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

[dumbass370]

RSX swap didn't help, error is stil 0xa0403034

 

[superprokill1]

Hey all, coming here back again. After a successful recap with tantalizers, I am now stumbling across error 0xA0801103. This is part of the "Thermal ALERT System", but in my case it's a bit strange. The console turns on for a few minutes, before eventually turning off. I think it would turn off immediately without any notice such as a triple beep when overheating.
I used to have issues with the RSX, but after 3D printing an eraser mod alternative for the RSX, it worked wonderfully after and didn't overheat anymore (I suspect this had to do with my shitty soldering job with the caps, since I think there's a giant blob of solder below one of the caps legs in the negative pad, which raised the metal plate just slightly enough to cause issues with the RSX. I suspect that if I had to do the same with the CELL, it would at least give me a notice or try to turn up the fans before shutting down without notice, right?)
Now, what I think could be the issue, is with the thermal pads. They were old, were exposed to air for months while they were taken out of the components they were protecting, and when assembling/disassembling the console, they would fall off immediately. I tried putting them back on, but, could it be that 1103 is related to these thermal pads in any way? If so, would repadding help?
I was able to get this error by putting my PS3 on an AC set to 16C while I did all the work to run ps3toolset and get the errors out of the console, so, it's definitely thermal and not an erroneous code.

 

[Ganjo]

RSX swap didn't help, error is stil 0xa0403034

The RSX you put on was bad. If it was a 90nm then theres a 90% chance its the RSX.
A solder ball didnt attach or merged.
If RSX_VDDR is missing it will give 403034
The CELL is bad or needs a reball
The SB is bad or needs a reball

 

[Ganjo]

since I think there's a giant blob of solder below one of the caps legs in the negative pad, which raised the metal plate just slightly enough to cause issues with the RSX

Gonna need to see pics.

If the CELL VRMs overheat they drop a different error code. Id check for solder splashes or trace damage.

 

[Ganjo]

I had an A01 today that had 403034 and 902120 errors. Replacing the RSX fixed the problem but when I first started it up I had a 603040 error. After removing the top metal shield to test voltages I noticed the system turned on normally. After inspecting the metal shield I could see one of the metal tabs that are supposed to touch the board had folded over and was causing a short. After I bent it back into shape the console booted normally.

Just thought I would add this in case someone comes across the same issue.

 

[WailkyS5M5]

Hello everyone,
I thought of talking about my CECHC04 with a weird issue as I couldn't find much on the Internet about it

Basically, it has around 10 days of runtime, I have repasted it undervolted etc. Rsx is under 55C Cell too this that still using its original NEC/TOKINs, no 1001 or 1002 errors when playing games, im using a custom syscon fan curve

At first I undervolted it to CELL 1.125V and RSX 1.15V by using the syscon clamp method. It ran fine and stable I played Yakuza and Yakuza 2 PS2 mode fine with it

Its like when I soldered in permanent syscon wires, I was testing around further with undervolting and my limit was CELL 1.0875V and RSX 1.05V. Many games ran completely fine for hours so I deemed this to be stable (?)

Now here comes the issue, whenever I turn on the console on a fresh new day (overnight console was not powered on, no standby) it would boot normally to the xmb, but as soon as I load any ISO, homebrew or app it goes to the black screen, HDD light and wifi light blinks for a little bit, then HDD light turns off wifi shortly after...

... and im met with YLOD, syscon errors 1601 and 1701 in the same time frame as a pair

Now here is the thing, when I turn it back on again and try to load whatever gave me ylod it works fine and I can use/play whatever i launched for hours and switch games etc. Till I turn it off myself. Even the most demanding games out there work absolutely fine

Now out of common sense I thought it would be a udnervolting issue. So I raised CELL firdt back to 1.125V and it still did it

It was up until I set the RSX to 1.0875V that it didnt do it anymore BUT I once left it without power for a night, a full on day and a half a day and the same symptoms happened.

I remember also that when it YLOD i was on my phone sesrching for anythinf about it and around 10 mins after I turned it on and it took me another try to make it load fine

So it is definetly something time related.... but what is going on? 1601 1701 are so generic that it tells you absolutely nothing sadly

Im running a noBD firmware with solely the Daughterboard attached (previous owner botched connector clamp on mobo and a few pins) Im able to load ISOs with my setup but I cant remarry the daughterboard/drive

On webman I set the xmb to have 350/400 clock speeds and apps, games 500/650

My CMOS battery is dead but im about to replace it with a new one by resoldering those wires to a new CR2032 battery.

My SD CARD reader is kind of weird, i checked the ribbon cable and it seemed to be fine but for some reason the memory stick pro light is always on and the CF light has a faint glow to it, Sd cards wont read (back then it did work fine so it must be something during reassmeble)

I left it disconnected as I was worried it could give more issues?

What could I try next to diagnose this issue? Its like can I just cope with it or might it lead to something terrible with time...

 

[Ganjo]

Hello everyone,
I thought of talking about my CECHC04 with a weird issue as I couldn't find much on the Internet about it

Basically, it has around 10 days of runtime, I have repasted it undervolted etc. Rsx is under 55C Cell too this that still using its original NEC/TOKINs, no 1001 or 1002 errors when playing games, im using a custom syscon fan curve

At first I undervolted it to CELL 1.125V and RSX 1.15V by using the syscon clamp method. It ran fine and stable I played Yakuza and Yakuza 2 PS2 mode fine with it

Its like when I soldered in permanent syscon wires, I was testing around further with undervolting and my limit was CELL 1.0875V and RSX 1.05V. Many games ran completely fine for hours so I deemed this to be stable (?)

