Greetings!

After experiencing problems with my Aircraft Orientation indicator on the DJI Fly app, I came across some posts relating to an interference issue between the RC and the phones compass.

I decided to run a test myself, using my Samsung Galaxy S9+ and an Android compass app available in the Play Store (Compass Galaxy).

Much to my surprise, there is indeed quite significant inteference of the phones compass when it is brought into a position of mounting it onto the RC (pls. see video).



So here are my questions:
- What exactly could be causing this?
- Is this what is causing the Aircraft Orientation indicator not to indicate accurately?
- What effect could this have on the safe operation of the drone, in particular with regards to any return to home (RTH) functions?

Thank you.

I am going to try a gooseneck clamp for my phone to try and avoid have problems as a result of this. Also I want to be able to keep my arms lower while having the screen closer to my face. It clamps to the controller. I haven't flown with it yet but it clamps very well.

DJI Brother Posted at 11-30 19:57
I am going to try a gooseneck clamp for my phone to try and avoid have problems as a result of this. Also I want to be able to keep my arms lower while having the screen closer to my face. It clamps to the controller. I haven't flown with it yet but it clamps very well.

DJI Brother (aka djiuser_RDpozUn4JhQY) ......

Thank you so much for yet another advertisement.

Do you think you could also maybe answer some of my questions?

Thank you.

Only dji knows... and everyone is waiting for their questions to be 'officially' answered. Sorry for the bother.

Well, I tried it on Samsung A50. No effect in 'normal' orientation. However, there is significant shift if I bring controller to what is normally right-upper corner. My guess that's were the compass is. It looks like in your phone it's in the wrong place. Not anyone's fault. Don't know if this affects Fly App, sorry.

I am somewhat lost. Have connected the phone via 1 m cable to the controller and looked for changes in the orientation display when I altered the controller position  and left the phone in a fixed position. The controller movements did not change anything on the display and no interference with the phones compass was observed. Only the phone orientation changed the display. So I concluded that it is only the phone compass that is compared to the orientation data from the drone. My big surprise is that exposing the phone compass to a permanent magnet does not change the orientation display. only rotating the phone or the drone does. I wonder now whether the accelerometers in the phone play a role as well.
I am truly flabbergasted and wished that DJI would provide some serious explanation of the factors that affect the real time orientation display.

Wow is over 90 degrees change...
Try rotating the phone 180 degrees ans see what happens?
I'm using also a tablet holder like mentioned before, more comfortable and it may help.

InspektorGadjet Posted at 12-1 03:18
Wow is over 90 degrees change...
Try rotating the phone 180 degrees ans see what happens?
I'm using also a tablet holder like mentioned before, more comfortable and it may help.

I rotated the phone 180 degrees, that is with the top of the phone to the left of the RC and the bottom to the right, and although there was still a bit of interference, it was significantly less.

I too have a phone/tablet holder, and this too seems to reduce the problem. I might have to start using it.

To be honest the compass in my iphone seems to go crazy anyway with anything big, metal, etc.
Check with the holder and let us know the results.

That's not surprising! Phones have a built-in magnetometer (sort of compass) in order to provide the compass function. When you take a standard compass and you approach to some magnetic element (aka a piece of steel) you get a deviation in its indication; with a magnetometer is somehow the same.
Try to approach your phone to a piece of steel and you'll notice the same effect you have with RC.
By the way, I think that your MM is not using at all the phone compass, otherwise how could it RTH when it lose the RC signal? Definitely it has a compass system itsell (probably associated to its GPS sensor).

Have a nice new test!

emaneg Posted at 12-1 06:42
That's not surprising! Phones have a built-in magnetometer (sort of compass) in order to provide the compass function. When you take a standard compass and you approach to some magnetic element (aka a piece of steel) you get a deviation in its indication; with a magnetometer is somehow the same.
Try to approach your phone to a piece of steel and you'll notice the same effect you have with RC.
By the way, I think that your MM is not using at all the phone compass, otherwise how could it RTH when it lose the RC signal? Definitely it has a compass system itsell (probably associated to its GPS sensor).

Thank you for your response.

Yes, I do have knowledge of how a compass works.

However, I did not think the deviation caused by the RC would be so much.

It is good to know that the MM does not rely on the phones compass!

