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

Shorted components near antenna amplifier

punkkone

Active Member
Joined
Sep 11, 2019
Messages
42
Reactions
17
Age
37
Location
Sweden
Hi guys! I know this might be the wrong part of the forum, but I figured I'd have a higher success rate with asking people that are not afraid to take stuff apart :)

So, I bought a crashed Mavic Pro, front was badly smashed, front sensors, gimbal, shorted and burned camera board, torn ribbon cables.. I tried starting it up, and it just gets stuck on solid red light, no wifi or USB connectivity at all. I thought "well, that could be anything.." and placed an order for most of the parts that were obviously beyond repair.
What caught my attention here is that the motor startup motor-"jiggle" only moved the rear rotors, the front ones kept still.. Possibly an ESC board issue? And I'm fine with just replacing the ESC board as the prices are reasonable.. So I took the rest of it apart and found this.
After carefully cleaning the board and removing the shield for the wifi amp etc:


20190911_141615.jpg20190911_140809.jpg

Now I cannot for the life of me find what those IC:s marked "S 20" are? Or the components between the antenna connectors and those IC:s? So, what would be the most likely culprit for blowing the IC:s (or possibly the IC:s are just damaged by the resistors blowing out..)? What are those components? I'm willing to have a go at replacing the components before ordering a new core board as they are not exactly cheap..

Feel free to move the thread if it fits better somewhere else.

Cheers!
 

Mm, yes that was my initial thought, but I've been searching for hours and hours and I can't find anything that resembles these. The inputs seem to be the two corners on the side opposite of the corner marking, and the output the center pin on the side with the marking no?
Finding direct replacements would be very very neat...

Any idéa of what the components between the IC:s and the IPX connectors are? Looks like either some coil or a female connector to me?
 
Hey, what if the "S 20" is just a divider/combiner? That'd make sense.. A lot of more sense as there is a "S 20" north of one of the burned ones but not the other.. Hmm
 
Mm, yes that was my initial thought, but I've been searching for hours and hours and I can't find anything that resembles these. The inputs seem to be the two corners on the side opposite of the corner marking, and the output the center pin on the side with the marking no?
Finding direct replacements would be very very neat...

Any idéa of what the components between the IC:s and the IPX connectors are? Looks like either some coil or a female connector to me?

Antenna tuning caps? Are they really fried? Looks like they are glued down after adjustment. I’d check their values before pulling them.
 
Last edited:
The left one is fried for sure, all carbon gunk from melted plastic on the inside.. The one on the right is seems to be ok, measured continuity and resistance. I really don't get how this damage can even occur.. The drone is not water damaged, no obvious shorts except where the front sensor ribbon cable connectors were smashed, there I could see signs of a pretty bad short.

I mean, was the previous owner flying this in a thunderstorm or did the drone come in contact with power lines? xD (of course not, but it's just so weird)

But ok, if those are just pots (they are very very tiny) I could probably just bypass them and come back to that issue later. But the "S 20" conundrum is the big issue, the closest thing I can find, to what I believe them to be, are these: https://ww2.minicircuits.com/pdfs/SP-2U2+.pdf

Never came across a more elusive IC.. I might just roll the dice on getting a known bad board and transplanting the IC:s from there :/
 
Last edited:
The left one is fried for sure, all carbon gunk from melted plastic on the inside.. The one on the right is seems to be ok, measured continuity and resistance. I really don't get how this damage can even occur.. The drone is not water damaged, no obvious shorts except where the front sensor ribbon cable connectors were smashed, there I could see signs of a pretty bad short.

I mean, was the previous owner flying this in a thunderstorm or did the drone come in contact with power lines? xD (of course not, but it's just so weird)

But ok, if those are just pots (they are very very tiny) I could probably just bypass them and come back to that issue later. But the "S 20" conundrum is the big issue, the closest thing I can find, to what I believe them to be, are these: https://ww2.minicircuits.com/pdfs/SP-2U2+.pdf

Never came across a more elusive IC.. I might just roll the dice on getting a known bad board and transplanting the IC:s from there :/

I think you are right, maybe its a RF splitter/combiner on the final. I didnt think there was enough output power to do it, but I wonder if all that can fry like that if the antennas are damaged or disconnected?

Or if there is no sign of water being inside it, perhaps something conductive entered the drone (like glitter can do it), or perhaps there’s a cold solder joint?
 
I think you are right, maybe its a RF splitter/combiner on the final. I didnt think there was enough output power to do it, but I wonder if all that can fry like that if the antennas are damaged or disconnected?

Or if there is no sign of water being inside it, perhaps something conductive entered the drone (like glitter can do it), or perhaps there’s a cold solder joint?

