PS3 Fault finding YLOD with the SYSCON - First steps and Error reporting

[RCWD21]

Well I don't think is really cmos issues. It was if discharged completely, you can try one with 3v at least. But never seen battery issue on software side. This is hardware issue if sync controller does not work.

I have a brand new cord and a different controller im going to try and see if it changes anything. I'll update once I try it

Sent from my SM-G965U using Tapatalk

 

[RCWD21]

Ok so with powering it on and trying to get past the invisible clock screen, no luck! The controller flashes like it would be if it couldn't pair to anything and then actually shuts off. I thought they were maybe dead so I hooked up another ps3 and check and they pair and are fully charged.

I don't think there's anything wrong with the console hardware wise I think its purely down to a bad update at this point.

Sent from my SM-G965U using Tapatalk

 

[RIP-Felix]

Just to recap. @RCWD21 please confirm this is correct:

• CECHB01
• No video out HDMI or AV.
• Console boots to green LED, but turns off (without error) after 50s or so. It just goes back to solid red standby LED as if nothing happened. No beeps.
  
• No SYSCON errors.
• No shorts or blown fuses.
• Bootloader starts and console should be in XMB.
• Normal powerstate.
• Console not updating timstamp.
• Was on OFW ***EDIT: 2.35***, which does not have recovery menu. Was working fine then. After attempting update to 4.86 is when the above began. Updat must have failed, because if it worked he should be able to access recovery menu. But it doesn't work, which means the update failed.

Based on all of that, my guess is that the FW update went bad and you need to find a way to reflash it. I'm no help with that end of things.

BTW, the errors (1001, 1004, and 1701) were old timestamps and consistent with the console having been shut down incorrectly by previous owners. The becount showed that the console has many more PWR on states than shutdowns and an unreasonably low uptime. That indicates it was used as a server or array for number crunching.

 

[RCWD21]

Just to recap. @RCWD21 please confirm this is correct:

• CECHB01
• No video out HDMI or AV.
• Console boots to green LED, but turns off (without error) after 50s or so. It just goes back to solid red standby LED as if nothing happened. No beeps.
  
• No SYSCON errors.
• No shorts or blown fuses.
• Bootloader starts and console should be in XMB.
• Normal powerstate.
• Console not updating timstamp.
• Was on OFW 3.35, which does not have recovery menu. Was working fine then. After attempting update to 4.86 is when the above began. Updat must have failed, because if it worked he should be able to access recovery menu. But it doesn't work, which means the update failed.

Based on all of that, my guess is that the FW update went bad and you need to find a way to reflash it. I'm no help with that end of things.

BTW, the errors (1001, 1004, and 1701) were old timestamps and consistent with the console having been shut down incorrectly by previous owners. The becount showed that the console has many more PWR on states than shutdowns and an unreasonably low uptime. That indicates it was used as a server or array for number crunching.

Correct. It was updated from 2.35 to 4.86.

And it may have been used for number crunching given the person I bought it from was located in New York. Its incredibly clean inside and out!

Sent from my SM-G965U using Tapatalk

 

[RCWD21]

I also noticed that the there's a command called "firmud" for firmware update. Any idea about that?

Sent from my SM-G965U using Tapatalk

 

[vyktormvmpay25]

Not working recovery mode it means software problems. If working recovery mode and still shut down hardware issue.
For software issues you must extract correctly both nands flash memorys. There are good developers how can check it for you.
Don't play with firmware update while on UART syscon!
That command is for syscon firmware.

 

[RCWD21]

Not working recovery mode it means software problems. If working recovery mode and still shut down hardware issue.
For software issues you must extract correctly both nands flash memorys. There are good developers how can check it for you.
Don't play with firmware update while on UART syscon!

I have a teensy++ 2.0 and everything for it should be arriving here soon. Could you point me in the direction of a good place to read up on how to go about dumping the nand?

I already have the teensy.exe for flashing it.and the most recent NOR/NANDway master zip.

Sent from my SM-G965U using Tapatalk

 

[RIP-Felix]

Makes sense. It would have been an an air conditioned room with HEPA filters to keep dust out. It was common that arrays of hundreds of consoles would be daisy chained to gether for protein folding simulations or what have you. The CELL was the most poweful CPU for that kind of thing at the time and for $500 the B models were the better buy. That story makes the most sense.

Good news there. Since the workload is CPU intensive and not so much GPU, then that console should be reliable, If you can get it flashed.

 

[RCWD21]

Makes sense. It would have been an an air conditioned room with HEPA filters to keep dust out. It was common that arrays of hundreds of consoles would be daisy chained to gether for protein folding simulations or what have you. The CELL was the most poweful CPU for that kind of thing at the time and for $500 the B models were the better buy. That story makes the most sense.

