I previous shared a photo of the inside of the battery. Here are a few photos of the inside of the controller.


Unsuprisingly the controller is well made. What did surprise me was finding there's nothing inside the right side of the "T" antenna.


The two antanna each have a second BCP with two traces on it. I figure DJI is doing some RF vodoo with these unwired PCBs.


I strongly recommend NOT taking apart your controller. Besides likely voiding your waranty, there are lots of fagile connectors which could easily be damaged. It's also really hard to open these plastic pieces without leaving marks. I openned my controller so you don't have to.
There are lots more photos on my Hackday IO project page. I have more photos I haven't uploaded yet. If any of you have requests for detailed photos of a particular part, let me know what your request is. I'll try to take requested photos if I have time.

Hello there Duane Degn. Good day and thank you for sharing these photos and this information here at DJI Forum. Just a reminder that doing this process might affect the DJI FPV remote controller's performance when in use and can void the warranty of the said remote controller as well. Thank you.

DJI Stephen Posted at 3-21 19:45
Hello there Duane Degn. Good day and thank you for sharing these photos and this information here at DJI Forum. Just a reminder that doing this process might affect the DJI FPV remote controller's performance when in use and can void the warranty of the said remote controller as well. Thank you.

Yes, I realize there are all sorts of reasons NOT to take apart the controller. Hopefully other people will be satisfied with seeing these photos and not take there own controller apart.
I consider my warranty voided by doing this.

Thanks for sharing, its cool to see inside.

Thanks for sharing.  

I've added another log to my Hackday IO project. This log shows details about the small buttons used to detect when a gimbal has been set to manual mode.


As I've mentioned elsewhere, the gimbals use Hall Effect sensors. The magnet used with the sensor can be seen in the photo below. The PCB with the sensor had been removed with this photo was taken. It might be possible to reverse an axis by flipping this magnet. I don't think the power lines of a Hall sensor can be reversed like they can on a potentiometer so flipping the magnet might be an option for anyone wanting to reverse an axis.

Follow the link at the beginning of this reply to see more photos.

Again, do NOT take your controller apart! You will likely void the waranty. Hopefully these photos will satiate your curiousity.

I have photos of the chips used in the controller. I'll upload these as time permits. I promise I'm not dragging this out on purpose.

Here are a few more photos of the inside of the controller.




There sure seems like there is some serious hardware inside this controller. I'm still unsure how the comminication protocol works. Does the controller talk with the drone and the goggles? Does the controller receive data back from the drone and goggles? I don't think I know enough to know how to test this.
There are additional photos is my latest log on Hackday IO.

Thanks for your work.
Maybe you can tell me a little about the 4 screws, used to force the sticks centered? Sadly I lost one of them (F1).
It must have accidentally unscrewed one a little too much, so that it fell down unnoticed...
They are like M3 on the shaft, but the head seems to be smaller than that...

thanks Imp

You're right about the threads being M3. The pitch is 0.45 (there are also 0.35 M3 fasteners though I don't know if I've ever seen one).
Both of the screws have threaded sections of 11mm length.
The F1 screws are 13.4mm long. These screws are used to reduce the spring tension on the vertical axis of a gimbal.
The F2 screws are 18.5mm long. These are used to adjust the leaf springs pressed against the vertical axis of the gimbal. As these are tightened the vertical axis becomes stiffer.

If the appropriate F1 screw isn't screwed all the way in, you'll get an error message while in manual mode.  Other than this error message, I think the controller can be used reasonably well with all four screws removed. You'll just need to physically hold the throttle position when in manual mode.

There are a couple options for replacing these screws. If you've only lost one, you can use the corresponding screw from the other side of the controller. I don't think these screws are really needed other than for adjusting the throttle when manual mode is used.

You could also use a Dremel type tool to cut a M3 screw's head off and then cut a slot in the end for a flat head screw driver. I've made plenty of DIY grub screws this way. Alternatively you could purchase a M3 grub screw of the appropriate length. If you can't find the correct length, you could purchase a longer grub screw and cut it to size.

Here are a few batches of grub screws I found on Amazon.
Black 20mm M3 grub screws.
Stainless Steel 16mm M3 grub screws. (These might be too short for F2 screws but I bet they'd still work.)

Duane Degn Posted at 3-26 08:28
You're right about the threads being M3. The pitch is 0.45 (there are also 0.35 M3 fasteners though I don't know if I've ever seen one).
Both of the screws have threaded sections of 11mm length.
The F1 screws are 13.4mm long. These screws are used to reduce the spring tension on the vertical axis of a gimbal.

