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

[Timbo9876]

The console arrived to me allready dellided (without any obvious visible harm) and with rather fresh paste applied. I did another repaste and tried it without any change. I wasn't so thorough, meaning didn't replace the thermal pads but that was before i dived in the syscon method. Should i try again, should i go straight to reflowing? I don't really care about the outcome since it's more of a hobby than business, but i am not comfortable yet with trying a full RSX reballing (nor do i have the equipment other than a cheap-chinese rework station). Should i just declare it dead and move on? From what i have been reading, 3034 errors is associated with GPU/CPU contact faults. Has anyone succeeded with less complicated methods?

Your latest error was 3010. Now if you look at the YouTube recording your see I was able to determine that this error is pointing towards a CPU substrate issue. So get about 4 small pads and place them in the indentation on the back side of the CPU. Now add 2 large pads and place these on top, put the console back together clamping the pads to perform pressure behind the CPU chip. If you get a boot like i did, you need a CPU reball or a good reflow may well get the console running but may not last as long as a reball

 

[Bazilthemage45]

This has to be one of the more interesting brigups I've seen yet.

Hard to know what to make of it. I have been wondering for a while now if the inpeedance of failing BGA or Bumps could cause 1001 errors, or PLL unlocks. The 3034 seems to confirm this.

OTOH, perhaps the 1001/1301 are related to the tokins and the GPU soldering is failing simultaneously (3034/4402). Don't think I've seen a 1001 occur at step# 40 before. Can't remember off the top of my head anyway.

Thanks a lot for your time and guidance. For the fun of it, i am gonna order some caps and a BGA reballing kit and try to replace the GPU balls and then the tokins if necessary. That might provide an answer.

 

[Bazilthemage45]

Your latest error was 3010. Now if you look at the YouTube recording your see I was able to determine that this error is pointing towards a CPU substrate issue. So get about 4 small pads and place them in the indentation on the back side of the CPU. Now add 2 large pads and place these on top, put the console back together clamping the pads to perform pressure behind the CPU chip. If you get a boot like i did, you need a CPU reball or a good reflow may well get the console running but may not last as long as a reball

Thanks, i will try that in the weekend

 

[squeept]

This has to be one of the more interesting brigups I've seen yet.

Hard to know what to make of it. I have been wondering for a while now if the inpeedance of failing BGA or Bumps could cause 1001 errors, or PLL unlocks. The 3034 seems to confirm this.

OTOH, perhaps the 1001/1301 are related to the tokins and the GPU soldering is failing simultaneously (3034/4402). Don't think I've seen a 1001 occur at step# 40 before. Can't remember off the top of my head anyway.

I have a documented A0801001 with artifacting that was resolved with a GPU swap, and some GLOD 1001 that are the same.

1001 is looking inconclusive or RSX leaning to me, especially if you don't have access to an oscilloscope.

edit: working on a better system of organization for my documentation that will allow me to update previous sheets easily when I circle back around to stuff that's been on the shelf. I want to also be able to post my "failure" sheets as they come in, rather than a year later when I send them to be recycled. I may have to make my own thread for housekeeping, or I'll set something up on my website that I have just for email addresses and file storage...

 

[RIP-Felix]

Thanks a lot for your time and guidance. For the fun of it, i am gonna order some caps and a BGA reballing kit and try to replace the GPU balls and then the tokins if necessary. That might provide an answer.

That's going to be a lot harder than you think. You will not be able to do it with a heat gun, if that's what you're thinking.

I have had sucess with a T8280 and Hot air wand with 45x45mm BGA nozzel, but I killed 2 boards learning to use it. Even with a BGA rework station you are likely to kill the first one (at least).

There's also the need for baking. There's lot's of little tweaks an tricks you must learn and it's all too much in one go. I mean it!

If you are serious about learning to rework BGA, there's a lot of planni g that needs to go into it. You can't just buy a kit and get started.

 

[Bazilthemage45]

That's going to be a lot harder than you think. You will not be able to do it with a heat gun, if that's what you're thinking.

