GO, but with the mini does the same as in the video of the partner, positioning does not work.
There is no worse blind than the one who does not want to see, and this is a problem of DJI and its Firmware.

the1shark Posted at 12-12 12:15
Ok I gave up. Can you or someone else please guide me how to calibrate an Iphone compass ?

Tap settings
Privacy
Location services
System services

You’ll find compass calibration there, it may on but this doesn’t mean it is calibrated correctly

I stuff around with this since weeks now. If all calibrations are done and well and the phones gps has position locked and there is some distance between phone location and drone location the position is shown fairly accurate - s o m e t i m e s.
I can never rely on it and doing so would not only invite trouble but break the code of responsible flying. A gauge must be reliable or not be there at all.
And in the orientation matter I cannot even know if it shows correct or wrong. So I have to visually check the position in any way. What's the point of having an unreliable gimmick that only clutters the view, distracts from operation and erodes confidence in the controllability of the aircraft?
Fix or remove it!

JM.REYES Posted at 12-12 13:01
Hi all. Earlier I commented that the same thing happened to me, and in my town it is not so cold nor there is snow, as much we can be in 6 degrees. Also comment that the phone is NOT THE PROBLEM, because to me the compass works perfectly with the Phantom 3 and DJI GO, but with the mini does the same as in the video of the partner, positioning does not work.
There is no worse blind than the one who does not want to see, and this is a problem of DJI and its Firmware.

You say it’s a FW issue, how do you account for those who have a perfectly working radar ?

I’m confused are you saying we are all blind but you are the only one who can see ?

JM.REYES Posted at 12-12 13:01
Hi all. Earlier I commented that the same thing happened to me, and in my town it is not so cold nor there is snow, as much we can be in 6 degrees. Also comment that the phone is NOT THE PROBLEM, because to me the compass works perfectly with the Phantom 3 and DJI GO, but with the mini does the same as in the video of the partner, positioning does not work.
There is no worse blind than the one who does not want to see, and this is a problem of DJI and its Firmware.

Perhaps those other controllers have built in magnetic sensors of their own,  but the one for the Mini relies on the phone.  Just tossing that out there.

hallmark007 Posted at 12-12 14:29
You say it’s a FW issue, how do you account for those who have a perfectly working radar ?

I’m confused are you saying we are all blind but you are the only one who can see ?

No man no ... I mean DJI, because this bug is being reported from the beginning (just analyze the forum and you will see how many people are with the same problem) and DJI seems to look the other way. CAREFUL, I say it seems.
You can make a check, put GOOGLE MAPS, and to show you the way forward to some location, you will see how the compass works well. Forgive if I am wrong, but I think I understand that it is a good way to know if the phone is well calibrated.

JM.REYES Posted at 12-12 14:34
No man no ... I mean DJI, because this bug is being reported from the beginning (just analyze the forum and you will see how many people are with the same problem) and DJI seems to look the other way. CAREFUL, I say it seems.
You can make a check, put GOOGLE MAPS, and to show you the way forward to some location, you will see how the compass works well. Forgive if I am wrong, but I think I understand that it is a good way to know if the phone is well calibrated.

No worries I just hope all get sorted, dji have already said numerous times on this forum that they would issue FW in late December, and I expect they will. Good luck .

It has been established that the  Mavic Mini's RC affects some phones compass, for whatever reason. I have done some tests, and started a couple of threads as well as posted in other threads regarding this. I have calibrated everything, including the phones compass, and the phones compass works normally and accurately on its own. But the problem persists.

Please see this thread: Mavic Mini RC Affecting Compass on Samsung Phone

Unfortunately I am yet to get a direct response from DJI.

StubbornSwiss Posted at 12-12 18:29
It has been established that the  Mavic Mini's RC affects some phones compass, for whatever reason. I have done some tests, and started a couple of threads as well as posted in other threads regarding this. I have calibrated everything, including the phones compass, and the phones compass works normally and accurately on its own. But the problem persists.

Please see this thread: Mavic Mini RC Affecting Compass on Samsung Phone

Does the Mavic Mini R.C. use magnets internally on Joysticks, as Mavic Pro R.C. does?

Wondering if some phones, like Samsung, are more sensitive to MM RC magnets than to MP RC magnets do to positioning of phone and size of MM RC.

Sigmo Posted at 12-12 14:30
Perhaps those other controllers have built in magnetic sensors of their own,  but the one for the Mini relies on the phone.  Just tossing that out there.

Not sure why (say for example) Mavic Pro R.C. would need electronic compass, (magnetic field sensors in XYZ axis) and Mavic Mini R.C. would not.   

If say Mavic Pro R.C. did have a builtin electronic compass, one would expect R.C. would from time to time need it's compass calibrated like Mavic drone does.  However, I have never seen a message about calibrating a compass in MP's R.C., and don't remember seeing any method on how too.

Sigmo Posted at 12-12 11:45
Wouldn't standing next to a truck mess with the compass in the phone?

I've read somewhere on this forum, a post from a DJI representative, where they advised calibrating the phone compass with regard to this issue.