Oh my god, I just hope, DJI promotes you to best support staff member ever and pays you a ton of money.
(especially the money of the other support members, because you are the only one I have got a real and accurate answer from. I wrote mail, posted and even tried to call the support without success...)

I tried unscrewing the two screws from the pitch/roll stick, and the result was ....nothing. they just don't have any function on that side.
So , now I got my throttle tight again, and even got an additional screw.
Btw, I love your project on hackaday.io. Keep it up, we definitiflly need exact information about what hardware DJI is selling us, without information.
Thanks again.
greetings Imp

Thanks for sharing all this, awesome stuff. Any tips for completely removing the tension from the throttle? I've tightened F1 as much as possible and it still slowly re-centers.

djiuser_MyBlXwwjaemB Posted at 3-26 15:59
Thanks for sharing all this, awesome stuff. Any tips for completely removing the tension from the throttle? I've tightened F1 as much as possible and it still slowly re-centers.

Can you tighten F2 without causing an issue? Tightening F2 should provide more friction to the throttle and make it harder to move.

I was about to suggest removing the spring on the throttle but I don't think that would remove all springiness in the throttle. I'm not sure where the last bit of springiness comes from. It's possible the wires to Hall sensor provide a small about of return force when the throttle is moved away from the center. The wires to the rudder's Hall sensor is glued to the gimbal. This acts as a weak spring to return the throttle to center. I'm not sure this is the case but that's my present guess.

I tried removing the throttle return spring on my controller but I don't see a way of completely removing the spring without disassembling the gimbal. If the spring is unhooked from on just the end you can access, then it causes movement issues with the throttle.

My only suggestion for now is to tighten F2.

Here's a photo with some annotations to show the parts discussed above.

“ It might be possible to reverse an axis by flipping this magnet. I don't think the power lines of a Hall sensor can be reversed like they can on a potentiometer so flipping the magnet might be an option for anyone wanting to reverse an axis.”

Not a good suggestion, depending on the application, flipping the magnet can render the sensor useless (poles matter).

Interesting thread and your time to post pictures and information is appreciated but be careful with the information.

Some great info in this thread.
Thanks heaps.

Duane Degn Posted at 3-26 18:11
Can you tighten F2 without causing an issue? Tightening F2 should provide more friction to the throttle and make it harder to move.

I was about to suggest removing the spring on the throttle but I don't think that would remove all springiness in the throttle. I'm not sure where the last bit of springiness comes from. It's possible the wires to Hall sensor provide a small about of return force when the throttle is moved away from the center. The wires to the rudder's Hall sensor is glued to the gimbal. This acts as a weak spring to return the throttle to center. I'm not sure this is the case but that's my present guess.
[Image]

Thank you for the advice, adjusting F2 did what I needed. I feel silly for not trying that sooner!

djiuser_MyBlXwwjaemB Posted at 3-27 05:56
Thank you for the advice, adjusting F2 did what I needed. I feel silly for not trying that sooner!

"adjusting F2 did what I needed"

Thanks for letting me know. I was hoping that would work.

Great thread!

Do you know what type of batteries are in the controller? Lipo, Lion, NiCD, NMH? (And goggle batteries if you happen to know).
EDIT: just saw your website, looks like Lion 18650s in the controller.

Thanks!

Buster1 Posted at 4-7 11:52
Great thread!

Do you know what type of batteries are in the controller? Lipo, Lion, NiCD, NMH? (And goggle batteries if you happen to know).

"And goggle batteries if you happen to know"

I haven't managed to pry the goggle battery's case open. Occasionally DJI welds/glues enclosures to make them really hard to open.
I have DJI's RoboMaster S1 robot. Most of the robot was relatively easy to take apart but the battery enclosure had been glued closed. After getting the battery open, my S1's battery pack never fit in its slot well again.

The enclosure of the goggle batteries appears to be glued shut and I haven't used much force to try to open it yet. I'd be very very surprised if the goggle battery wasn't a pair of 18650 cells. The battery enclosure is just the right size to hold two 18650 cells.

I think the goggle battery has internal battery management circuitry for both charging and boosting the output voltage. Trying to measure the voltage output gives very strange readings when the battery isn't connected to the goggles. It doesn't behave like a normal 2-cell Li-Ion. The way the unloaded voltage jumps around makes me believe it has a boost circuit in line with the output. I haven't made an adapter to monitor the voltage while connected to the goggles.

