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

[SkaziChris]

Dual transistor replaced.
Still a black screen on hdmi - yes, I have re-tried to reset the resolution. No luck.

I'm thinking that it really might be some one small smd, that would cause it...
...and maybe we are wasting time here

 

[RIP-Felix]

I'm thinking that it really might be some one small smd, that would cause it...
...and maybe we are wasting time here

This is the rabbit hole to hell. It's already gone well beyond the point a technition would cut his losses and move on to easier repairs. But at this point it's a revenge project. There's a ninja rabbit in that console laughing and thumbing his nose at us.

I'm curious to see what could possably be causing it, but you have to decide how far down the rabbit hole you're willing to dig.

 

[usernameyeah]

Hi guys, I want to ask for some help with my DYN-001.

I don't know if I am missing something because I don't know much about electronics but I unable to find the PWRGD line/rail in this motherboard and I'm going crazy looking in the wiki or here (and as you can see, my english is poor). I was checking for some voltages but couldn't find much because there are barely schematics. And I ripped syscon pads and it's alternative points, so I can't read that anymore. But I took some photos before that.

The thing is, my main error is 20 3010 and it seems related to CELL and it comes with a 2 seconds YLOD. I don't know if there isn't that CPU buck controller in this mobo. I want that so I can see that power good thing. I checked syscon BE_INIT/ATTENTION pin and I think it's ok, but the voltage it gave me was 3.27 instead of 3.15 according to the wiki. Does that mean anything? I don't have an oscilloscope to test clock signals. So it leaves me with BE_PWRGD (which I don't know where is), BGA problems or the CELL itself is damaged internally. Also I haven't had any 1001 unlike the user that injected 3.3v into PWRGD. I was cheking for shorts but I don't know if I am measuring things correctly with every component.

Is IC6103/IC6201 only in fats PS3? What are some testpoints? How do I even know if my southbridge is fried or not?

I discarted the PSU because the fan spins and the standby LED works. Could it be some regulator of those?

There is other interesting thing about this, sometimes when I start the system it takes more time and then it shut off with a YLOD. When I could check the syscon I found out that in that time it repeats the same sequence of 80 1701, 80 14FF and 80 1802 errors. I'm not quite sure what does it have to do with all of this. Then, in the very first time I checked syscon codes I found 80 1200, then a lot of 80 2022, though I never used the AV port and the ps3 ran in 65 degrees in CELL, not even 70 and it was free of dust. But I noticed through webMAN that the CPU always ran 10 dregrees hotter than the RSX for some reason when I always saw that was the opposite for everyone, and sometimes when I turned off the PS3 it beep three times before shutting down indicating thermal problems, and again, the machine was not close to 70 degrees.

I have got mixed errors and almost every version of 30xx related to CELL, once I got a SB error, once a RSX error, sometimes a CELL-RSX error, I think I will give up because the most plausible cause for me is a BGA problem. I even did a ""reflow"" with a hairdryer but it didn't change anything more than triggering again 1200 thermal CELL error. I don't even think it could be the famous NEC/Tokin.

The last thing I noticed before it died was when I was playing TLOU, the game started to go laggy gradually to the point of stuttering and then it never waked up again.

I just put this here so maybe it could help with syscon errors meaning decoding because I think it is dead and makes me very sad. I thought in buying a fat since it has a lot more of documentation and it's service manuals and use this to practice deliding if it's beyond repair. Thanks in advance.

(Good job all of you have done in this thread, it has been a interesting reading even for someone that barely knows something about this. Keep it up!)

https://imgur.com/a/jZvaZ98

 

[Generic]

There's an unfortunate commonality to your story. You bought the console "used working." Then about a year later it started having the YLOD. That's already suspicious. That's about how long reflows tend to last.

Your errolog confirms the console was reflowed or reballed previously. Probably reflowed, since it only lasted a year. Although I suggest using PS3 Advanced Tools to dump your errorlog, it provides a bit more information and all 32 errors. You said you can get the console to boot.

Before the console was having similar livelocks and checkstops causing the BE Attention signal to go high (1701 caused by either a 1601 or 14FF). That's a warning sign of a BGA going out. Currently I'm thinking that as the cracks propagate in the solder balls the impedance fluctuates during use. This messes with the FlexIO calibration and causes those errors, along with visual artifacts and freezing. It could cause unstable VDDC power, but you'd get 1001 errors. That's the Hypothesis, not quite theory yet.