There's also the need for baking. There's lot's of little tweaks an tricks you must learn and it's all too much in one go. I mean it!

If you are serious about learning to rework BGA, there's a lot of planni g that needs to go into it. You can't just buy a kit and get started.

Yeah, i know, i didn't want to sound like i have a false confident on my skill (or in fact lack of it). I am aware of my chances, and i am taking my time trying to understand how things work. Considering that even professional level equipment won't cover for my lack of experience i just wanted to try reballing without spending on something i would barely use. I really appreciate your advise though.

 

[Aran3a]

Hi All

i have bumped into an issue i have not seen before

the unit is a CECHC02 with the error 3040 which refers to the system Flash...

I checked the incoming voltage rails (using the service manual) to the Starship 2 chip and the 2 Flash chips and they all seem to be ok, after doing some research on the net the only things I can find on it were caused by messing with the CFW and were fixed by reprogramming the flash chips from backup

I have spare chips I can swap onto the board but as I understand the firmware needs to be for that specific unit and the flash is not interchangeable…

I don't know the history of this unit because I bought it "for parts or repair" so I also don't have a backup of the Flash

Anyone have any idea what is going on or what I need to do to bring it back?

 

[Oleg-ps3]

Hi everybody.Tell me what is the error ps3 super slim errlog 0000000 CODE CLOCK A0805FFF 2986C076 FFFFFFFF FFFFFFFF FFFFFFFF FFFFFFFF FFFFFFFF FFFFFFFF

 

[Cypha]

would like to note as havent seen much info on certain error code fixes/combos logged yet, error 3034 and 4401 cell related seemed to work after doing a 40nm frankie this is to help others as recently been buying ylod bc ps3s with there launch box it had been opened however and rsx was delidded but not cell so was confused as why it was reporting that "could be they was bending and twisting the board or pressure testing, but gave it a shot and it worked from doing a frakie. looking forward to bringing a mod soon to the table once i feel its worth it and work!

 

[Timbo9876]

would like to note as havent seen much info on certain error code fixes/combos logged yet, error 3034 and 4401 cell related seemed to work after doing a 40nm frankie this is to help others as recently been buying ylod bc ps3s with there launch box it had been opened however and rsx was delidded but not cell so was confused as why it was reporting that "could be they was bending and twisting the board or pressure testing, but gave it a shot and it worked from doing a frakie. looking forward to bringing a mod soon to the table once i feel its worth it and work!

The heat required to do a Frankenstein mod could have flexed the board back into shape. I know that gently heating the board to 160 degrees over a 15 minute period will bring the console back to life that is showing this error.

What I would like to see is some actual testing of the Frankenstein mod. A heat fix like the one I have mentioned above can last years if done correctly. So the question is, is it actually the Frankenstein mod that is fixing the console or is it the heat that's applied to the motherboard that is bringing the console back to life?

Is there a way to test the RSX chip to see if its actually dead? If not then we could be in a position where the board bends over time causing the original BGA issue it had in the first place.

 

[DeadEnd]

would like to note as havent seen much info on certain error code fixes/combos logged yet, error 3034 and 4401 cell related seemed to work after doing a 40nm frankie this is to help others as recently been buying ylod bc ps3s with there launch box it had been opened however and rsx was delidded but not cell so was confused as why it was reporting that "could be they was bending and twisting the board or pressure testing, but gave it a shot and it worked from doing a frakie. looking forward to bringing a mod soon to the table once i feel its worth it and work!

Dude, where are the pics of your graphics operator chip that you promised in the summer?

The heat required to do a Frankenstein mod could have flexed the board back into shape. I know that gently heating the board to 160 degrees over a 15 minute period will bring the console back to life that is showing this error.

What I would like to see is some actual testing of the Frankenstein mod. A heat fix like the one I have mentioned above can last years if done correctly. So the question is, is it actually the Frankenstein mod that is fixing the console or is it the heat that's applied to the motherboard that is bringing the console back to life?