"Wouldn't standing next to a truck mess with the compass in the phone?"

In this case it could.  The guy is standing fairly close to truck.  When he turns, he is moving RC & Phone closer to truck, and that is when compass radar goes wrong.  When he turns back away from truck, compass radar returns to normal.

Something I am wondering about what is on his truck.  Does his truck have electrically adjusted mirrors inside a plastic shell?  Electric D.C. motors may have magnets, and as he turns, RC & Phone get closer to mirrors.

Then again, he does only rotate Phone idenpdent of RC, and Phone does not appear to move closer to truck or truck's mirror.  Pointing back to magnetic inteference between R.C. and Phone.

HedgeTrimmer Posted at 12-12 20:30
Not sure why (say for example) Mavic Pro R.C. would need electronic compass, (magnetic field sensors in XYZ axis) and Mavic Mini R.C. would not.   

If say Mavic Pro R.C. did have a builtin electronic compass, one would expect R.C. would from time to time need it's compass calibrated like Mavic drone does.  However, I have never seen a message about calibrating a compass in MP's R.C., and don't remember seeing any method on how too.

If you can't find any procedure for calibrating an RC's compass, that is good evidence that the RC doesn't have a compass.  So much for THAT idea!  ;)

OK.  So here's the crude Boy Scout analysis.

I just took out, I kid you not, my old Boy Scouts of America National Council compass, probably from the 1960s, made by Silva (a name you compass folks might recognize).  I always had that thing with me when we were doing antenna installations and site surveys for radio paths back in my radio telemetry days.

Anyhow, playing with that compass and the DJI Radio Control unit for the Mini, it does INDEED have something or things that are magnetic inside, and it does affect the compass profoundly.

So just because your phone's compass is calibrated, don't imagine that it will be when it's brought near to your DJI Mini's radio controller.

I took the controller and compass out into the field away from our building and other metal objects just now, and I could get the compass to point south if I wanted to, with it clamped into the arms of the remote control.

So this is not going to be a firmware update fix, I fear!


My incredibly sophisticated test rig!

That's sitting on a huge steel desk, so tests there are not useful, but out in the field next to my work office, you can see the problem very clearly.

Sigmo Posted at 12-12 20:52
If you can't find any procedure for calibrating an RC's compass, that is good evidence that the RC doesn't have a compass.  So much for THAT idea!  ;)

OK.  So here's the crude Boy Scout analysis.

Not surprising, based on Mavic Pro R.C. having internal magnets for joy sticks as input to field sensors.
You can use a magnet to cause joy stick input, even though you don't physically move joy sticks.

If you want a more modern method of detecting magnetic fields.  There are free to very cheap Magnetic Field Apps available for both iOS and Android smartphones.
One of advanages of Magnetic Field Apps is they can (depending on which App) provide field level strength in X,Y,Z axis and/or combination of all three.  Along with ability to chart and/or record.  

Sigmo Posted at 12-12 20:52
If you can't find any procedure for calibrating an RC's compass, that is good evidence that the RC doesn't have a compass.  So much for THAT idea!  ;)

OK.  So here's the crude Boy Scout analysis.

"I took the controller and compass out into the field away from our  building and other metal objects just now, and I could get the compass  to point south if I wanted to, with it clamped into the arms of the  remote control."

At this time I can only suggest a possible work around.  Calibrate your SmartPhone clampled in R.C. as you would when flying your drone.  Once calibrated changing position SmartPhone clamps in R.C. could result in SmartPhone compass being off.


Making sure you when you calibrate your SmartPhone you are not wearing or carrying anything magnetic (Ferrous metal).
When you fly, don't wear or carrying anything magnetic.  Or there is potential to throw SmartPhone's compass calibration off.


Do SmartPhone calibration as you did compass test.  Out in field at least 10-meters from any source of magnetic interference.  Overhead or buried powerlines, burried metal piping, chain link fencing, etc.

HedgeTrimmer Posted at 12-12 22:33
"I took the controller and compass out into the field away from our  building and other metal objects just now, and I could get the compass  to point south if I wanted to, with it clamped into the arms of the  remote control."

At this time I can only suggest a possible work around.  Calibrate your SmartPhone clampled in R.C. as you would when flying your drone.  Once calibrated changing position SmartPhone clamps in R.C. could result in SmartPhone compass being off.

Exactly.  And that's what I recommended doing up earlier in this thread.

Gotta cal the phone compass under the actual use conditions.

Now I'm tempted to rip the controller apart to see what's inside!  ;)  That WILL happen eventually.  I can never resist.

Some phones are affected worse than others, based on where their sensors are located in the phone/tablet.  What is odd is Gorilla man has his phone (Galaxy S7 Edge) on a mount, well away from the joysticks.  His mount is non-metallic.  