Good news there. Since the workload is CPU intensive and not so much GPU, then that console should be reliable, If you can get it flashed.

That's my hope, I'd really like to get this thing going again and keep it alive. It's one of my favorites in my collection right next to my a01 model that runs cooler than any other console I have.

Sent from my SM-G965U using Tapatalk

 

[RIP-Felix]

It's one of my favorites in my collection right next to my a01 model that runs cooler than any other console I have.

It't off topic, but you must only have hot AF consoles if an A model is the coolest of the bunch. Or maybe you meant the coolest in the other sense of the word. If so, it's hard to disagree with that assessment. I'd go with my tricked out Dreamcast, though.

 

[RCWD21]

It't off topic, but you must only have hot AF consoles if an A model is the coolest of the bunch. Or maybe you meant the coolest in the other sense of the word. If so, it's hard to disagree with that assessment. I'd go with my tricked out Dreamcast, though.

No it literally runs 10c cooler than all the others, no idea why lol and I want a dreamcast one of these days.

But back on topic lol I know you said that nand flashing and stuff is out of your realm of expertise, but could you point me in the direction of some good reading? Like I posted a little while ago I've already got the master zip for the nand and Norway stuff and the teensy. I just want to be as well versed as possible before everything arrives.

Oh and seriously, thanks a ton for helping with all this tonight! Its been a huge headache since I discovered all the syscon stuff

Sent from my SM-G965U using Tapatalk

 

[vyktormvmpay25]

Teensy2++ was a great thread on ps3hax but vanished. If someone here have saved those files along with tutorials I don't really know. Probably live info on ps3 devwiki on flash hardware section.
Wayback machine for teensy. baileyscream thread from ps3hax.

 

[RIP-Felix]

No it literally runs 10c cooler than all the others, no idea why lol and I want a dreamcast one of these days.

But back on topic lol I know you said that nand flashing and stuff is out of your realm of expertise, but could you point me in the direction of some good reading? Like I posted a little while ago I've already got the master zip for the nand and Norway stuff and the teensy. I just want to be as well versed as possible before everything arrives.

Oh and seriously, thanks a ton for helping with all this tonight! Its been a huge headache since I discovered all the syscon stuff

Sent from my SM-G965U using Tapatalk

Nope, not my experience area. I enjoy fixing broken consoles, but this is a new one for me. Everything I can think of requires the recovery menu. You're gonna have to unbrick it manually. @M4j0r @sandungas are more experienced in the hackier side of things. Maybe they can help.

Keep us updated, though. It's a fascinating case. I'm sure @Pacorretaco will be pleased he turned out to be right after all. The 1001,1004, and 1701 error turned out to be old errors from when the console was part of a DATA farm being improperly shutdown, not tokins...lol. If nothing else the SYSCON provided us with a lot more information than you had before. It may have prevented you from harming a good console unnecessarily! So I would say it wasn't a waste of time or money. I'd say the headache was worth it. Besides, it gave you a taste of the type of coding/hacking you'll need to get used to...(knock on wood)

 

[vyktormvmpay25]

I have assumed those errors went after all tests have been done along with delid. Then reading issues with 2.35 then 4.86 remembering that been there. Now been like 8 years back can't remember if I've done 3.55 via dongle in 3.55 or have patched for 3.55 with flasher . On first unit I have done something, then second I updated in stages because of that otheros issues. I remember first was hard sorted then second only via USB updates. It was there in ps3hax, again vanished. Some useful info
https://www.psdevwiki.com/ps3/Flash
https://www.psdevwiki.com/ps3/Teensy++_2.0

 

[sandungas]

@RCWD21 i also think is software, there is a test you have not tryed yet, i guess is not going to work but just incase... try to remove the hdd and boot normally without it... if you see a warning text in the screen is a nice signal (it means the firmware is loading partially)

Another thing you can do by now (while waiting for the flasher and flash clips) is to use the syscon UART to enable the southbridge UART... then solder your "UART to USB adapter" to the tesptads of the southbridge and take a look at the info displayed
As far i did read... the southridge UART reports info about the software bootchain (lvoldr, lv0, bootloader, etc...), if there is some stage of the firmware "broken" maybe you are going to be able to see it there
But thats all i know about the southbridge UART, i cant help with it

------
About the NAND flasher and software... there is an introduction in wiki, most important thing is the software setup and run the command "info"
Basically... the "info" command is like doing a "ping" to the flash chip/s to read his identifyer... if the flash replyes with some valid info thats a victory in itself (it means you have everything ready and well configured)

NANDway.py COM1 0 info

Next thing you should do (as another test) is to run the command indicated as "dump Filename [Offset] [Length]" with small values, lets say... something like this:

