Nintendo Logo and then shutdown

I got inspired but many repair videos especially tronicsfix :slight_smile: and recently decided to try it myself. I was quite successful in repairing a few switches I got from ebay but I’m now clueless with one of them:

At the start it did not display anything on the screen or when it was docked. I discovered a bend pin in the lcd connector, which led me to replace the connector and now the display works and shows the nintendo symbol for 2 seconds and goes dark afterwards. It does not seam to shutdown completely since I can hear a hissing sound from (I guess the speakers) until it goes silent when I press the power-button for several seconds.

I checked for shorts and took diode mode readings all over the board (e.g. M92, CPU, PI3) and compared them to a known good switch and the values in other topics, but I wasn’t able to find anything suspicious.

The switch charges normally and also jumps into fast charge.
I checked the voltages on various testpoints that I found in other topics but everything matched up as far as I can tell.

With the help of a repurposed Joy-Con-Slider I managed to put the Switch in RCM mode, it was detected by my computer but I got no response when I tried to inject a payload, neither the 0x0000 nor the 0x7000 answer.

I read that emmc could be the wrong one or defective but popping it off and trying to inject a payload yielded the same results as above.

I would be fantastic if somebody could give me a hint what I could try next, help would be very much appreciated.

Severence suggested to me before about bisdumpkey and see what it say about the emmc

Thanks for the suggestion!

I tried to send the bisdumpkey.bin but it behaves the same and displays nothing.
my terminal says:

Found a Tegra with  Device  ID: ...
Setting ourselves up to smash the stack...
Uploading payload...
Smashing the stack...

and nothing afterwards

The fact you aren’t getting any of the codes in tegraRCMGUI after injecting the payload is somewhat concerning.

It’s very interesting though that it is stepping into fast charge, would be curious to know the voltage readings at the inductor outputs of the CPU/GPU buck regs, you should have voltage at the CPU buck (output) at least, if not then that Max IC would be sspect.

Before that though, I take it you have injected a payload on another Switch with success? Just checking that this isn’t some driver related issue

Will say, I had one case such as this matching your symtoms and it turned out to be the SoC at fault, luckily it was a joint issue and a reball resolved.

Your question is a good one. I havent had success until today and started trying around with different cables and different versions of the fusee-injector-thing. Just now I successfully injected the TegraExplorer.bin on another unpatched switch today, so my setup seams to be working, but I noticed that neither the interface, the web version nor the python command line tool gave me any response code like 0x0000 or 0x7000, successful or not, is there some kind of a verbose option? Another of my working patched switch behaved in the same manner as the “nintendo-logo-shutdown”-one and gave no response. The serial number suggests that it is an unpatched one but it’s possible that the motherboard was swapped. So I can’t really discern if the lack of a response is a usual-patched-switch or an unusual-unpatched-switch behavior.

Where can I find inductor outputs of the CPU/GPU buck regulator on the board?

I take it your using a Linux OS if your using the Python/Web ver?

Sorry I’m not familiar with these versions as such it might be completely normal for them not to spit out a code, I would find a Windows PC and use the GUI version where it is known to provide this code after sending the payload.