Hardware BSOD cause

[Ttm3]

So I've got a about 5 or 6 switches that have BSOD. 2 of them I can put pressure on the processor to avoid the blue light but with the others that doesn't work. Is there any way to test to see which component is causing the BSOD. I'd rather not reball the soc if I don't have to.

Also what components can cause BSOD? I've heard processor, ram, emmc, and m92t36. This is one area I've always avoided but I think I'm ready to take the plunge

 

[SheriffBuck]

@FXDX is your man here. Appears to have a great track record in nursing BSODs back to life.

I've got a BSOD unit here. It's patched, so I'm waiting to fit an SX Core to see if I can get CFW mounted and perhaps that can tell me a little more about what is going on. I'm still holding out that the NAND is corrupted somehow so I can avoid applying heat.

Perhaps someone could offer some advice on a reflow rather than a complete re-ball.... ???

 

[Canna]

The blue screen can be caused by software also. Bad update/Corrupt fw etc..
Have you tried factory rest on them?

Holding both volume buttons and pressing power once whilst holding volume buttons still should show the factory reset menu.

 

[Ttm3]

I haven't tried that. I'll do it when I get home from work. Does it matter if it gets blue screen as soon as I touch the power button

 

[FXDX]

The blue screen can be caused by software also. Bad update/Corrupt fw etc..
Have you tried factory rest on them?

Holding both volume buttons and pressing power once whilst holding volume buttons still should show the factory reset menu.

BSOD - Nand corrupt or bad eMMC, bad soldering memory ram, bad soldering SoC Nvidia Tegra, logic board bent.

 

[Canna]

BSOD - Nand corrupt or bad eMMC, bad soldering memory ram, bad soldering SoC Nvidia Tegra, logic board bent.

Nand corrupt or bad eMMC (Like I said bad software bad FW,)
what's your point?

--------------------- MERGED ---------------------------

I haven't tried that. I'll do it when I get home from work. Does it matter if it gets blue screen as soon as I touch the power button

Possible depends on the situation.

Can it RCM ? If not Hardware issue.
Does hekate payload work,? have you tried to send payloads at all ?

Reminder, Hekate does not need emmc or sd card to work

 

[FXDX]

Nand corrupt or bad eMMC (Like I said bad software bad FW,)
what's your point?

My point is that if it's bad software, bad FW as you said, you don't solve the problem just with a simple hard reset. The problem requires a much more complicated solution and here we are talking about a nand restoration after a software or hardware reading that is not simple at all.......
90% of BSOD cases are hardware problems !!

 

[SheriffBuck]

@FXDX is your man here. Appears to have a great track record in nursing BSODs back to life.

I've got a BSOD unit here. It's patched, so I'm waiting to fit an SX Core to see if I can get CFW mounted and perhaps that can tell me a little more about what is going on. I'm still holding out that the NAND is corrupted somehow so I can avoid applying heat.

Perhaps someone could offer some advice on a reflow rather than a complete re-ball.... ???

So my SX Core has arrived and I've fitted it to my patched V2 Nintendo Switch that is suffering from a BSOD. The fit appears good and I'm getting a green light from the board, however I'm getting a repeated boot.dat error on the screen, regardless of SD card, or reader daughter board used. In using the official boot.dat from Xecutor.

My plan once I get this working is to try and do some diagnosis on whether the NAND is causing the BSOD, or if its the CPU/RAM....

Any suggestions on what the boot.dat error could be?

Sent from my SM-G975F using Tapatalk

 

[SheriffBuck]

OK. Big breakthrough this morning that I thought I would share.

I have overcome the SX boot.dat issue that was plaguing my unit. Combination of a firmware update and using the latest beta (Xecutor are not at all clear that for SX core you need to use the 3.0.x line of firmwares!). My patched BSOD Switch is now booting into SX OS fully.

From this I launched the OFW, expecting a BSOD, but to my surprise, the unit booted up and straight into the OS!

I'm doing a NAND backup right now before I do anything more (I wasn't able to do that because of the BSOD) and will start to do some more diagnosis to determine the cause of the BSOD without SX Core installed. My current theory is that I've now got a fuse mismatch which is causing the BSOD, and not a physical CPU/RAM/FW issue. Will test that theory over the next day or so and report back.

 

[SheriffBuck]

Update from today.

Booted into OFW and upgraded to 10.4. Rebooted and was able to get the device to boot without the SX Core attached.

Result. BSOD cured. Must have been a fuse issue.

Hope these findings help someone.

 

[lpoolm]

Update from today.

Booted into OFW and upgraded to 10.4. Rebooted and was able to get the device to boot without the SX Core attached.

Result. BSOD cured. Must have been a fuse issue.

Hope these findings help someone.

Can you give me some pointers on what you did for the fuse mismatch please? I have two boards I would love to save
Thanks

 

[SheriffBuck]

As its a patched switch, I used the SX Core to bypass the fuse check that normally happens on boot. Once I got the stock firmware running, I let it apply its updates which would have triggered a revised number of fuses to be burned. One further boot using SX Core to ensure that the OS had correctly updated, and then allowed reboot without disrupting fuse check and it worked. Hope that helps explain the process.

Sent from my SM-G975F using Tapatalk

 

[Ttm3]

Just an update. I've had terrible luck reballing the tegra. However I've "cured" 3 bsod by reflowing the ram chips.

 

[SheriffBuck]

That's great news.

Can you offer a step by step on the process for anyone that may want to have a go at doing this operation themselves please? If certainly appreciate how to reflow the ram....

S.

Sent from my SM-G975F using Tapatalk