I would bet that very soon you are going to get a 3034 and 1.5s YLOD on every boot. The console likely needs reballed.

thanks for the detailed answer. Looking very closely at the board the nec/tokin capacitors seem to have slight traces of burning/melting haha.

 

[Killuminati]

But you fixed one of the checksum mistmatchs, see how it matches the llast 2 bytes of the area you readed with the value that appears in the line ?

Addr:0x000039fe should be 0x7350

The cheksum of this area (after the byte you wrote to unlock internal mode) is 0x7350... is just syscon stores it in the EEPROM in little endian (from right to left)

...

Thanks so much for you reply, really appreciate it guys!
I'm sorry, I couldn't access my hardware for two weeks, right after I was getting into this topic.

Anyways I read your reply and ran the command you suggested and this is the output:

>$ r 3E00 200  
r 3E00 200
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
-----------------------------------------------
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FD FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF F
>$

so it seems that the range needs to be filled and the FD changed?

The secound faulty checksum could've been the reason why my PS3 is still in the red blinking state, right?

Thanks and have a good time

 

[RIP-Felix]

I don't like the matrix display. for me, it's more confusing than it is helpful. Maybe if you look at code a lot, but for the uninitiated its overwhelming.

I'd just use the csum command and fix the errors it complains about. Fix all of them and SYSCON will let you out of standby. After that we'll see what nonsense is waiting.

 

[sandungas]

Thanks so much for you reply, really appreciate it guys!
I'm sorry, I couldn't access my hardware for two weeks, right after I was getting into this topic.

Anyways I read your reply and ran the command you suggested and this is the output:

>$ r 3E00 200  
r 3E00 200
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
-----------------------------------------------
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FD FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
FF FF FF F
>$

so it seems that the range needs to be filled and the FD changed?

The secound faulty checksum could've been the reason why my PS3 is still in the red blinking state, right?

Thanks and have a good time

What you posted in the forum is not complete, the output of the command "r 3E00 200" should display 32 horizontal lines (with 16 bytes in each line)... and all the bytes should be FF except 1 byte at the end with 00

Right now we can see you have a byte with value FD, this is what you wrote by mistake... but check again because we are not sure about the others, if you have the PS3 near you try this 2 commands:

>$ r 3E00 100

And

>$ r 3F00 100

The output of the first command is going to look like what you posted (filled with FF... except the FD)
And the output of the second command should be filled with FF... except 1 byte at the end with 00

 

[sandungas]

so it seems that the range needs to be filled and the FD changed?

Yeah, but you just need to overwrite the byte with value FD
I calculated the offset by looking at what you posted... but before writing in it is better to "read" it for a confirmation that the offset i calculated is fine

This way you are reading 1 byte at offset 0x3EE6

>$ r 3EE6 1

If the result if FD it means we found it, so you can proceed to overwrite it by using this command, this writes the byte FF at offset 0x3EE6

>$ w 3EE6 FF

After that run the "eepcsum" command to see if syscon complains about checksums

 

[Killuminati]

Yes I was wondering as well why the output was cut, but this was the actual output on my Mac. Even tried again to confirm.
Anyways, tried it on another Computer and now the output is complete. So the error was on my side, sorry.

The output was complete with all FF and the last with 00, except the one FD.
Fixed it with the commands from your second post, and the checksum was correct.

Now the red light blinking stopped and the light even turns green when pressing the power button. Will reassemble the console now to see if anything works again. Seems like the syscon got corrupted while in storage, don't know how this happens.

Right now I am super happy and really thankful that you guys helped me so far I hope that my console might run again after years in storage, but we'll see.
I'll report how it works out.

