DJI Mavic, Air and Mini Drones
Friendly, Helpful & Knowledgeable Community
Join Us Now

DIY rear compass cabling fix

Logger

Well-Known Member
Joined
Dec 1, 2016
Messages
1,160
Reactions
622
Age
59
There may be a relatively simple DIY fix to the rear compass noise issue. I have not attempted it yet. However I have discussed the idea with knowledgeable people around here and it does seem that it might work. So let's discuss the idea....

Mavic-Teardown-Rear-cabling.jpg
^^^^^ So here is an early build mavic.

4-mavs-inside-screws-hilited-jpg.5508

^^^^^ and here one that I suspect was manufactured after DJI became aware of the compass EMI issue. Image courtesy of @Thunder-drones who posted this around Christmas time.

It can be seen in the second image how the rear power cabling has been twisted, presumably to reduce EMI/RFI. It has also been re-routed slightly, omitting passage through the two grey retaining lugs at the inside of each yellow circle.

I am quite interested in modifying mine to twist the rear motor power cables in the belief will reduce the rear compass interference.

I see two way to twist the cabling.
a) un solder the rear motor power wires, twist the cable and re solder.
b) Remove rear Arms while keeping cabling attached and rotate them to twist cable then re attach

Option b being less problematic as it does not requiring soldering the motor power wires.

So I have several questions:

1)Has anyone actually removed a rear arm? @Thunder-drones maybe? How difficult is it? Do you think doing so would facilitate the means to twist the power cabling to the main board.

2) Does anyone know precisely where the rear compass is? I have included the image below to show the front compass module and integrated lead.

3) Are their other shielding methodologies that might be easier to achieve DIY?

4) Has anyone who has encountered compass redundancy warnings popped the top on their Mavic. If so would be very interested to see if you have the twist? ie does this fix actually appear to work.

There is no need to point the Warranty implications of doing this. Conversely, sending a Mavic in to DJI opens the possibility of receiving an exchange/refurb unit back which may or may not be desirable.

So I am interested to discuss this idea and here and am seeking suggestions and input...


Mavic-Teardown-inside3.jpg
Front compass highlighted. Note untwisted rear power leads all the way to middle of aircraft.
 
Last edited:
  • Like
Reactions: Mr Spock
1)Has anyone actually removed a rear arm? @Thunder-drones maybe? How difficult is it? Do you think doing so would facilitate the means to twist the power cabling to the main board.

I have removed dozens of them and it's alot easier than the fronts. 6 screws on the top cover, 4 underneath. Pop it up from the back, remove the GPS wire with tweezers or needlenose. Only 3 screws hold in the arm, 1 outside and 2 inside. take them out, then twist the wires to your hearts delight. You will be amazed at how well the rear arm sits into its saddle. It just pops right out straight up, but when you put it back in and screw it down, it stays in really snugly. Seems to me it was designed to be easily replaced with only 3 solder joints and 3 screws.

There are 2 glue spots you will have to break in order to twist them all the way to the main board, but you can glue them back with some shoe goo or other silicone type adhesive.
 
You ripper. Sounds very promising. Can you keep an eye out for the prevalence of twisted rear end vs straight from now on? Do you reckon this premise has legs or a waste of time?
 
Last edited:
  • Like
Reactions: G284L
Just wondering how many twists you will be able to get before the wires are too short or just too tight? Are the individual wires in the factory twisted ones longer than the untwisted ones?
 
  • Like
Reactions: tomrink550 and 1dr1
Could someone explain the noise and any effects other then noise in detail?
 
Just wondering how many twists you will be able to get before the wires are too short or just too tight? Are the individual wires in the factory twisted ones longer than the untwisted ones?

Looks like the straight wires might not have have much play in them to do more than 5 or so twists. Might have to add some wire to the main board then solder the existing motor wire to that. I never try to splice in the middle of a DJI wire because they are coated under the insulation and you have to strip an already thin and fragile wire just to be able to make it take solder.
 
You ripper. Sounds very promising. Can you keep an eye out for the prevalence of twisted rear end vs straight from now on? Do you reckon this premise has legs or a waste of time?

These are 3 of my inventory all bought within the past few weeks. All twisted wiring.

Rear legs missing because thats what people have been breaking most and I replace it for them.

3 mavs rear.jpg
 
There may be a relatively simple DIY fix to the rear compass noise issue. I have not attempted it yet. However I have discussed the idea with knowledgeable people around here and it does seem that it might work. So let's discuss the idea....

