Crash after first boot logo

Sorry it won’t let me edit on mobile but it’s still a no-go after swapoing just about everything but RAM. The seemingly mysterious GC ASIC must hold system specific software/encryption as changing it made it so I don’t even get the first logo, just screen on for 2 seconds and then off.

So I’m gonna throw the old one back on and then swap RAM. If that doesn’t do it I’m down to hunting random regulators/components…which may just prompt me to abandon the board and transplant the SoC/eMMC to another just to be 110% sure all is good there.

If it’s not the RAM and (educatedly assumed/tested) not the SoC/eMMC, it means that some tiny, random component can not short entirely but give a 1 logo crash. Update soon.

Wow thats insane. What about the other two Max ICs? (on is the small fuel gauge, not sure if you already replaced that) and another small Max IC is South of the CPU on the B side.

All I think off if not the RAM is the CPU.
I mean even if it’s not Shorted, there Could be some other Problem

1 Like

Yeah, fuel guage was the first replacement as it was the only chip to show anything looking loke damage. The board originally did nothing with draws that pointed to the PMICs.
Fuel guage was a little chipped (you know how those MAX ICs are basically glass) so I swapped it and that got it to boot to the first logo and then crash. So, an improvement. I then swapped M92, no change. BQ, no change. Battery and LCD is 100% good so I made this post as I’ve never had a “first logo & crash” board fixed with what I had already done so - almost out of boredom and wanting to find info to help the community - I decided to swap everythinf I could.

So far I’ve done all of the above, all MAX ICs, the GC ASIC, the IC above the sound chip, PI3USB, the backlight IC and the IC near the LCD connector. I also removed all non-essential parts that Nintendo did themselves on other boards.

Still first logo and crash. I’m doing RAM next, likely tonight or tomorrow. My issue being that I have tons of Samsung RAM but this board has SKHynix. I don’t think it should be an issue if I change both but I’m not sure if the system uses “RAM Learning” at first boot to lock itself into a RAM type as Hynix has a different resistance than Samsung (at least in all RAM I’ve dealt with in GPUs and other devices).

So we will see! I’ll report when done. If that fails, I’ll probably double check all readings on the random voltage refulators/mosfets and just start swapping those, too. It’s about 18 total to my count. If that fails, it’s just going to be a case of transplanting the SoC/eMMC as the issue would have to be in the PCB itself. Rare but I have seen it a few times

Update soon.

1 Like

Yeah after All the work or even before there Could be some damaged trace insife the board.

Have you already checked for the present voltages if you turn it on? Maybe you See missing voltage and can relate which rail is faulty

1 Like

It does matter I’m afraid, fused to the manufacturer, so it it’s Hynix, you have to stick with Hynix etc :frowning:

That all being said, It’s unlikely to be Ram related in the first place I’m afraid.

1 Like

I am remembering someone with no boot issue had Bad caps left of the CPU

1 Like

Well dang. I suspected that may be the case but wasn’t sure if Nintendo set that or not. I knew they couldn’t be mixed and matched but wasn’t sure if they are set at manufacture to be one or the other and only that.
So I’ll scratch that plan until I get some good Hynix modules (I’m pretty sure all I have is Samsung atm).

As a test/discovery board in my eyes, I think I will wait until then despite RAM not being high on my list of suspects as well. I agree that it’s highly unlikely but why not try at this point and in this scenario? Kinda just wanna swap to rule out another possibility/variable.

I appreciate the reply as you’re clearly very knowledgable about Switch! Appreciate it, along with all other replies. I’ll update if I eventually swap RAM or find anything else interesting.

Yeah, all appear to be present. That’s the weird thing about it. Everything tests fine in that regard.

As for those caps, I’ll check that out as well. I appreciate the replies, too! Truly. Anyone that offers any help is appreciated and especially in a scenario like this where my motive is more about helping everyone more than myself.
I’ve always heard that these “one logo crash” scenarios described as a case of being about 3 or 4 things before getting a big “I dunno”, hehe.

