I have a switch that won’t boot into anything (OFW, CFW)
OFW will just remain black when either booting normally or selecting OFW in hekate
CFW will get to the atmosphere logo and freeze at said logo
It is unpatched and I am able to get into hekate and will include an image of the fuse info
The switch had no p13usb originally and required a port replacement (a pin was pushed back into the port grounding it)
No shorts were found anywhere on initial inspection before any IC replacements
What I tried:
Put a new p13usb (was missing)
Replaced M92t36 in case of damage from the previous usb port
Replaced Bq24193
Replaced the fuel guage IC
Two different firmware nand rebuild methods (had no issues dumping anything)
Additional information:
Diode readings always taken with red probe on ground
I should mention I get all voltage rails other than 0v8 GPU, don’t think the switch boots far enough to enable that line, anyone know a way to test if the 0v8 GPU rail enables?
So I had some time to mess with this switch some more, noticed I was missing one of the small voltage rails coming from the max77620A, I went ahead and replaced the IC and ended up with a 0.0a draw
Reflowed the max77620A after finding an OL on one of the caps near the IC, after the reflow I had the correct reading on said cap
After plugging in the charger, the amp draw shot straight up to 0.48 with nothing else connected and produced a warm SoC, Figured I nudged it too hard so I removed the max77620A again
Reballed by hand the original max77620A in case the one I got from ali was faulty (it did have marks on it as if it was a used ic when I received it)
With the newly reballed max77620A(original) I receive all expected voltage rails but I am back to the 15v 0.0a draw with no response when attempting to power on, I suspect the ‘replacement’ max77620A was fine when I first put it on as I had the same 15v 0.0a
Anyone have any ideas on what could cause a 0.0a draw?
Back to square one with this switch after replacing the original max77620A, accepts payloads, can dump nand etc.
Black screen when starting the switch normally,
Black screen when selecting original firmware in hekate,
Freezes on full logo atmosphere screen when selecting custom firmware in hekate
So either a ram or cpu issue, hopefully just a connection issue
Should I just reball the ram? lastly reball the soc if ram doesn’t solve the issue?
I guess no one has seen an issue like this one?
What happens if you try to boot L4t Ubuntu? This will validate your CPU/GPU regulators if it boots and also wifi after. If L4t doesn’t boot then you could potentially still have issues at the CPU/GPU regs but yeah your probably right, SoC/Ram issues most likely
Didn’t think of trying l4t, will try that as soon as my new micro sd card arrives, I gave my last one away.
The cpu regulator does output 0.95v so that one is probably fine?
I did change the gpu regulator in the beginning, I’ll make it a habit of using l4t to know if they are good in the future
Forgot to mention the amp draw is 0.08a with only the nand connected, gets stuck somewhere in the boot cycle
yeah it makes it highly likely it’s fine. Very rare it’s output would come up if the IC is bad.
this is unusual given that the (secondary) CPU rail is coming up… afair high current charge starts just prior to this (or alongside it at the least). I suppose it’s possible you have a secondary issue
I reckon wait and see what the charge behaviour is if it can boot into L4T and we’ll go from there
Yes the max 77621aewi below the soc is the one that comes up.
The gpu one near the nand I will test once I get l4t going in the next couple of days
Happen to know anything about the 1v8 alt rail? on this board the resistance to ground is ~850k where two other boards are within the ~250k range, black probe on ground
I was originally thinking you were referencing your 1V8PDR but realised after you noted “alt” 1V8 rail. Can you highlight this rail your reffering to on an image? (there is like 4/5 1V8 rails on switch unrelated to each other)
Ah I have a feeling me and @SheriffBuck have talked about this before and I think he identified this one not me (I can’t remember for the life of me what it was for… maybe joycon related?) so perhaps he can chime in and let you know. But based on your readings I’m guessing it’s nothing to be concerned by, likely slight board differences. If it was leaning lower rather than higher I’d dig deeper. Does this rail measure 1.8V following prompt?
Edit - Oops I see in his image it’s relating to digitizer. probably nothing to worry about then (at least not at this stage)
I setup l4t and it freezes on the switchroot logo screen, the same way the switch freezes on the atmosphere screen when trying to boot the cfw.
While in the frozen logo screen there is no voltage on the cpu regulator but I am not sure if there should be at this point
It doesn’t definitively mean the CPU reg is bad it would have just validated it was good if L4T had booted. Often times this style of fault your seeing, it won’t boot into HOS but it will boot into L4T and after you can verify a few other things like charging, (joycons I think), WIFI/BT etc.
So it’s possible it’s your CPU or GPU regulator but given the output is on on the CPU reg it’s somewhat unlikely. Does it come on when you’d expect? just following first logo?
I’m leaning more towards a SoC issue as you suggested earlier just because you mentioned the bent pin in the USB.
You can narrow Ram down a hair (though not fully) by running Biskeydump - for whatever reason Biskeydump can esasurbate the Ram fault issues (this can sometimes be SoC joint issues too) and trigger graphical artifacts on the display (even when Hekate won’t - strange right) so might be worth checking that.
Though given how easy it is to swap the CPU reg over you might just want to do it to rule it out if you have some more of these IC’s handy.
I did swap out the cpu regulator at the beginning when I wasn’t aware it only activates when the switch is pretty much booted, I can swap it out again and see if it makes any difference.
Biskeydump has no issues running
Since I don’t get any normal logos (just the frozen atmosphere or l4t logos) I can’t say if the gpu regulator comes on when expected, I do know it is not on if I remove the nand module which keeps the switch in the first stage of boot i believe, so I can assume that regulator is fine
this is unusual given that the (secondary) CPU rail is coming up… afair high current charge starts just prior to this (or alongside it at the least). I suppose it’s possible you have a secondary issue