NANDway.py COM1 0 dump c:\myflash.bin 0 100

This is going to create an small file (so is fast), open it in a hexeditor to see if it looks like valid data (as far i remember appears the name "sony entertainment..." at the first bytes), and when you are ready make the full dump (this takes time because are 256mb)

NANDway.py COM1 0 dump c:\myflash.bin

Dump it a couple of times (and compare them), just to have a confirmation that you are not having any interference in the wires
Also, download the last version of the "pyps3checker" tool and scan your dump with it, that tool could identify some problems of some damaged areas in the flash

In this subforum there are a bunch of threads related with hardware flashing ---> https://www.psx-place.com/forums/downgrading.101/

 

[RCWD21]

@RCWD21 i also think is software, there is a test you have not tryed yet, i guess is not going to work but just incase... try to remove the hdd and boot normally without it... if you see a warning text in the screen is a nice signal (it means the firmware is loading partially)

Another thing you can do by now (while waiting for the flasher and flash clips) is to use the syscon UART to enable the southbridge UART... then solder your "UART to USB adapter" to the tesptads of the southbridge and take a look at the info displayed
As far i did read... the southridge UART reports info about the software bootchain (lvoldr, lv0, bootloader, etc...), if there is some stage of the firmware "broken" maybe you are going to be able to see it there
But thats all i know about the southbridge UART, i cant help with it

------
About the NAND flasher and software... there is an introduction in wiki, most important thing is the software setup and run the command "info"
Basically... the "info" command is like doing a "ping" to the flash chip/s to read his identifyer... if the flash replyes with some valid info thats a victory in itself (it means you have everything ready and well configured)

NANDway.py COM1 0 info

Next thing you should do (as another test) is to run the command indicated as "dump Filename [Offset] [Length]" with small values, lets say... something like this:

NANDway.py COM1 0 dump c:\myflash.bin 0 100

This is going to create an small file (so is fast), open it in a hexeditor to see if it looks like valid data (as far i remember appears the name "sony entertainment..." at the first bytes), and when you are ready make the full dump (this takes time because are 256mb)

NANDway.py COM1 0 dump c:\myflash.bin

Dump it a couple of times (and compare them), just to have a confirmation that you are not having any interference in the wires
Also, download the last version of the "pyps3checker" tool and scan your dump with it, that tool could identify some problems of some damaged areas in the flash

In this subforum there are a bunch of threads related with hardware flashing ---> https://www.psx-place.com/forums/downgrading.101/

I may just solder directly to the teensy as some of the components are taking forever to arrive...

I forget the command, I think it was "task" but it showed a bunch of percentages and some of them were only up to 60% or so. When I get home later I'll pull it back up and post it. I'm not sure if it's useful info in this case though

Sent from my SM-G965U using Tapatalk

 

[Pacorretaco]

Can't say isn't themal panic of rsx @RCWD21.
It will never show images without delid both.

Actually because his system is on the power ON state (and the hardware is probably OK)
He could just ask the SYSCON for the temperatures

tmp 0
tmp 1
duty get 0


But yeah all things considered it's very possible that the problem was the update... Especially if this was used in a processor farm with linux OtherOS. To which all the evidence points towards.
The update might have gone wrong precisely because of that, because that 4.86 update had to break all the Other OS functionalities that sony really wanted to disable desperately.
I wonder what was there on those OtherOS partitions. I would have thought twice about deleting hehehe.
Maybe the recipe for the krabby patty?

 

[Pacorretaco]

I really hate phats boards and if I

Hahaha, you know I was actually considering sending you a bunch of these boards for some time right? I think you mentioned you were looking for COK 002 s because yours had funny CPU or something.

Inductors

So the idea is to isolate the RSX from the power delivery circuit, and then check resistances huh. Ok. I was thinking of

Spoiler

Removing the tokins

But that's reasonable I guess. I was suspecting the IC voltage regulator mostly.
And now I remember @squeept had a board with abnormal voltage on RSX too. But was a total heatgun catastrophe with toothpaste flux.
My board has most likely been reflowed too, but actually doesn't look that bad to me. Maybe the Voltage regulator IC which is near the RSX could have been damaged by whomever used hot air there?

 

[vyktormvmpay25]

I may just solder directly to the teensy as some of the components are taking forever to arrive...

I forget the command, I think it was "task" but it showed a bunch of percentages and some of them were only up to 60% or so. When I get home later I'll pull it back up and post it. I'm not sure if it's useful info in this case though

Sent from my SM-G965U using Tapatalk

Ah forgot one way simple that can probably will keep unit open as glod. Simply put SB tristate on gnd. We should test that. If an rsod form it's working with nor. Can't remember for Nand. I use to read out of board, found easier to manipulate like that. It should work fine on board with teensy.

 

[vyktormvmpay25]

I can find boards here, but sadly same, all tested on reflow mostly. Like I said I didn't want to put to much effort on phats. I'm more in to slims.