I have to wonder why they are using the compas information at all. The phones they are recomending have pretty decent gps so they know where the RC is, and the AC has its own gps that is constantly feeding telemetry back. With that info you can calculate and polt the direction between the 2 sets of gps co-ordinates. In my longer range fpv days i built an antenna tracker that was able to do this with a single telemetry link feeding gps data and the antenna tracker on a fixed location. this was processed by an arduino so this is no major processing over head.only reason I could see for using the phones compas would be to cover devices with out gps. but then when did you last see a devive that had magnetometers but no gps?

laazarus@gmail.com Posted at 1-9 10:10
I have to wonder why they are using the compas information at all. The phones they are recomending have pretty decent gps so they know where the RC is, and the AC has its own gps that is constantly feeding telemetry back. With that info you can calculate and polt the direction between the 2 sets of gps co-ordinates. In my longer range fpv days i built an antenna tracker that was able to do this with a single telemetry link feeding gps data and the antenna tracker on a fixed location. this was processed by an arduino so this is no major processing over head.

The idea is to align antennas.  A pilot in one GPS spot can face any direction.  A compass is needed in addition to GPS.

Minnesota Posted at 1-9 10:13
The idea is to align antennas.  A pilot in one GPS spot can face any direction.  A compass is needed in addition to GPS.

no it isnt. the purpose of the antena tracker I described is to keep a directional antenna pointing at a model in the air with out any interaction from the pilot. It did this very well and I can assure you there was not a compass involved at any point. just the known position of the ground station and the telemetry stream from the model.

laazarus@gmail.com Posted at 1-9 10:34
no it isnt. the purpose of the antena tracker I described is to keep a directional antenna pointing at a model in the air with out any interaction from the pilot. It did this very well and I can assure you there was not a compass involved at any point. just the known position of the ground station and the telemetry stream from the model.

I need to correct my self. my comparrison with my long range setup did miss a vector. (its been a while since its flown with all the legal stuff relating to fpv now) . My ground station was fixed and part of the setup once its location was set was to manually point it at the model. this is the bit that would need a compass if it were to start moving about and changing orientation. The only thing it could track reliably if it moved location would be elivation. So Minnesota you were right (and I need to sit and think about my stories of the past more before I post them).

I make two tests with Samsung Note 9 and iPhone 7 and Dji Mavic Air controller. Note 9 is influenced magnets in controller, but only when is telephone USB-C connector is orienting to cable from mavic controller. iPhone 7 is not influenced in any orientation. Situation is documented on my videos.

laazarus@gmail.com Posted at 1-9 10:10
I have to wonder why they are using the compas information at all. The phones they are recomending have pretty decent gps so they know where the RC is, and the AC has its own gps that is constantly feeding telemetry back. With that info you can calculate and polt the direction between the 2 sets of gps co-ordinates. In my longer range fpv days i built an antenna tracker that was able to do this with a single telemetry link feeding gps data and the antenna tracker on a fixed location. this was processed by an arduino so this is no major processing over head.only reason I could see for using the phones compas would be to cover devices with out gps. but then when did you last see a devive that had magnetometers but no gps?

Without compass craft can’t no what direction it’s traveling in, this is the trusted system dji have always used, there has been some progress with other systems which we are now seeing coming into some drones and maybe the day will come when no compass will be needed.

Great comments and feedback DJI'ers!!

Has anyone tried setting their device to use true north as opposed to using *default (magnetic north) .  I am going to give this a test now and will post back .

Minnesota Posted at 1-9 10:13
The idea is to align antennas.  A pilot in one GPS spot can face any direction.  A compass is needed in addition to GPS.

It goes along with a person walking with GPS unit which has a built-in compass.
You can be walking North with GPS unit pointing East.  The GPS will show your track moving North, but compass of GPS will show it pointed to East.  
Similar, a drone can be flying North with its camera facing East.  The flight App will show drone's track moving North, but show drone is facing East.

Of interest, on elimination of magnetic compass.

With large drones (2-meters or more), it is possible to have two GPS receivers with separate GPS antennas at opposite ends of drone, and determine from difference of GPS signals which way drone is pointing to less than 1-degree.
A few tech hobbyists are working on 1-meter spacing, and discussing possibility of 0.5-meter achieving better than 3-degree accuracy.

OneWireAT1 Posted at 1-9 21:12
Great comments and feedback DJI'ers!!

Has anyone tried setting their device to use true north as opposed to using *default (magnetic north) .  I am going to give this a test now and will post back .

I tried to set True north in my Iphone 6, but it looks it locks the compass and makes it not "work"... maybe I'm wrong.

Arcicorsa Posted at 1-9 11:39
I make two tests with Samsung Note 9 and iPhone 7 and Dji Mavic Air controller. Note 9 is influenced magnets in controller, but only when is telephone USB-C connector is orienting to cable from mavic controller. iPhone 7 is not influenced in any orientation. Situation is documented on my videos.

iPhone 7

Interesting tests.
Thanks a lot

Mgozer Posted at 1-10 01:18
Interesting tests.
Thanks a lot

Setting the "true north" to off iPhone 6s worked for me.
Mini and the controller are on .400 version.


I do the compass calibration one time when i come to the location.

Read almost all of the comments.  The issue posted here is well known by the amount of youtube videos posting the same issue.  The culprit is the Mavic Mini Remote Control.  My compass on my Galaxy 8+ worst just fine.  I can google my address and find North and my compass will point straight north in my home.  Now when I place the Remote control near the phone the compass goes off by at least 30 degrees or more!  The Remote does not even have to be turned ON.