Thanks and kind regards

 

[Cypha]

The real reason for ylod is the rsx, reason why sony hasnt mentioned anything as it would of cost them billions to fix as nvidea told them about the issue and the bad design by making the rsx die at a angle was a terrible idea reason why there was a couple rsx revisions and less of the cell, and thats also why the super slim is all at a angle on the board like the rsx to align it, but you maybe wondering how have reflows revived it etc etc and tbh the reason is whatever is in the rsx it somewhat revived it little enough but as more use it completely fries itself thats why reflows only work for the one time, but since we can swap the rsx as its not married to the boards my very clever friend who worked for a graphics comapng has made a new graphics chip for the console and will release it on his channel he made it about 4 years ago and he wanted to wait for how long it would break or ylod and hasnt after hes been using it heavily and sharing to friends and family to keep using it for movies games etc we are both happy to release the GO ( graphics operator) graphics chip only compatible with all the backwards compatible a,b,c and e models as he said the other models are kinda useless and waste of time as they was cheaped out

 

[Shawn Shakir]

The real reason for ylod is the rsx, reason why sony hasnt mentioned anything as it would of cost them billions to fix as nvidea told them about the issue and the bad design by making the rsx die at a angle was a terrible idea reason why there was a couple rsx revisions and less of the cell, and thats also why the super slim is all at a angle on the board like the rsx to align it, but you maybe wondering how have reflows revived it etc etc and tbh the reason is whatever is in the rsx it somewhat revived it little enough but as more use it completely fries itself thats why reflows only work for the one time, but since we can swap the rsx as its not married to the boards my very clever friend who worked for a graphics comapng has made a new graphics chip for the console and will release it on his channel he made it about 4 years ago and he wanted to wait for how long it would break or ylod and hasnt after hes been using it heavily and sharing to friends and family to keep using it for movies games etc we are both happy to release the GO ( graphics operator) graphics chip only compatible with all the backwards compatible a,b,c and e models as he said the other models are kinda useless and waste of time as they was cheaped out

When is this happening?

 

[Cypha]

Also you maybe wondering why does replacing the nectokins make the console work for sometime this is because people are using higher uf allowing more power freedom that boosts the rsx traces eventually burning them out and having a glod as they have fried the rsx hope this shines some light and yes the ps3 design is a complete disaster reason why there was 2 other console revisions so that somy could chill and a fresh start for something that can run the same firmware

When is this happening?

Next week at the latest

 

[RIP-Felix]

... since we can swap the rsx as its not married to the boards my very clever friend who worked for a graphics comapng has made a new graphics chip for the console and will release it on his channel he made it about 4 years ago and he wanted to wait for how long it would break or ylod and hasnt after hes been using it heavily and sharing to friends and family to keep using it for movies games etc we are both happy to release the GO ( graphics operator) graphics chip only compatible with all the backwards compatible a,b,c and e models as he said the other models are kinda useless and waste of time as they was cheaped out

The level of skill and expensive machinery needed to make a processor package is rediculus. I want details! Siffener window, underfill chemistry, die manufacturer and process, solder chemistry, VRAM manufaturer, and exactly who they convinced make it? The idea anyone can do it on their own is pretty rediculus. No offense, but this is too good to be true. And we all know what that means...

Proove it!

PS: it's a bit early for april fools isn't it?

 

[Cypha]

The level of skill and expensive machinery needed to make a processor package is rediculus. I want details! Siffener window, underfill chemistry, die manufacturer and process, solder chemistry, VRAM manufaturer, and exactly who they convinced make it? The idea anyone can do it on their own is pretty rediculus. No offense, but this is too good to be true. And we all know what that means...

Proove it!

PS: it's a bit early for april fools isn't it?

The level of skill and expensive machinery needed to make a processor package is rediculus. I want details! Siffener window, underfill chemistry, die manufacturer and process, solder chemistry, VRAM manufaturer, and exactly who they convinced make it? The idea anyone can do it on their own is pretty rediculus. No offense, but this is too good to be true. And we all know what that means...

Proove it!

PS: it's a bit early for april fools isn't it?

Dont worry im posting pictures of the GO (Graphics Operator) tonight you will be impressed by its design also hes added more capacitors to try down clock the rsx as we all know sony went from i think 2.4 to 3.2 reason why the rsx is the main issue as they overclocked it to much reason why the cell and rsx gets so hot as the rsx is operating quicker and more power which makes the cell work more faster for the rsx reason why the cell is as hot as the rsx but sometimes on a good console is 10degrees cooler
Btw he didnt make it on his own as that would be hard and a lot of stress he designed and configured it and had a lot of help with others making it come true

Ps heres my water cooled ps3 aha and ive noticed for the waterblocks they have to be tightened to a specific tightness to stop a ylod so it's definitely the solder balls

 

[RIP-Felix]