Is there a way to test the RSX chip to see if its actually dead? If not then we could be in a position where the board bends over time causing the original BGA issue it had in the first place.

Not this again...

 

[Timbo9876]

Yep this again!

I'm sorry, I will keep my opinions to myself and you can live in a world of fantasy, just like everyone did with the Tokin hysteria. It wasn't the act of replacing the tokins that fixed the console, it was the heat that was required to remove them. This is likely to be the same scenario, now only time will tell. Its interesting that computer booter only gives a six month Warranty on his Frankenstein mods. This is the same duration a typical reball will last.

 

[alty75]

I think you're misunderstanding my point. I didn't say heating random shit won't cause random devices to randomly work. Just that if the heat is fixing things here, it's NOT the TOKINs that are getting fixed, it's something else. They aren't the same chemistry as ones where that can actually do something.

It's an issue with a filter - there is no fooling the scope here. It's the one right tool. Or two, but I don't have a spectrum analyzer. That's why it's the only proof I'm willing to accept that the heat isn't, say, temporarily reconnecting a BGA or bump nearby.

I've tested this theory many times. A cap issue in the log and apply 5-10 seconds of heat with the wand @ 100 C and it boots. I highly doubt that is enough heat nor time to reconnect a bump or bga ball with very little heat from the want going any further than the caps area but it certainly allows the console to boot. Stops booting on the next few boots and then reapply a little direct heat to the caps in question and it boots again. I've seen and done it many times for fun lol

 

[RIP-Felix]

He's trollong. Please nobody bite.

 

[alty75]

He's trollong. Please nobody bite.

Huh!
Me?
Just giving some real world experience lol. Go back under your rock if you are pointing your finger at me lol

 

[Shawn Shakir]

Got 3034 on a COK-001 after reballing a 40nm from a slim. Placed same RSX on two different consoles with same results.

VDDC - 2.7
VDDIO - 49.1
FBVDDQ - 228.5
PLL_VDD - 5.591K
RC_VDDA - 3.6
VDDR - 426.8
YC_RC_VDDIO - 12.5

bringup
[SSM] state: 0000 -> 0101
Bringup Mode #0 (0xFF)
[SSM] ssmCb_OnStartingBePowOn() called.
[SSM] Bringup mode : syspm_stat=00000000/00000000
[POWSEQ] PowerSeq_Setup called.
[SSM] state: 0101 -> 0201
[POWSEQ] AV Backend Setup
[SSM] state: 0201 -> 0102
[SSM] state: 0102 -> 0202
[SSM] state: 0202 -> 0103
[SSM] state: 0103 -> 0203
[SSM] ssmCb_BeforeBeOn() called.
[SSM] state: 0203 -> 0104
Psbd_SbTransMode_Half:0x20e2
>$
[POWERSEQ] Error : BitTraining RSX:RRAC:RX3:GLOBAL1:RX_STATUS
[SSM] state: 0104 -> 0304
[SSM] ssmCb_AfterBeOn2() called.
[SSM] PowSeq Fail : Detected !
[SSM] state: 0304 -> 0700
[POWSEQ] AV Backend Letup
[SSM] Shutdown mode : syspm_stat=00000000/00000000
[ERROR]: 0xa0403034
[POWSEQ] PowerSeq_Letup called.
[SSM] state: 0700 -> 0600
(PowerOff State) (Fatal)

 

[DeadEnd]

Yep this again!

I'm sorry, I will keep my opinions to myself and you can live in a world of fantasy, just like everyone did with the Tokin hysteria. It wasn't the act of replacing the tokins that fixed the console, it was the heat that was required to remove them. This is likely to be the same scenario, now only time will tell. Its interesting that computer booter only gives a six month Warranty on his Frankenstein mods. This is the same duration a typical reball will last.