Possibly, the left antenna housing is broken clean off and the antenna is dangling freely. Never seen actual board damage due to shorted/damage antennas/coaxial cable though but I guess it's possible. The attenuation/resistance for the complete circuit might be way off with a shorted antenna, but I haven't studied the circuit closely enough to determine if that might be the culprit yet.

I'm wondering what would happen if I just desoldered the damaged components and hooked antennas directly to the wifi-board. Sure RF mode wouldn't work, but it might let me test other things when all the other broken components are replaced.
Wouldn't surprise me if it gets pissed because gain levels would be out of spec but who knows..
 
So, bird is finally in the air and flies fine. This is what I did, bypassing the multiplexers (at least that is what I figured they are, might be RF combiners) and effectively bypassing the wifi-board. This will likely remove some attenuation and in effect produce more gain (about 1dB with stock antennas). Shouldn't affect tuning at any noticable level. (I work as a technical architect with home routers, wifi, networking stuff etc. so I have at least a decent insight into the world of hardware engineering with RF component, but that said I'm still a layman at drone stuff)

Anyways if anyone else ever encounters this:

1. Use a microscope and hot air solder station. I didn't, just a standard iron and my phone for magnification.. It sucked..
2. Leave the coax-cables attached to the wifi board or leave the wifi board disconnected or else you will have magnetic interference on compass #1

IF you for some reason wanted to (I can't imagine why) you could hook up two separate antennas directly to the wifi board.

1578658120929.png
 
Last edited:
  • Like
Reactions: zappy54 and AMann
My guess would be they are band pass filters. These would be to help block extraneous signals on other frequencies. Blocking this sort of RF can improve range as the amp front end doesn't get loading up with garbage. It will work without them, or with them bypassed, but range may suffer due to the inability to block other frequencies.

The "hint" that leads me to conclude the above is their location relative to the antenna coax fittings, and that band pass (or rejection) filters typically have just two connections (like a capacitor, inductor, or other simple component, as compared to amps which will have more connections including power, input and output etc.).
 
My guess would be they are band pass filters. These would be to help block extraneous signals on other frequencies. Blocking this sort of RF can improve range as the amp front end doesn't get loading up with garbage. It will work without them, or with them bypassed, but range may suffer due to the inability to block other frequencies.

The "hint" that leads me to conclude the above is their location relative to the antenna coax fittings, and that band pass (or rejection) filters typically have just two connections (like a capacitor, inductor, or other simple component, as compared to amps which will have more connections including power, input and output etc.).

Good thinking, but no I don't think so, the grey components before the burnt chips looks more like filters of some sort (multilayer lowpass or bandpass), the burnt ones are for sure something to combine the RF signals as they alone handle the signals for both the wifi board and the core board RF. The burnt ones also have two inputs, one output and three ground connections which to me sounds more like a diplexer/multiplexer/combiner.
Range seems pretty much where I would expect, about 5000' through dense spruce and pine forest blocking LOS (in CE mode, just set it to FCC but weather hasn't allowed for testning)
 
Last edited:
Also, if you look at the picture above, the third visible "S20" chip makes no sense if it's a filter, but as a multiplexer it does (for combining 2.4GHz and 5.8GHz signals on the core board RF circuitry)

But then again I can be dead wrong! Hell if I knew this stuff for real I'd probably be making money at some drone manufacturer.. :D
 
Hi,
Interesting work so far. Actually I remember someone did mention on the forums here that those are diplexers / multiplexers. My guess is wifi + regular 2.4Ghz Occusync mixer. Since you mention you can get them to switch on etc, maybe you can try to switch to Wifi + Phone. If it does not connect then probably you are right.
 
Hi,
Interesting work so far. Actually I remember someone did mention on the forums here that those are diplexers / multiplexers. My guess is wifi + regular 2.4Ghz Occusync mixer. Since you mention you can get them to switch on etc, maybe you can try to switch to Wifi + Phone. If it does not connect then probably you are right.

Well, the coax cables from the wifi board are disconnected from the core board (So no passthrough to the actual PCB-antennas in the front legs) and kept connected on the wifi side just to mitigate magnetic interference
(yes I know it sounds stupid but leaving them disconnected from the wifi board peaked the compass interference) and the amps on the wifi board are probably dead by now, but I don't care for that functionality so.. :) )

And yes, what you are saying is exactly the same conclusion I came to, they are multiplexing the signal from the wifi board (802.11) and the signal from the core board (occusync)
 
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Members online

Forum statistics

Threads
130,598
Messages
1,554,236
Members
159,603
Latest member
refrigasketscanada