In a related thread on Aircraft Orientation ... people with these phones are reporting problems.  They somewhat suggest in appeared with the last FW update.  
Xiaomi mi A2. Android 9
Xiaomi Redmi 5 plus android 8.01
Xiaomi redmi Note 7
Samsung Galaxy S7
iphone8
Samsung Galaxy S10
Samsung Note 10+
Iphone XR

Minnesota Posted at 12-13 00:28
Some phones are affected worse than others, based on where their sensors are located in the phone/tablet.  What is odd is Gorilla man has his phone (Galaxy S7 Edge) on a mount, well away from the joysticks.  His mount is non-metallic.

I've got a mount sort of like that.  I need to play with it and see if it can get the phone far enough away from the remote to make things work well.

But I haven't had good weather to be able to try with my mini and phone to see if there's a problem to begin with!

The mount I have is mostly aluminum, but it does have some steel as well.  If it's the right formula of stainless steel, that may all be relatively non-magnetic, too.  But there are springs in mine, and those are most likely just carbon steel, so the mount might make things worse!

It will be interesting to see if my phone has this problem to begin with, and then see if the mount helps or makes things worse.  I got the mount to use with a tablet of mine, but alas, the tablet is not compatible with the Fly app, anyhow.  But the mount does have the advantage of allowing me to mount my phone without prying it out of its protective case.

I can't center the phone up, though.  Either the power button or the utterly-useless "Bixby" button get's pushed continuously the way it's designed.  I need to machine a special "Note 8" holder that avoids the buttons.

Interesting stuff!

I don't think it's a magnetic problem. As I was trying to explain and as the video implies, the problem is due to the phone landscape mode meaning it's top part is pointing to the East when the North is ahead of you. The DJI app doesn't take this under consideration and the end result is falls reading of a 90 degree offset.
Theoretically if I will mount my IPhone XR vertically, reading will be fine.

HedgeTrimmer Posted at 12-12 22:24
Not surprising, based on Mavic Pro R.C. having internal magnets for joy sticks as input to field sensors.
You can use a magnet to cause joy stick input, even though you don't physically move joy sticks.

Was this an untruth , you posted this after been told that there are magnets in MP RC , but you vehemently disagreed that they could even be moved by normal household magnets, until I posted a video showing this.
You say you tested with very strong magnets and nothing happened, was that not true .


Maybe that app is not correct and measurements where totally wrong, might not be a good idea advising others to use it.

Minnesota Posted at 12-13 00:44
In a related thread on Aircraft Orientation ... people with these phones are reporting problems.  They somewhat suggest in appeared with the last FW update.  
Xiaomi mi A2. Android 9
Xiaomi Redmi 5 plus android 8.01

not really. I've a redmi note 7 and the radar indicator works fairly well (about +- 20° around real position).

I'll do some more checks in order to understand better the radar behaviour

It only takes very small house hold magnets to effect RC sticks, see below.

I've been meaning to do some testing, just haven't had time and I live in a metal box on top of that. I'm relatively new to DJI products, but I did suspect magnetism could affect the RC. Hallmark007's video that he shared demonstrates that perfectly. I actually suspect that the answer is in a firmware update. As the1shark pointed out with his explanation of phone rotation while conntected to the RC, I found the same thing with my last flight. I was in an area where there was no magnetic interferance, or at least none that registered on the app on my phone and the compass heading was off by 90 degrees. Thinking about calibration, I put the phone (iPhone Xr) back in the holder landed the mini, checked the compass app, which was accurate,  but did a calibration while in the RC. This had no postive affect. Of course this is my opinion and could be totally wrong but I think it can be fixed with an update.

hallmark007 Posted at 12-13 04:50
Was this an untruth , you posted this after been told that there are magnets in MP RC , but you vehemently disagreed that they could even be moved by normal household magnets, until I posted a video showing this.
You say you tested with very strong magnets and nothing happened, was that not true .

Do Read what I wrote right after your Outline Box on screen snap.

As for your entire post, re-analyze it and think it through.  I'll give you a chance first, to delete it or modify it.  

LOL, can we talk about compass orientation and how to fix it now?? Please??

JodyB Posted at 12-14 08:49
LOL, can we talk about compass orientation and how to fix it now?? Please??

I fully agree that’s what we are here to discuss, thanks for the reminder .

I actually think both you guys have some really great qualities. Just the tempers, possibly past issues or whatever, really none of my business. To be honest,  there are times where I think throw some gloves on them and put them in a ring and see who walks out, but that wouldn't solve anything here. I really think you guys would make a hella good team if the logic was combined!!!

JodyB Posted at 12-14 09:06
I actually think both you guys have some really great qualities. Just the tempers, possibly past issues or whatever, really none of my business. To be honest,  there are times where I think throw some gloves on them and put them in a ring and see who walks out, but that wouldn't solve anything here. I really think you guys would make a hella good team if the logic was combined!!!

I have to agree here. Both highly helpful and qualified users
I'm quite surprised with DI Forum, I'm personally a very active person in forums, love to learn and participate, show my findings... But some times it gets visceral over here lol
I thought 3D printing nerds, and audiophiles where at the top, but this may be a winner!

Jody thanks for showing your results in the compass calibration with controller I haven't had a chance to test it myself.