I'm sorry, there is no need to be condescending to each other over this. I may react a bit crass , it's just you are not the first one to argue about the theories. But how do you imagine the heat to fix a console when the whole gpu is replaced ? Once the original part is gone, you can't prove it anymore. And let's say that something fails later, how are you going to prove that it's not a new fault that developed and had nothing to do with the "heat fix" ? And all of this is also ignoring the fact that GPUs of that day had issues...But this was already covered quite well in the Felix's video.

6 months is a pretty standard warranty for a repair shop. Sony offered even less irc. Another reason could be that the early models are in general not highly reliable. GPU swap can extend their life... But the system is not fault proof after that, because new faults can develop. And getting returns or diagnosing the same console over and over again is not exactly good for any business. It's a problematic console in general, and I can bet that's why not many people are offering the repairs.

I will for once agree with Paco on this one, the gpu swap is not the ultimate fix. I've talked about it too. It shouldn't be treated as such.

 

[DeadEnd]

Got 3034 on a COK-001 after reballing a 40nm from a slim. Placed same RSX on two different consoles with same results.

VDDC - 2.7
VDDIO - 49.1
FBVDDQ - 228.5
PLL_VDD - 5.591K
RC_VDDA - 3.6
VDDR - 426.8
YC_RC_VDDIO - 12.5

bringup
[SSM] state: 0000 -> 0101
Bringup Mode #0 (0xFF)
[SSM] ssmCb_OnStartingBePowOn() called.
[SSM] Bringup mode : syspm_stat=00000000/00000000
[POWSEQ] PowerSeq_Setup called.
[SSM] state: 0101 -> 0201
[POWSEQ] AV Backend Setup
[SSM] state: 0201 -> 0102
[SSM] state: 0102 -> 0202
[SSM] state: 0202 -> 0103
[SSM] state: 0103 -> 0203
[SSM] ssmCb_BeforeBeOn() called.
[SSM] state: 0203 -> 0104
Psbd_SbTransMode_Half:0x20e2
>$
[POWERSEQ] Error : BitTraining RSX:RRAC:RX3:GLOBAL1:RX_STATUS
[SSM] state: 0104 -> 0304
[SSM] ssmCb_AfterBeOn2() called.
[SSM] PowSeq Fail : Detected !
[SSM] state: 0304 -> 0700
[POWSEQ] AV Backend Letup
[SSM] Shutdown mode : syspm_stat=00000000/00000000
[ERROR]: 0xa0403034
[POWSEQ] PowerSeq_Letup called.
[SSM] state: 0700 -> 0600
(PowerOff State) (Fatal)

Well you forget to mention that you took the 40nm out of a dead slim and didn't check its errors before placing it onto the COK board. but in this case the bittraining error reveals the internal issue with the GPU. Either it died in a slim or during the swap. They can fail too, just not as often.

 

[alty75]

Well you forget to mention that you took the 40nm out of a dead slim and didn't check its errors before placing it onto the COK board. but in this case the bittraining error reveals the internal issue with the GPU. Either it died in a slim or during the swap. They can fail too, just not as often.

If it was off an KTE then there is a high possibility that it was foobarred prior to lifting it to use elsewhere

 

[Pacorretaco]

Not this again...

Yep this again!

He's trollong. Please nobody bite.

Hmm, did I miss some kind of party or something?

In short, once you see 3034, it's game over for the GPU. There's 3034 in all of your consoles.

Well, that might be true for people that can't reball, never reballed or don't want to reball. It is a very difficult thing, so unfortunately is most people.
In other words those who can't or won't lift the chip from the board without destroying. Then yeah.

But actually I didnt come here to argue, so nobody worry... There can be peace again.
Anyway I came here for different reason:

OK, In wiki, after rereading, this should be nec tokin error the A0801001 ....

1001 is looking inconclusive or RSX leaning to me, especially if you don't have access to an oscilloscope.

Yes, these are old stories but I noticed that lately, some people are still getting confused with this 1001 error, meaning the Wiki was still not clear enough.

So I tried to clarify the entry a bit:

If somebody have something to remark or add to make it better feel free to discuss it. I think @RIP-Felix was the last person to write there

Cheers