The bottom line here is that the Mavic Mini remote controller affects the compass of the cell phone.  On or off it does not matter.  The culprit is the material inside the controller - obviously magnetic.  There are a few youtube videos out there with Mavic Mini owners demonstrating how the remote affects the compass in the phone.

So - what is DJI going to do about this?  Major flaw!  And WILL lead to unsafe flying.

hallmark007 Posted at 1-9 11:42
Without compass craft can’t no what direction it’s traveling in, this is the trusted system dji have always used, there has been some progress with other systems which we are now seeing coming into some drones and maybe the day will come when no compass will be needed.

If I understand it correctly the new Skydio 2, which has a wait list til about the end of this year if you want to order one, does not have a compass. That is why you are seeing all these crazy videos of it flying right down the opening of a huge metal bridge.

Yesterday I noticed that the iphone 6S is not affected by this compass interference (vs Mavic Mini remote).
Funny that the iphone 6 is instead affected a lot!

OSM Posted at 1-11 20:02
Read almost all of the comments.  The issue posted here is well known by the amount of youtube videos posting the same issue.  The culprit is the Mavic Mini Remote Control. My compass on my Galaxy 8+ worst just fine.  I can google my address and find North and my compass will point straight north in my home.  Now when I place the Remote control near the phone the compass goes off by at least 30 degrees or more!  The Remote does not even have to be turned ON.

The bottom line here is that the Mavic Mini remote controller affects the compass of the cell phone.  On or off it does not matter.  The culprit is the material inside the controller - obviously magnetic.  There are a few youtube videos out there with Mavic Mini owners demonstrating how the remote affects the compass in the phone.

So - what is DJI going to do about this?  Major flaw!  And WILL lead to unsafe flying.

I agree with you that DJI should solve the issue.. if it happens (as it does) even with RC off that means that it's the RC itself that creates magnetic "interference"

... so what could DJI do at SW/FW level?
NOTHING! they should redesign (assuming it was possible) the RC and replace it to everyone.. for free

Being realistic.. it cannot&will never happen..

What do you all think about?

spanzetta Posted at 1-19 01:29
I agree with you that DJI should solve the issue.. if it happens (as it does) even with RC off that means that it's the RC itself that creates magnetic "interference"

... so what could DJI do at SW/FW level?

Spanzetta, your are 100% right in saying that DJI will NOT redesign the RC and replace it for free.  This is just another flaw that Quality Control did not detect.  Unfortunately, it is a major flaw and could impact safety of flight.

OSM Posted at 2020-1-11 20:02
Read almost all of the comments.  The issue posted here is well known by the amount of youtube videos posting the same issue.  The culprit is the Mavic Mini Remote Control.  My compass on my Galaxy 8+ worst just fine.  I can google my address and find North and my compass will point straight north in my home.  Now when I place the Remote control near the phone the compass goes off by at least 30 degrees or more!  The Remote does not even have to be turned ON.

The bottom line here is that the Mavic Mini remote controller affects the compass of the cell phone.  On or off it does not matter.  The culprit is the material inside the controller - obviously magnetic.  There are a few youtube videos out there with Mavic Mini owners demonstrating how the remote affects the compass in the phone.

No it is not interference because now some air2 and mini2 with "smart controllers" also have this issue. And how come they all have exact 90° deviation in various kinds of situation?

It is and easy to prove. Just calibrate your phones compass away from the controller or anything metallic. Then point your compass on your phone to North then gradually bring your phone near and around your controller. On a lot of phones you will see the compass deflect as it is affected by the controller.

It just makes the antennae orientation feature totally unreliable unless your phone seems to be shielded some how...

Three things. Might get away from metal viechle , powerlinrs.
Reverse phone in mount in controller.
Wonder if different phones have deferred locations inside where the compas is. Some people have problem and some don't.
DougWonder if the latest update is still doing it.
To cold here to go outside and check here.
I just seen this is an old post.

What about SC? It doesn't have portrait/landscape mode, and nothing to" mount"
How come some SC users have this problem and some don't?

I think Hallmark and others have said they have had success getting correct orientation by calibrating their compass while it was in the RC holder. I have tried that in the past with my Samsung S9 but it didn't work.

However today I had repeated success using that technique after trying a few things.

The difference today was I got a UE Boom Speaker (you could use a small magnet) and held it near my phone until my compass app...

https://play.google.com/store/ap ... ompass&hl=en-GB

...showed I was totally out of whack and needed to calibrate (A green infinity symbol shows on the bottom)

Then keeping the phone in the RC cradle I did the figure 8's and afterwards phone was calibrated correctly and stayed that way.

If I turn everything off you have to do it again but for the sake of carrying a small magnet and repeating this at the start of each flight I can have correct compass orientation.... so pretty happy.

May help others having issues with this.

So I need to rescind my previous post :-(

Had another session today and results were just not consistent not matter what I tried.  So I'm back to my original thinking with the Mavic Mini and Samsung S9. Unfortunately if using the RC cradle with phone flipped or not and with compass calibrated while the phone is in the cradle you seem to sadly not be able to trust that the drone direction indicator (if you can see it!) will allow you to accurately point your RC antennas at the drone.

A different phone or phone holder may have successful results.

I fly VOS 95% of the time so it's not a biggy but is a little annoying when tools don't work like they should.

Did anyone find a cure.  I am so overly upset off right now.  I tried to explain to the slow people at tech support what was happening.  He acted like this was new to them and wanted me to send a video.  I told them to look it up.  He said he needed one of mine doing it.  I would love to send this turd bucket back to them.  Mine does the same thing yours is doing