View attachment 6696
^^^^^ So here is an early build mavic.

4-mavs-inside-screws-hilited-jpg.5508

^^^^^ and here one that I suspect was manufactured after DJI became aware of the compass EMI issue. Image courtesy of @Thunder-drones who posted this around Christmas time.

It can be seen in the second image how the rear power cabling has been twisted, presumably to reduce EMI/RFI. It has also been re-routed slightly, omitting passage through the two grey retaining lugs at the inside of each yellow circle.

I am quite interested in modifying mine to twist the rear motor power cables in the belief will reduce the rear compass interference.

I see two way to twist the cabling.
a) un solder the rear motor power wires, twist the cable and re solder.
b) Remove rear Arms while keeping cabling attached and rotate them to twist cable then re attach

Option b being less problematic as it does not requiring soldering the motor power wires.

So I have several questions:

1)Has anyone actually removed a rear arm? @Thunder-drones maybe? How difficult is it? Do you think doing so would facilitate the means to twist the power cabling to the main board.

2) Does anyone know precisely where the rear compass is? I have included the image below to show the front compass module and integrated lead.

3) Are their other shielding methodologies that might be easier to achieve DIY?

4) Has anyone who has encountered compass redundancy warnings popped the top on their Mavic. If so would be very interested to see if you have the twist? ie does this fix actually appear to work.

There is no need to point the Warranty implications of doing this. Conversely, sending a Mavic in to DJI opens the possibility of receiving an exchange/refurb unit back which may or may not be desirable.

So I am interested to discuss this idea and here and am seeking suggestions and input...


View attachment 6699
Front compass highlighted. Note untwisted rear power leads all the way to middle of aircraft.

Awesome pics and detail. I've been wondering if a simple DIY fix was possible too. By watching compass interference in app and hunting around with small metal object, I've found the rear compass is in the "upper" body piece (removed in those pics), and it's pegged against the left side. Seems to be very close to rear left side in flat region cutout for blades to pass.

With a tiny bit more ambition I'll take my Mavic apart to try this.
 
...By watching compass interference in app and hunting around with small metal object, I've found the rear compass is in the "upper" body piece (removed in those pics), and it's pegged against the left side. Seems to be very close to rear left side in flat region cutout for blades to pass..
Now that would makes sense. So uses the same connection as the GPS I gather.
 
Could someone explain the noise and any effects other then noise in detail?
-------------------
Mavic - Flight Data Specialist
-------------------
Mavic - Flight Data Specialist? :p

PS I will defer to @BudWalker . He is the man to ask.
 
Last edited:
Mavic - Flight Data Specialist? :p

PS I will defer to @BudWalker . He is the man to ask.

Make sure to raise your hand and ask him politely when it's ok to eat, too.

Btw, not sure why your panties got twisted to begin with. I asked the question you quoted because you posted a thread as if everyone knew about a compass noise issue. I know I had never heard of it until trying to understand what it was you were wanting others to get involved in.
 
  • Like
Reactions: kentfin
These are 3 of my inventory all bought within the past few weeks. All twisted wiring.

Rear legs missing because thats what people have been breaking most and I replace it for them.

View attachment 6704
Thanks for that. So seemingly if I open mine and find it to be in non twisted config there we will be no good reason not to try an emulate those.

Keep the feedback and suggestions coming.
 
This thread is on the right track! As a quick late at night investigation, I took the single screw out from each rear arm near the pivot without taking anything else apart. I thought, if there is enough wire slack maybe I could pop the arm off, give a few rotations, and put back. What I found is there was barely enough wire slack to slide the arm off and see a tiny bit of wire. Not enough that I felt comfortable spinning the arm. (maybe could have though) I pushed each arm back on, tightened the screw, and went for a flight for no good reason. What I found was that no matter how hard I tried, I could not get compass error and could not get redundancy switch. Even going sport mode, full elevator + full throttle at same time, no compass error. Now in the past I would easily have gotten compass switch every time. I proceeded to fly two full batteries in sport mode, not able to get compass to switch.

Now then using @BudWalker 's miraculous and life saving CsvView, I compared the rear compass noise under full speed sport mode to previous .dat captures I had. The plots are below, and you can see previously I had about 250 counts pk2pk on rear magZ.Back and I now have about 150 counts pk2pk.

