Nintendo Switch not turning on:draw 0.45A ,found 2 caps shorted around M92T36(PIN 18)

[mabot17]

@Severence

Or you can bridge the corresponding pins of the flex connetor directly.

Or there is a testpad somewhere, though i forget where off the top of my head.

that’s cool i will check on this

But prior i think i have spotted a cold joint on solder ball under max77261,the one that i have changed
time todo a reflow with amtech flux to be sure

[Severence]

@mabot17

But prior i think i have spotted a cold joint on solder ball under max77261,the one that i have changed
time todo a reflow with amtech flux to be sure

Good spot those ICs are a real pain to peak under because of the surroundings

Though this is almost certainly not going to affect your previous measurments so don’t think that’s the primary issue here (because the ball/pad is open if anything and not shorted to anything else)

[mabot17]

@Severence

hough this is almost certainly not going to affect your previous measurments so don’t think that’s the primary issue here (because the ball/pad is open if anything and not shorted to anything else)

After reflow,nothing changed so far,solder seems good now

@Severence

Or there is a testpad somewhere, though i forget where off the top of my head.

switchbrew orgTestpads
i found this link,so power button tespads are in Cluster E11 and J4.
What should i do after battery connection?

[Severence]

@mabot17

i found this link,so power button tespads are in Cluster E11 and J4.
What should i do after battery connection?

So see how it says “active low” it means pull low to activate essentially.

You’d just bridge either of those points (take your pick) to ground to “press the power button” which is effectively what your doing when pressing the power button, pair of tweezers will work.

Will say though, I’ve not verified a lot of the TP listed on this site, so do it at your own risk, or buzz these points to the power/vol flex connector and see if they are the same as the pin here to verify if your not sure

[mabot17]

@Severence

Can you plug in battery and prompt the console to boot and measure the voltage at these two locations and let me know the readings?

I think i’ve made it right,i have 3,3V on big cap and 1,8V on the small one,are those those M92 pin18 VDDIO and 3v3 pin 6 rails?

[Severence]

Well done

@mabot17

are those those M92 pin18 VDDIO and 3v3 pin 6 rails?

So pin 18 of the M92, is connected to 1V8PDR,

Pin 6 of the M92 is connected to 3V3PDR (the big cap I highlighted near EN IC)

The left side of the resistor I asked you to measure is a 1.8V enable signal (which is in sourcing it’s voltage by way of 1V8PDR but is not a direct connection to it) from the SoC to tell the EN IC to enable the 3V3PDR output.

[mabot17]

@Severence

The left side of the resistor I asked you to measure is a 1.8V enable signal (which is in sourcing it’s voltage by way of 1V8PDR but is not a direct connection to it) from the SoC to tell the EN IC to enable the 3V3PDR output.

So if i understood,the Soc is alive right?

But now what can i do?Test again M92 voltages with this method?

[Severence]

@mabot17

So if i understood,the Soc is alive right?

It’s got a heartbeat it’s at least partially ok, but no guarantees just yet

@mabot17

But now what can i do?Test again M92 voltages with this method?

Right, you should see after prompting the console to boot 3V3PDR at pin 6 now, which means your EMMC and other ICs on this rail will now have power unlike before.

You might wanna try now with the bl/screen connected

Why your USB is/was not prompting the console to boot earlier though is another issue.

If it’s not a USB connector issue then think it’s going to come back to one of the three ICs i mentioned earlier. I hope you’ve got a jig on order it’s so much easier

[mabot17]

@Severence

Right, you should see after prompting the console to boot 3V3PDR at pin 6 now, which means your EMMC and other ICs on this rail will now have power unlike before.

YES i have both 3v3 and 1v8 up and running!!!
i have to tell you,now i’m excited

[mabot17]

@Severence

You might wanna try now with the bl/screen connected

tried with everything hooked up and nothing coming on the screen,awsome
i have 3v3 and 1v8 on M92
still 0 amps from usb connector

[mabot17]

@Severence

If it’s not a USB connector issue then think it’s going to come back to one of the three ICs i mentioned earlier. I hope you’ve got a jig on order it’s so much easier

RCM jig is coming tomorrow,
even if I don’t know how to use it yet, but if I understand it should allow me to understand what state the console is in via the usb cable connected to the pc, and do some checks, I’ll give it a chance at this point.
If there was a sure way to know the Soc is fried, I wouldn’t really be wasting my/your time, even though I’m learning a lot from this experience.

