PS3 Need the experts help: PS3 shuts down during one game without beeps or Syscon error codes

[Tamás Orosz]

I have a weird PS3 issue - the PS3 shuts down during Killzone 2 only, no 3 beeps, only a solid red light. Any ideas?

When looking for SYSCON error logs, I see a bunch of old A0801001s and one A0801004, but no new error entry is being generated after the shutdowns. I'm curious what the issue could be. This is an OG phat 60GB PAL console (I live in europe) and I noticed this weird issue while I was playing Killzone 2 singleplayer and during the end of the first big battle sequence, right as the game would load the next cutscene when you get out of the tank, the PS3 shuts off completely, without the usual 3 beeps, instantly switching to a red light the second the screen goes black. Now the PS3 would not start with the PS button on the controller, only by pressing the power touch button on the front.
The weird thing is that it only does this with Killzone 2, but not always at the same place. Sometimes at the end a of a loading screen, sometimes during gameplay, but I cannot really reach the end of the first chapter. Any other game works fine, playing for hours, Uncharted games, even the Last of Us. Now I checked this Killzone disc with a stock slim, no issues, completed the first 2 chapters. I reinstalled the patch, I deleted the save data, same issue on the phat. I did a database rebuild and file system check also. Usually when we saw similar issues, you can suspect 2 things:

• NEC/Tokin caps are failing, and the PS3 cannot get the correctly regulated juice to operate during a taxing game. But this is always followed by the 3 beeps and SYSCON error logs, but there are no beeps, or flashing red light. Just a solid red light after the shutoff in an instant.
• The other culprit can be a failing PSU, but when a PS3 shuts off during gameplay with a PSU issue, it's usually followed by no light whatsoever and you have to switch off/on the main power switch on the back to have a red light again and then you can power the machine on again.

Other info: When the shutoff happens and I restart the machine, it says that it was turn of incorrectly and it needs to do the database rebuilding. Temps are perfect, during gameplay, 60c the max it reaches (both on CELL and RSX). It runs the latest 4.91 evilnat CEX CFW. The issue still happens if I copy the game to the HDD.
So what could this be? I have no idea, I've never seen this combination of issues on any PS3 forums, and I thought I was well versed in the wide range of issues you can encounter. I would love the hear your opinions. And thank you in advance for anyone to help me!

 

[Tamás Orosz]

@RIP-Felix I love your work in the PS3 scene! Sorry for tagging you directly, but as in Star Wars when Ben Kenobi sees the hologram recording of Leia "you're my only hope...". I looked everywhere and I did not see this issue elsewhere on forums, youtube or reddit. Do you have any ideas what could cause this? Should I try a PSU swap? Without the SYSCON logs I feel like I have no lead on this. I tried an HDD swap, that did not fix it. My main suspect here is the PSU.

 

[spla]

Good day, are the temps for the GPU/CPU within a manageable range when it shuts down? I know someone with such an issue, just curious if it could be the die overheating.

 

[Tamás Orosz]

Good day, are the temps for the GPU/CPU within a manageable range when it shuts down? I know someone with such an issue, just curious if it could be the die overheating.

Hi! Thank you for chiming in. Yes the temps are perfect. This is a delidded, modded machine, with good quality thermal paste and carefully managed fan curves. When the shutdown happens the RSX and CELL are at the 59-62 celsius range. And the shutdowns are happening at various parts of the mission, but always at the same general moments. Before cutscenes, at a certain checkpoint, things like that. 9 out of 10 times I can replicate it at the end of the first battle with the tanks. Sev climbs out of the tank after the battle and boom. As you see in the video. Can it be that at this part in the game, the PSU would need more juice and some caps are failing in that and that's the reason of the shutdown? The problem is, that I don't get SYSCON error codes. It's really hard to get a lead on this without it.

 

[Tamás Orosz]

It appears that it was the PSU. I swapped in a PSU from another unit (same CECHC04 model). And it works fine now. Even with the digital backup of the game. Although the mystery is not entirely solved because, just to confirm the theory, I put the "faulty" PSU back to the donor machine and guess what. It works fine too. A stress tested both machines for hours. Go figure.... Maybe there was some debris in the PSU connector that was dislodged during the swap? I don't know. And I hope I'll never know cause it will never reappear.

 

[spla]

It appears that it was the PSU. I swapped in a PSU from another unit (same CECHC04 model). And it works fine now. Even with the digital backup of the game. Although the mystery is not entirely solved because, just to confirm the theory, I put the "faulty" PSU back to the donor machine and guess what. It works fine too. A stress tested both machines for hours. Go figure.... Maybe there was some debris in the PSU connector that was dislodged during the swap? I don't know. And I hope I'll never know cause it will never reappear.

Any chance you swapped them back to see if it worked with the original PSU again?

 

[Tamás Orosz]

Any chance you swapped them back to see if it worked with the original PSU again?

Well I would rather not. If both machines are happy now, I'm happy too. If the issue reappears on the donor machine, then I'll know. But for now, I think I'm good. Do you have an explanation, or theory of what this could have been?

 

[GuilloteTesla]

Well I would rather not. If both machines are happy now, I'm happy too. If the issue reappears on the donor machine, then I'll know. But for now, I think I'm good. Do you have an explanation, or theory of what this could have been?

Maybe one of your consoles asks for a little bit of power that the PSU couldn't deliver, pushing it to the limit. It could be some electronic component that is faulty or doing some crazy things which makes the PSU do random stuff.

It could also be a bad connection. We will never know until the same behaviour happens again.

 

[Tamás Orosz]

Maybe one of your consoles asks for a little bit of power that the PSU couldn't deliver, pushing it to the limit. It could be some electronic component that is faulty or doing some crazy things which makes the PSU do random stuff.

It could also be a bad connection. We will never know until the same behaviour happens again.

Yeah... I hope I'll never know... but I still kinda would want to know. Yeah these are all sound theories. Thank you for sharing! Appreciate it!

 

[Tamás Orosz]

Maybe one of your consoles asks for a little bit of power that the PSU couldn't deliver, pushing it to the limit. It could be some electronic component that is faulty or doing some crazy things which makes the PSU do random stuff.

It could also be a bad connection. We will never know until the same behaviour happens again.

I realized what the issue was. The PSU that caused issues was an APS-231. I just noticed it now. That PSU is known to cause random shutoffs during certain games, and yeah I can also confirm that. The weird things still is that it does not cause shutoffs in a different CECHC04 unit... weird...

 

[Raidmax]

APS-231 units should be discarded completely IMO. They are the second most unreliable units besides the original ZSSRs, and even in a CECHG unit they heat up excessively. An APS-226 or 227 should be used on any unit in the CECHA-CECHG range.

I know this might be a hot take, but I did test this with 2x G04s. Both would heat up in excess with the 231s, both ran considerably cooler after switching both to 226 and 227 units respectively.

 

[Tamás Orosz]

APS-231 units should be discarded completely IMO. They are the second most unreliable units besides the original ZSSRs, and even in a CECHG unit they heat up excessively. An APS-226 or 227 should be used on any unit in the CECHA-CECHG range.

I know this might be a hot take, but I did test this with 2x G04s. Both would heat up in excess with the 231s, both ran considerably cooler after switching both to 226 and 227 units respectively.

For me it was exactly the opposite. In my CECHC04 with the APS-231 installed the CPU was constantly on 65c with 35% fan speed. Now after I switched to an APS-227 the CPU is 68c with the same fan speed. So for me it was actually cooler with the 231.