This tells me the compass interference is highly sensitive to the routing of the small amount of wire slack inside the Mavic body. Most likely all I did was slightly move the wire by pulling on it. I'm not sure my "fix" will last, and you can see front compass is still much cleaner than rear. However I'd be very confident now that just adding some twists to the rear motor wires would totally fix the issue.

Old .dat prior to my wire pull (full speed sport mode)
upload_2017-2-19_23-30-40.png


New flight after my wire pull (full speed sport mode)
upload_2017-2-19_23-33-40.png
 

Attachments

  • upload_2017-2-19_23-32-14.png
    upload_2017-2-19_23-32-14.png
    54.9 KB · Views: 260
Ok, it's late but I couldn't stop there. I removed each rear arm again with the single screw. I pulled out gently as far as I could until it could freely rotate. I gave each arm 3x rotations clockwise allowing wire to twist, and then replaced. I went back and flew full speed sport mode. Again no compass switch. Looking at the .dat with CsvView the magZ.Back is now down to about 100 counts pk2pk. It also is not that different from the front compass at this point. Thank you @Logger and @Thunder-drones for the wonderful insight you have given on this issue!

New flight after 3x rear arm rotations (full speed sport mode)
upload_2017-2-20_0-12-49.png
 
  • Like
Reactions: Mr Spock
I had to go further. I pulled off the rear arms again and gave them 2x more clockwise twists. This makes 5x total. At this point, the last rotation was feeling fairly tight and I was putting some strain on the wires that made me uncomfortable. I screwed the arms back on and went flying in sport mode. The plot below from CsvView shows the further improvement was very mild. My magZ.Back noise is now down to about 75 counts pk2pk. If I gave a recommendation, I'd say 3-4 turns gave most of the improvement without feeling too tight.

New flight after 5x (total) rear arm rotations (full speed sport mode)
upload_2017-2-20_1-1-40.png
 
Now do we know from which serial number on these twisted wires were applied in production?`

Do we know the production date from the serial number anyway??
 
Wow @ScrappyMavic you have been busy!

Those results look terrific. I cant wait to try it with mine now. I think I might go the whole hog though and pull the top. Given we have @Thunder-drones clear and simple instruction on how to do so. I know you have been careful with the number of turns. Does it have any discernible effect on the folding and unfolding the rear arms?

Also how old is your unit - approx delivery date might be of use. Mine is a start of NOV delivery. I will pop my lid tomorrow and report back once ascertain if I am straight or twisted :p

@Egika it would be interesting to know wouldn't it. At this point I am thinking earl/mid DEC deliveries based partly on when the issues were first discovered and when @Thunder-drones posted the image of the twisty unit. But I am only guessing.
 
Wow @ScrappyMavic you have been busy!

Those results look terrific. I cant wait to try it with mine now. I think I might go the whole hog though and pull the top. Given we have @Thunder-drones clear and simple instruction on how to do so. I know you have been careful with the number of turns. Does it have any discernible effect on the folding and unfolding the rear arms?

Also how old is your unit - approx delivery date might be of use. Mine is a start of NOV delivery. I will pop my lid tomorrow and report back once ascertain if I am straight or twisted :p

@Egika it would be interesting to know wouldn't it. At this point I am thinking earl/mid DEC deliveries based partly on when the issues were first discovered and when @Thunder-drones posted the image of the twisty unit. But I am only guessing.

The folding of the arms is just fine after the rotations. The wire is tight and slack used up only when the arm cylinder is pulled out of it's socket. It's here you have to rotate carefully and worry about wire tightness. Once you slide the cylinder back into it's socket, there is plenty of slack for normal arm folding. The arm cylinder has to slide about 0.5-1.0 inches out of it's socket before you can freely rotate the arm. So the reassembled state is perfectly fine.

My Mavic shipped on Nov 14.
 
  • Like
Reactions: Mr Spock
Mine was purchsed on January 5th and it also has 200 units PP of noise when flying in sports mode.
I have not had any problems with my Mavic so I am not sure if I want to pull it apart just yet.
It is kind of surprising that engineers of the caliber and experience that DJI uses did not think of twisting the wires from the start.
 
Last edited:
  • Like
Reactions: Mr Spock and _milo_

DJI Drone Deals

New Threads

Forum statistics

Threads
130,585
Messages
1,554,095
Members
159,585
Latest member
maniac2000