Gray Volk Posted at 11-30 21:52
Well, I tried it on Samsung A50. No effect in 'normal' orientation. However, there is significant shift if I bring controller to what is normally right-upper corner. My guess that's were the compass is. It looks like in your phone it's in the wrong place. Not anyone's fault. Don't know if this affects Fly App, sorry.

Hi. I just want ask if A50 is working fine with the mavic mini because im planning to buy one. Did you encounter any issues so far? I will really appreciate your response. Thank you so much.

That's interesting, as the Samsung Galaxy S9+ is on the approved list.  If the DJI Fly App relies on the phone's compass, it needs to be rock-solid in every orientation of the phone, both landscape and portrait, as well as any angle relative to the horizon.  How was this solved with all the other DJI drones?

After doing some tests yesterday I managed to get correct reading of the compass.

Basically, with iOS Compass APP opened, the DI FLY APP did get the correct orientation, but as soon as I move the RC the arrow would go from Front to indicate the AC is behind me (which is wrong). Since the compass keep telling the AC is behind no matter what, I swapped my phone 180 degrees... and it works...

Now the USB connector of the phone is at the right, the orientation shows correctly as long I keep my body in the same position, that means let say I take off with my body facing north, I have to keep my body in that direction for the whole flight duration or the orientation is wrong.
Hope it helps!

I think this video will explain what is going on with the compass issue.  Just want to warn Mavic Mini owners - it's going to break your heart and I don't think you can get your money back

https://youtu.be/DTwWgsvImgo

djiuser_PzAAe5ng30ya Posted at 2019-12-8 18:52
Hi. I just want ask if A50 is working fine with the mavic mini because im planning to buy one. Did you encounter any issues so far? I will really appreciate your response. Thank you so much.

Hi, u know if Samsung a50 work fine with mavic mini now? I interested to buy. Thanks you

djiuser_PzAAe5ng30ya Posted at 2019-12-8 18:52
Hi. I just want ask if A50 is working fine with the mavic mini because im planning to buy one. Did you encounter any issues so far? I will really appreciate your response. Thank you so much.

Hi you know if Samsung a50 work fine with mavic mini now? I interested to buy one. Thanks you

The answers to your 3 questions is ...

-This is caused by interference coming from the sticks which movements is registered by means of magnets. This together with an unfavorable placements of your phones magnetometer.

-Yes, this is what's causing the wrong indication in your Fly app as the reading comes from the phone compass.

-This error will not affect you Mini at all or any RTH functions as the Mini has it's own compass. The Fly app indicator main purpose is to show you the RC pointing direction relative where the Mini are up in the sky so you can point the antennas correctly for the most optimum signal strength.

... you mainly have 4 alternatives to remedy this:

1. Try to position your phone differently in the RC holder (rotate 180 degrees)
2. Use a additional holder that gives more distance between the RC & phone
3. Try another phone with a more favorable location of it's magnetometer
4. Calibrate your phone's compass while it's attached to the RC (the phone compass will then be off when used without the interference from the RC)

I have the same results on an iPhone 6+. I usually use a phone extension bracket. I've also just opened the compass app while connected and that seems to recalibrate the phone compass. One other thing to watch out for are watch bands...mine has a strong magnet in the band.

I wonder if recalibrating the PHONES compass with the controller present and running might help. I will test this.

Before this post goes viral again.

YOU HAVE TO CALIBRATE the Mobile Device compass after it has been attached to the Remote Control.

The source of the interference is the 18650 cell inside the RC, there's plenty of evidence on the net about it.

To calibrate the MD+RC assembly move it in a figure of 8 pattern or rotate in all 3 axys, much like if it was an aircraft consider yaw, pitch, roll.

m80116 Posted at 2-15 12:36
Before this post goes viral again.

YOU HAVE TO CALIBRATE the Mobile Device compass after it has been attached to the Remote Control.

I think depending on the phone this may or may not work or works then slowly  de-calibrates. For many phones there is no issue but for many there are. I just think you can't rely on the compass indicator in DJI Fly and way better knowing where North is and using the map itself  in DJI Fly to know where your Drone is - assuming you can't see it!    

Tbh in my experience once I've discovered the calibration issue the indicator stays put and and it's no different then say using the compass for Google Maps. Sometimes it's off by some degrees but still giving a coarse orientation that's roughly correct, that can be due to map misalignment, local magnetic field deflection or offset by the device itself.