[Severence]

@mabot17

allow me to understand what state the console is in via the usb cable connected to the pc

The USB is just for the initial payload upload after that it’s not needed, the Hekate payload will provide some insight into a few components on the board.

Hopefully you still have USB comms though based on your previous tests. It might be it doesn’t.

@mabot17

If there was a sure way to know the Soc is fried

you’ve been performing all the preliminary tests to validate the SoC without even realising it
Those USB pins you measured earlier can sometimes be an indication but sometimes a bit of a false trap, the rails produced by all the Max ICs etc which are all fine. there’s a few more points which you can test in circuit but tbh it’s very dependent on board rev and the results are not 100% conclusive, there is no guaranteed way to know for sure without removing the SoC and measuring the corresponding rails out of circuit.

[mabot17]

@Severence

you’ve been performing all the preliminary tests to validate the SoC without even realising it

lol sorry i’m really ashamed of myself

I have tried a few things in the meantime that the jig arrives. I have installed tegraRCMgui on the pc and the APX drivers.
I did 2 tests, one I connected the switch to the pc via usb cable without battery and without emmc and nothing happens, then I connected the good battery and the memory, and still nothing, I don’t know what to think, it should see it if it were in autoRCM quite right?
How should I do with the jig?

[Severence]

@mabot17

I did 2 tests, one I connected the switch to the pc via usb cable without battery and without emmc and nothing happens, then I connected the good battery and the memory, and still nothing, I don’t know what to

You may have to prompt the console to boot manually as we know the USB isn’t going to do it. with battery connected and EMMC disconnected.

It may be the USB is faulty (and the dodgy readings your got earlier) on D+/D- will prevent USB comms, or this could be caused by the SoC or something elsewhere.

If that’s the case though I’d expect “USB not recognized” message on the PC.

[mabot17]

@Severence

You may have to prompt the console to boot manually as we know the USB isn’t going to do it. with battery connected and EMMC disconnected.

ok did that but nothing happens

[Severence]

@mabot17

ok did that but nothing happens

What’s your resistance across both pins/pads? black probe on one and red probe on the other

And also what’s your resistance to ground on these pins/pads with red on ground?

[mabot17]

@Severence

What’s your resistance across both pins/pads? black probe on one and red probe on the other

108Kohm

@Severence

And also what’s your resistance to ground on these pins/pads with red on ground?

form left to right 89Kohm and 17Kohm(red probe on gnd)
91Kohm and 35Kohm(black probe on GND)

I found something interesting following the signal path,
an EMI filter i think is blown,because on one side it gives me OL,what do you think?

[Severence]

@mabot17

108Kohm

I see, This is quite low, again, would normally be high megs.

@mabot17

form left to right 89Kohm and 17Kohm(red probe on gnd)
91Kohm and 35Kohm(black probe on GND)

But the issue is your diode mode readings from earlier and resistance readings aren’t agreeing with each other which is making me suspect a meter issue here in this case which maybe skewing readings somewhat.

@mabot17

an EMI filter i think is blown,because on one side it gives me OL,what do you think?

Yeah that’s definately an issue,

I would replace the filter and just as a matter of course take those resistance measurments again as it will surely change them, don’t plug it in just yet, though if they remain in the high Ks both polarities, no real damage should occur with USB connected

It’s also entirely possible the diode has failed also

[mabot17]

@Severence

I would replace the filter and just as a matter of course take those resistance measurments again

i don’t have a donor boards atm so i cannot replace now the filter.
Can i short circuit at least the blown side just to see what happens or is it dangerous for the circuit?

[Severence]

@mabot17

i don’t have a donor boards atm so i cannot replace now the filter.
Can i short circuit at least the blown side just to see what happens or is it dangerous for the circuit?

So I would jump the filter on the side which is open,

Take those readings again with your meter no power, if the readings remain strange, remove the diode just below it and repeat taking the measurments with your meter again,

If everything measures fine (high megs) in all polarities then it should be fine for testing purposes

I’d also check the diode on the other side of the board

You could also rule out USB removing the filter and taking diode/resistance readings at the two points again should be OL