While the 2-Cell goggle pack works to power the goggles, a non-boosted 2-cell Li-Ion or LiPo pack doesn't supply enough voltage to power the goggles. I haven't tried this myself, but I think DJI correctly specs the goggles as needing a 3-cell to 6-cell LiPo pack.

I have taken apart the Fly More multiple pack charger. I'll try to make time to upload some photos of the inside of the sequential charger in the next day or two.

This thread is nice

Duane Degn Posted at 3-23 19:07
I've added another log to my Hackday IO project. This log shows details about the small buttons used to detect when a gimbal has been set to manual mode.
[view_image]

Hi Duane,

Nice info and nice pics.
Do you have pics of the hall sensors? Im looking for the reference of the sensors, changing the sensors for another ones with the inverted polarity should invert the axis .

It would be so easy to do this throught an firmware update, but it seems like DJI don't want to listen to their customers...

Thank you Duane for all the work, and risking your warranty in the process. We do appreciate it. Nice info.

roviraleix Posted at 2021-12-11 08:37
Hi Duane,

Nice info and nice pics.

I've heard that turning the magnets will reverse an axis.
https://forum.dji.com/forum.php? ... 711&pid=2644484

I'm accustomed to left stick L/R being rudder (yaw), right stick L/R being roll (sideways), left U/D being altitude, right U/D pitch (forward/backwards). Haven't been able to train myself another way. There's no custom mode for that. Do you know what wires need to switch for swapping the left and right sticks for their L/R (horizontal) functions? I feel like it's one of the yellow/purple/pink trio.

WiscoSailor Posted at 3-22 05:33
I'm accustomed to left stick L/R being rudder (yaw), right stick L/R being roll (sideways), left U/D being altitude, right U/D pitch (forward/backwards). Haven't been able to train myself another way. There's no custom mode for that. Do you know what wires need to switch for swapping the left and right sticks for their L/R (horizontal) functions? I feel like it's one of the yellow/purple/pink trio.

What you described sure sounds like mode 2 to me. Isn't that the default setup?
If you need to switch wires, I'll be glad to try to help but what you described sounds like the standard setup.

“ The two antanna each have a second BCP with two traces on it.

“ The two antanna each have a second BCP with two traces on it.“ Dou you have a photo of this unwired board? Size of traces are important for me (in mm) any detailed info?

Hi Duane,

Sorry for the late response.
Do you think there could be a way to increase the spring tension on all axes?

Is there any chances 3rd party hall gimbals will work with Remote 2? Like M9 or Literadio? I've tried old DJI (black) gimbals on Remote 2 - doesn't work.

350D Posted at 3-3 14:20
Is there any chances 3rd party hall gimbals will work with Remote 2? Like M9 or Literadio? I've tried old DJI (black) gimbals on Remote 2 - doesn't work.

Do you have a multimeter? The joysticks might not use the same voltage. The pin order might also have changed.
As I mentioned on Hackaday, I don't have any experience with the Remote 1.

Good luck.

Duane Degn Posted at 3-4 07:58
Do you have a multimeter? The joysticks might not use the same voltage. The pin order might also have changed.
As I mentioned on Hackaday, I don't have any experience with the Remote 1.

Yes I have and I've tried to measure all pins resistance to each other while rotating new gimbals - no any numbers changes. There is no even a short circuit on any of pins.

UPDATE: I've tried TS16S Hall gimbals with swapped PWR and SIGNAL pins and it works on new controller! Also i've tried to swap same pins on old DJI gimbals and it doesn't work. Well, at least i have PoC for hall gimbals now. I want old gimbals to work on new remote more than ome 3rd party ones...

Hello everyone, I went ahead and made a in depth video of how to take apart the remote controller 2 to invert the throttle and pitch controls. The only other video on youtube doesnt show step by step. My video is not a high quality production but I hope it helps some of you! Cant believe DJI has not come out with as software update... now that the Avata 2 is out and RC 3 I hope they come up with an update.. If not this video is for you
https://youtu.be/SxEbhOioJAM?si=tMWMQaChvCbS5-rU

DJI Stephen Posted at 2021-3-21 19:45
Hello there Duane Degn. Good day and thank you for sharing these photos and this information here at DJI Forum. Just a reminder that doing this process might affect the DJI FPV remote controller's performance when in use and can void the warranty of the said remote controller as well. Thank you.

He LITERALLY stated that

curious? are the gimbals the same throughout the different controllers 1, 2, and 3? I've been looking around online, and I haven't seen any info on this. when you look up the gimbals for each controller, they look identical. I'm needing to replace a gimbal on my controller 2 but they are out of stock everywhere.