More than trying to get this board up and going, I wanted to hopefully add another bit of potential things to look for when others in the future have this issue.
So any and all advice and thoughts are appreciated. I’ll report back if/when I find anything as I need to wait until I have some Hynix RAM to do that swap. As Severence said, it’s unlikely the RAM but I think we both assumed that as is. His confirmation is definitely helpful so at this point I’m in a “let’s just do it to do it” mindset :slight_smile:
I’m good at reball and install so worst case is nothing changes. One less thing to wonder about.

Thanks to you both again and I’ll report back as soon as I have something to add!

1 Like

I guess it could still be a software issue? Is it a patched board?

2 Likes

@Dwhdwh did you tried booting into hekate with TegraRCM? Maybe you can reflash the firmware

1 Like

@insomniac

It’s an unpatched board. It could be software as the only thing I can go on is read errors. It shows as ok when mounted on anorher V1 Switch but that only shows so much. I even tossed it on an actual eMMC breakout reader and all sectors show as good/accessible but that still doesn’t rule out a bad write. It’s something I view as definitely possible but not likely based on what I’m able to see.

@zyrex

Yeah, I tried and it won’t let me due to the system not staying on. This is what lead me to swap the MAX ICs to begin with as I thought it could just be a power issue. It gets to RCM and I’m jjst fast enough to send the payload before it turns off due to that slight delay in boot>RCM.

Sounds like a Problem with the BIOS like on PCs.
Not sure if the switch has a dedicated BIOS Chip and even if, you arent probably able to reflash or change it.

1 Like

I think that is all handled by eMMC and SoC on immediate power press as opposed to a traditional BIOS chip (could be wrong, not an expert wrt Switch).

You should be able to get into Hetake (5.0.0) without an eMMC connected at all, so there is clearly something wrong other than software

2 Likes

My view as well. I can’t rule out what I can’t rule out (haha) but I’m pretty much in the camp of it being purely a hardware issue.

Hey there!
I have recently encountered the same “Crash after the Nintendo logo” issue with my Switch, yet I am unable to find a solution. Most of your description fits for my case as well.
I believe, you havent found a fix for it yet? :sweat_smile:

1 Like

Have you changed the two caps left of the CPU already?

1 Like

@R8197 Maybe you can try this, not much to do:

1 Like

Hi. No, I haven’t found the solition yet but I am waiting on some new Hynix RAM modules to rule that out.
I’ll be sure to post as soon as I find the solution (if I do).

For now, the best solution is to make sure the battery and LCD are good. If so, them replace the M92 and BQ chips. That seems to fix some of these consoles but clearly not all of them.

Also, yes, I swapped the two caps on my unit. You may want to try that on yours as well. Finally, the last tip I found was from a Chinese repair site where changing the MAX77621 chip under the eMMC board fixed a first logo crash systems. For me that didn’t help but it can help some.

Best of luck and I’ll report back if I find a solution.

1 Like

Sorry for the late reply!

@zyrex I have inspected those caps and even though the right one seems damaged, both of them give the right values when I measure them with my multimeter. Do you maybe know the exact name of those capacitors, so I may order a replacement online, just to be sure?

@Dwhdwh I wish you best of luck! As much as I understood, the blackscreen after the Nintendo logo may be caused by many things, hopefully swapping the RAM will fix it for you!

I found many shorted caps around the MAX77621 chip you mentioned, thank you for the hint! Nevertheless the caps around MAX77620a on the backside of the board are shorted as well. I am not quite sure if it is the right name of the IC or not, since I am not at home right now.
The one IC surrounded by the inductors on the image, I would recommend you to check the caps around it, just in case.

image

Also, I dont want to take up the spotlight in your post, yet I am also not quite sure if I should create a new one for my question, but do you know by any chance, where to find schematics for the Nintendo Switch? I bet it would make things surely easier :sweat_smile:.

1 Like