Dont worry im posting pictures of the GO (Graphics Operator) tonight you will be impressed by its design also hes added more capacitors to try down clock the rsx as we all know sony went from i think 2.4 to 3.2 reason why the rsx is the main issue as they overclocked it to much reason why the cell and rsx gets so hot as the rsx is operating quicker and more power which makes the cell work more faster for the rsx reason why the cell is as hot as the rsx but sometimes on a good console is 10degrees cooler
Btw he didnt make it on his own as that would be hard and a lot of stress he designed and configured it and had a lot of help with others making it come true

I would very much like to hear this from the horse's mouth. Any way you can convince your friend to join the forum and tell us about it? I'm interested in the technical details. The RSX has been a black box for a long time. I cannot imagine anyone being able to clone it without technical documents of the original. I would be extremely interested in seeing these or whatever reverse engineering notes exist.

If this is true, it's one of the most exciting developments in the PS3 scene - like ever!

 

[Cypha]

See this is exactly why he doesnt join ps3 reddit and these forums as people and complete c*nts about everything as the ps3 community are act like selfish old men from facebook so calm your shit and let us make some real changes to the ps3 and fix what none has bothered as people laught about old gen as he posted a full video of the GO graphics in working order and his ps3 collection but they got his account banned as they didn't believe he was the creator keep acting like a smart guy and he wont bother showing it, im currently waiting for him to get back home from work for him to send the pictures of it because it's incredible design its sleek matt black and it has some support around the rsx to stop it from moving by drilling tiny holes around the rsx area like a ps3 controller battery plastic holders if u know what i mean just trust the process

Remember stuff did take alot of time especially when we all have lifes to deal with to make money so we couldnt work on it 24/7 on like a couple of hours a day and 97 rsx later from taking them apart and testing

 

[marciolsf]

ps3 community are act like selfish old men from facebook

I dunno man, I've been on this thread from the beginning, and I've always been proud of how freely information has flown in and out of it. I share @RIP-Felix healthy skepticism (this does sound pretty huge) and excitement (it does sound pretty huge!). We've spent a lot of time digging into the hardware are always looking for more information. Anything that you feel comfortable with sharing, please do!

 

[Cypha]

I dunno man, I've been on this thread from the beginning, and I've always been proud of how freely information has flown in and out of it. I share @RIP-Felix healthy skepticism (this does sound pretty huge) and excitement (it does sound pretty huge!). We've spent a lot of time digging into the hardware are always looking for more information. Anything that you feel comfortable with sharing, please do!

I completely understand and it is a big announcement because we need to catch up now as all the other older gens are basically all solved and same with the xbox 360 even Microsoft told us the issue on there documentation that it was the solder balls and they solved that in the jasper version and so on, and the ps3 is still here not being 100% and it's because sony doesnt want to tell is as they are worried about having to pay billions to fix the issue with nvidea as the rsx is the issue but you might ask them why the slim and super slim last for ever and the only faults are them is the power supply or fuses is because they have tweaked the power down alot so this means the rsx isnt pushing as hard and gives it a longer life as the electrons etc isnt rushing through the traces and solder balls at a higher paste causing wear and tear over time i honestly this helps with alot of people as we need to get this issue done and dusted so we can all relax as many people adore this console just like i do

 

[DeadEnd]

I would very much like to hear this from the horse's mouth. Any way you can convince your friend to join the forum and tell us about it? I'm interested in the technical details. The RSX has been a black box for a long time. I cannot imagine anyone being able to clone it without technical documents of the original. I would be extremely interested in seeing these or whatever reverse engineering notes exist.

If this is true, it's one of the most exciting developments in the PS3 scene - like ever!

Don't be naive. The kid is just having some fun...

 

[Cypha]

Don't be naive. The kid is just having some fun...

Kid… im 27 your mad that we fixed the issue before you and hmmm you wanna know why it's because my friends friend knows ken and he admitted its the rsx reason why he left sony as he couldnt cope with the issue but since hes been working on robots hes been kindly enought to give us documents of what the e3 2005 ps3 and how the old cooling was going to be like as as u saw in e3 as there was only one vent and that was on the side and disk tray was at the top rather on the bottom, he literally gave us the full documents and how everything was made a processed and most of all we been able to have the concept design and its incredibly better than the 60gb