FPV drones fly in acro mode and do not need nor use magnetometer nor GPS receiver.
You are comparing apples with potatoes. Those are all very different.
In case of DJI their products are fully integrated and tested by experienced engineers. With ArduCopter, inexperienced users place unschielded, untested, electronics near each other and wonder themselves why they interfere with each other.
Someone that knows about EMI and EMC simply will not compare apples with potatoes. He just knows that to get DJI like EMC he needs to spend a couple of dozen hours in a EMC chamber with measument equipment adding filter after filter. After spending the effort, you get the results.
There is no way around that.
Or … You just keep the magnetometer and GPS receiver away from the other electronics like everybody else
Of course we all know it takes alot of trial and errors to get the best outcome for us DIY drones. EMC is not cheap to do. But I’m sure there are basic principle advice people who knows well could give. After all we are here to try to make our drones better or at least learning to.
I actually tried ardusimple. The antenna seems not very affected by the interference. I assume it has to do with the type of antenna as well. In additioa, ardusimple doesn’t have compass which is another added point.
I am suspecting internal antenna design might do just as well but I need to do more test. As the ceramic bottom helps protects it from interference.
I been for days frustrated with the compass going haywire.
When i calibrate it still show high interference.
Even after I add more height to my gps.
Then I realised setting the external compass as first piority then I did the compass motor calibration again it suddenly work.
My internal compass is just too near the power lines but there is nothing I can do.
Once I fix the external compass further away from the power wires it worked.
Hope it helps anyone facing the same problems.
The normal solution Is to just disable the internal compass, it’s too close to power wires to be useful so trying to use it for redundancy is pointless
As @geofrancis stated this has been known since compass’s have been included on flight controllers/IMU’s. The original APM even had a GPS module on it which was soon recognized as a terrible idea. The only reason FC’s have them now is if they use a 9-axis IMU.
i did a lot of tests. also to augment the ground plane with a fine copper mesh. i also shielded the 3d printed holder with EMI tape. nothing works right, just the stylus that raises the GPS. however I’m working with a small 4" copter with LiDAR and sonar. very difficult environment! on a 10" it would be easier. my aim would be to eliminate the delicate GPS stylus
I can see your challenges on such a small platform.
The grounding plate / mesh needs to be overhanging the sides of the GPS receiver or be raised at the sides to be fully effective.
But have to admit I haven’t done any tests as to how much overhang is required.
Just had an idea:
There used to be this special, somewhat conductive 3D printing filament available. - Used to be applied to build cases for electronic equipment /boards.
This may fix the issue with creating a RF shielded platform whilst allowing you to create a small platform compared to other solutions.
One example:
Another thing that can help with your problem:
Twist all wires if possible. Especially the battery and ESC wires. This will reduce RF interference.
I used one of these https://www.3m.com/3M/en_US/p/d/b00050680/
The wire mesh usually I used in any ventilation opening.
Carbon fibers are electrical conductive, technically the emi shield ideally should be grounded with the structure. A small piece can only take insufficient charge. These are just theory, I guess a lot of trial and error is required.
My gps antenna is about 5cm to 8cm from my fc, thats the lowest it can go. I can see your drone is much smaller than mine which makes the gps look exceptionally long.
I have little space because the copter is small, but the gps could still work well even with the printed support. The problem is the EMI lidar for GPS and partly the sonar for the compass.
I shielded the ildar with this conductive tape and with other EMI tape, on the bare part of the lidar PCB I also put a copper mesh connected to the GND of the FC. I shortened and shielded the signal cables as much as possible.
Here you see something:
I had a benewake and lightware usually the casing grounds the lidar. Becareful that it doesn’t have any material that electrically connects to your gps as well.
Gps should be isolated from anything electrical conductive especially you are using electrical conductive pla. It defeats the purpose if your gps antenna are also connected to the source of EMI.
One major issue is the shielding of RX/TX and especially I2C wires to the GPS unit. You need to carefully shield them from the plug at the GPS board to the FC and ground the shielding. I use self adhesive copper tape and solder a wire to GND close to battery.
A digital VTX (HDZero/Walksnail) and even a GoPro also disturbs GPS reception massively. Keep a distance of at least 5-6cm…
Just invest in some shielding around cables, ESC, FC via copper tape or aluminum sheets and your arducopter flies much more robust. GPS lock time reduces drastically, more satellites and no GPS glitches anymore…
