Hi everyone, I’ll briefly explain the problem I’m having with this edm-010 board, it arrives with a completely dead 0amp of absorption, checking the board under the microscope I have two capacitors in short circuit right under the southbridge, I change the southbridge and the board starts up, I reassemble everything to carry out some tests and start some games, but after about 5 minutes I get a freeze frame of about 20 seconds and the PS5 turns off generating the following error CE-108262-9, while in uart I get as per the attached image two errors relating to the XDPE phases (I also replaced that but the problem continues to occur). I don’t know if it can be useful but when it gave me the freeze frame the first time when I next restarted the PS5 a pop up appeared with: too many USBs connected, but in reality nothing was connected, do you have any ideas? thank you very much to anyone
Can I consider an APU faulty?
You replaced the VRM controller or the VRMs?
I only replaced the XDPE14286A, do all the vrms need to be replaced? Or just phase 2 and 6? If so, how can I recognize phase 6 and 2?
Thank you very much
two of the eight VRMs are visible separated towards the APU. They don’t have a coil on the RAM side like the other six ones.
The error code string looks a bit funny, because it countains only the error code and the sequence number. Nothing else. No timestamp, no powerstate, no upcause,… The sequence number is at 2004 = Failed to init periphal subsystem. For me it would indicate a pretty early shutdown in the pre-start or power to board phase.
ok… what do you advise me to do? this is a card that was completely dead, I replaced the cxd90061g, since then the system starts correctly, and the problem occurs only when a game is started, after about 5 minutes the image freezes and it turns off after 30 seconds.
Is it worth trying to replace the phase mosfets? If so, which ones are the 2nd and 6th? Or should I just change them all?
The two separat VRM phases have no coil on the RAM side.
The other six have each a coil.
Before pulling VRMs I would check and measure the surounding components from each VRM and compare it to the other ones.
here I am, I’ve been really busy, I checked everything also comparing the components with a working ed033, but everything is ok, I seriously think that it is the APU that does not receive current through those phases, the funny thing is that I have 2 ps5 edm033 and edm 010 with the same problem, and both I replaced the cxd90069g and cdx90061g, I also noticed that sometimes it can’t even turn off, it stays in a frozen image even for hours until I remove the power, and in uart there are no errors, I also replaced its ssd controller, but nothing. my story on the APU could be correct, so I stop wasting time on it?
You have three PS5s with the same problem?
At all three replaced the southbridge and ssd controller and have no UART codes?
It is possible that VRMs could still send power good signal while failing to deliver the ordered power. But a freeze whithout shutting down is interesting.
Wasting time is always a question of you own goals.
the first ps5 would not boot 0amp and no boot sequence with bench power supply, short circuit right under the cdx90061g, replaced and the ps5 boots, installed a game to test and it freezes, the ps5 does not always turn off, I noticed that the ps5 also freezes on the home screen! so I tried replacing the ssd controller with its ram module placed next to it, but nothing has changed, it also freezes when I try to do a factory reset it reaches 99% and stays there for hours! I have two units with this problem an EDM-010 and an EDM-033, the EDM 033 in uart gives me errors in phases 2 and 6 as I imagine previously, but only because after it freezes it turns off (I think it generates errors when it restarts) I’m about to give up! will I try to change the ssd chips? do you know if they are interchangeable? or are they married to the console?
The SSD NAND chips in red contain unique data and shouldn’t be replaced
But I don’t think that is an SSD issue if UART is outputting VRM errors.
ok, and how do you fix this problem?
It is a bit to much all over the place for me:
UART error codes → VRM both phases and XDPE
CE-108262-9 → from my notes more a GDDR6 RAM issue
To many USB connected
Now SSD controller or SSD NAND chips
but if it’s a gddr6 problem, uart should signal me a gddr6 problem? instead no! when it happens and it doesn’t turn off and stays on, in uart the error isn’t even signaled
Hard to tell. If the southbridge notice a problem at preboot an error code is written to the NOR. And usually it stops at this point at shuts off. If this is an issue which isn’t occuring every time, it is possible that the check sometime go through and stops at the next issue or recognize no issue at all and than fail with OS running.
And failing at OS running could lead to shut down/freezes without the southbridge getting the info/signal loss (Power good signal).