Switch Lite modchip FAIL. No Boot, Instinct V6S flashes white.

[Sir_Charles]

This is my second modchip install, first time on a Lite. Seemed to be a lot easier than OLED, but then it wouldn't even turn on. I hope I didn't totally brick my younger brother's Switch Lite! Removed the chip and still no boot (FPCs still soldered on). With the Instinct V6S connected it flashes red>white>red>white>purple>white (long), and after about 10sec it goes off but pressing power button again doesn't work, I need to hold power(hard shutdown) then it'll do the same led error code.

I checked all values and they seemed ok, except B was low compared to the OLED.. here's all the diode readings I got:
(Red Probe+ | Red -)
SP1 ≈ 0.024 | 0.031 (20.4ohm across cap)
SP2 ≈ 0.024 | 0.031 (20.4ohm)
3.3V ≈ 0.880 | 0.457
D ≈ 0.715 | 0.450
A ≈ 0.710 | 0.500
C ≈ 0.714 | 0.450
B ≈ 0.792 | 0.427

Another strange issue is that D reads 1.5 something when measured from the modchip instead of directly from the solder point (found out about this after). Also the FPC came with the PicoFly 3-in-1 board but I checked and the pin out should match the Instinct V6S, need to double-check.

I don't have a USB-C power meter yet, just ordered one, but what else should I check for now? Is there a diagram of expected resistances/common places to check for shorts?




(SP2 looks bridged to the right cap but it's not.)



Reddit post with video

 

[slimeinacloak]

I took some measurements on a lite (no modchip or flexs installed, battery, daughterboard, game card slot not plugged in but everything else plugged in)

Black/- on GND | Red/+ on GND
A - 0.782 | 0.516
B - 2.666 | 0.439
C - 0.726 | 0.460
D - 0.725 | 0.460
SP1 - 0.022 | 0.029
SP2 - 0.025 | 0.028
3.3v - 0.888 | 0.460

Everything other than the B look roughly ok.
I'd recommend taking the board out fully and doing some diode testing again, ones to compare to can be found here (open with FlexBV Free):
https://gbatemp.net/threads/xzzs-switch-boardviews-cracked.663854/post-10581574

 

[Sir_Charles]

Removed all the flex cables. RST/B point still reads around 0.795 | 0.428, and still no signs of life/no boot. I even tried plugging into my PC to see if it would go into RCM but got nothing (my first time using TegraRCM so maybe I just didn't use it right?).

I've got a dead short on the bigger cap below the card reader/headphone connector.

And I'm not sure if these readings are ok (continuity mode) for this area.


This is all I could find so far on the top side.. I'll pull the whole board to check everything around the M92T36 and other main chips when I have more time tonight. USB-C amp meter coming in tomorrow hopefully, but right now I have no clue how I even borked this Switch Lite! Is there anything that would cause a low RST/B reading like this?
I really hope this is fixable, it would suck to need a new mobo and lose all my bros save data.

 

[Sir_Charles]

Got the amp meter! With the official charger it shows 15V and only 0.31A max.


Still no charging icon on screen or anything else. Same current readings with or without battery connected.
The shorted cap in my previous post is apparently supposed to read 0.00 according to the PCB viewer.

Checked all the caps around the M92T36 and BQ chips. The scary lone "CPU" cap on the bottom left in the pic beeped on me but only in diode mode, not continuity.


Nothing else was shorted, but the diode readings I got were way different compared to the PCB viewer XZZ dump, maybe I need to do the readings with red prove to ground instead?

I still have no idea how installing the modchip could've caused this, I only removed the gamecard board and the CPU lid before doing all the soldering work, also taped over the CPU flex to prevent any contact to the lid. But I did notice once when I tried to pop off the lid while the battery was still connected it ended up powering on the modchip as if I pressed the power button, that was odd but it happened after I found out the Switch wouldn't boot so the damage was already done.
I guess I'm gonna give a shot at replacing the M92T36, hopefully that fixes it, if not then I'll try the BQ next.

 

[slimeinacloak]

Yes normally red probe is on ground for measuring in diode mode, 15v is a good sign the M92 is ok as it is responsible for negotiating to the voltage. Have you checked if it's a display/backlight issue? Can be done by not having the chip connected and trying to boot the console and see if you can hear anything when changing volume/touching on the screen

 

[Sir_Charles]

Yes normally red probe is on ground for measuring in diode mode, 15v is a good sign the M92 is ok as it is responsible for negotiating to the voltage. Have you checked if it's a display/backlight issue? Can be done by not having the chip connected and trying to boot the console and see if you can hear anything when changing volume/touching on the screen

I didn't even touch the backlight/display connectors until after the Switch was already dead, so I don't think it's either of those, also no sound at all. It just stays at 15V 0.031A, no attempt to auto boot when plugged into charge. And battery was fully charged before I attempted the mod chip install, still at 4.183V.

Here's the new readings with red probe to ground:


And for around the MAX chip? Can't really read what it says on it.


Also found a lot of info about the Switch's boot stages from this site: https://www.retrosix.wiki/first-stage-boot-short-checks
But unfortunately I don't have a bench power supply to do any voltage injection tests, and it's talking about the original Switch not the Lite so things are totally rearranged. At least now I know the M92 isn't even needed for boot so I'll try to remove it and see what happens.

Another source of helpful info:

 

[Sir_Charles]

I compared all the new diode readings for M92 & BQ to the XZZ pinout and they seem close enough, plus 0.30-0.90V for each point comparatively.

So I'm thinking it's that oddly low RST/B point, and looking at the XZZ diagram it goes directly to the MAX77620A and should read 0.409V (mine read 0.795 | 0.428, so close enough with red to gnd but still off for black to gnd). I'll check it again when I have more time but just in case anyone else comes across this here's info about the MAX77620 (taken from the previous vid):

I'm not sure if it's okay to check voltages with just the fully charged battery plugged in, or battery and a standard 5V/2A power brick? Using the official 15V one seems a bit much for probing around...
Also not sure if it's possible to do the 10k ohm resistor battery bypass on the Lite for voltage injection, the battery/connector is different too.

Also all points connected to RST/B:

 

[slimeinacloak]

To do the battery bypass on the lite you can use a 10k resistor across these 2 test pads

 

[Sir_Charles]

To do the battery bypass on the lite you can use a 10k resistor across these 2 test pads

I'm so stupid..
Found the problem: I accidentally flipped one of the resistors to the bottom right of the B point on its side, bridging it to the one next to it. After looking at the PCB viewer I finally noticed there should be 2 resistors side by side. Tried to heat it and push it back down but it ended up getting obliterated.
At least now the B point reads correctly at 2.528 | 0.427 with that bridged resistor removed.


Comparing to a OLED board scan with component values it appears to be 100kΩ and 10kΩ for the left and right respectively. Looks like I'll need a digital scope to do this, those 0201 resistors are unbelievably tiny.

 

[slimeinacloak]

Having a microscope from the beginning and being able to see the issue easier wouldve saved you a lot of trouble lol, glad you were able to find the issue and I hope everything is normal after getting the new resistors

 

[iiTzSamurai]

Looked like you had bridged sp2 with the cap beside it