Now here comes the issue, whenever I turn on the console on a fresh new day (overnight console was not powered on, no standby) it would boot normally to the xmb, but as soon as I load any ISO, homebrew or app it goes to the black screen, HDD light and wifi light blinks for a little bit, then HDD light turns off wifi shortly after...

... and im met with YLOD, syscon errors 1601 and 1701 in the same time frame as a pair

Now here is the thing, when I turn it back on again and try to load whatever gave me ylod it works fine and I can use/play whatever i launched for hours and switch games etc. Till I turn it off myself. Even the most demanding games out there work absolutely fine

Now out of common sense I thought it would be a udnervolting issue. So I raised CELL firdt back to 1.125V and it still did it

It was up until I set the RSX to 1.0875V that it didnt do it anymore BUT I once left it without power for a night, a full on day and a half a day and the same symptoms happened.

I remember also that when it YLOD i was on my phone sesrching for anythinf about it and around 10 mins after I turned it on and it took me another try to make it load fine

So it is definetly something time related.... but what is going on? 1601 1701 are so generic that it tells you absolutely nothing sadly

Im running a noBD firmware with solely the Daughterboard attached (previous owner botched connector clamp on mobo and a few pins) Im able to load ISOs with my setup but I cant remarry the daughterboard/drive

On webman I set the xmb to have 350/400 clock speeds and apps, games 500/650

My CMOS battery is dead but im about to replace it with a new one by resoldering those wires to a new CR2032 battery.

My SD CARD reader is kind of weird, i checked the ribbon cable and it seemed to be fine but for some reason the memory stick pro light is always on and the CF light has a faint glow to it, Sd cards wont read (back then it did work fine so it must be something during reassmeble)

I left it disconnected as I was worried it could give more issues?

What could I try next to diagnose this issue? Its like can I just cope with it or might it lead to something terrible with time...

Your RSX undervolt is too low. The 1.0875v setting, which is closer to 1.045 if you check with a DMM, is too low for a 90nm RSX. The other issues are common undervolt problems where if you leave it turned off for a day or two you can get YLOD or really odd behavior from the console. I have always fixed it by increasing the RSX voltage.
If you want to prove or disprove its the undervolt just set it back to FF and see what happens, dont do half a job and raise the setting slightly and say its not the problem. You should set both to FF, verify system stability, then undervolt 1 processor at a time with long testing sessions. When everything is stable unplug the console for 1-2 days then retest.

 

[WailkyS5M5]

Your RSX undervolt is too low. The 1.0875v setting, which is closer to 1.045 if you check with a DMM, is too low for a 90nm RSX. The other issues are common undervolt problems where if you leave it turned off for a day or two you can get YLOD or really odd behavior from the console. I have always fixed it by increasing the RSX voltage.
If you want to prove or disprove its the undervolt just set it back to FF and see what happens, dont do half a job and raise the setting slightly and say its not the problem. You should set both to FF, verify system stability, then undervolt 1 processor at a time with long testing sessions. When everything is stable unplug the console for 1-2 days then retest.

Yes, its the next thing I will try. I removed that underclocking on XMB little gimmick as I think the Values jump from low to stock values when entering something and it might stress out the RSX... I noticed different symptoms after doing that

 

[Cosmic-R]

so one thing i noticed, iirc the error code #0 is the very last one right ?. in my case is 5FFF
i will just explain the chain if events and maybe someone can better explain to me what really happened

basically i was playing a game like usual (i had webmans fps overlay active dunno if it had anything to do with it)
i then quit the game, let the ps3 idle on xmb for like 3 minutes then proceeded to boot up a homebrew
console shut down completely to NLOD, i just pressed the power button, console beeped to red stand by, press power button again and console boots up without problem and asks to check filesystem as usual.

last time this happened was around 3-4 years ago ? that time i was booting up a game from a mounted .ISO, same deal again, console boots without problems try same iso again and works without issue.

so with my very limited knowledge i would assume i got some sort of system crash, or a kernel panic, or something with the hypervisor or something like that, because the rest of the error codes are just 2022 another lone 5FFF (from 3-4 years ago) and FFFF dated 1970.

console is KTE-001, HFW 4.86, HEN. other than those very rare hiccups only when booting stuff console is in perfect working order

 

[oiamlaffin]

I recently got an interesting CECHB00 from Japan. Was listed as a working console, but when I received it that was only kind of true.

Noticed the console had been opened, it powered on, booted to XMB then after about 45sec it YLOD. I decided to pull it apart and get syscon errors and found an absolutely atrocious attempt at replacing the back tokins with capacitors. but thankfully no shorts or major damage to the board. Pulled the board out and found the entire cooler shroud was melted/warped, this lines up with what I found in the syscon errors later.

Pulled syscon errors to find 1001 and 1002 being the most recent, but the backlog also showed signs of a major thermal event (hence the melted cooler shroud) and 3034 errors. Found it interesting to see south bridge thermal error 1214, followed up by 3034 shortly after. The following codes show the console got really hot, throwing pretty much every error for thermal possible.

I decided to throw some proper tokin replacement boards in, swap the entire cooler for a non melted one and even undervolted it a bit. So far its working flawlessly, I'm going to keep using it and see if the 3034 comes back! Has anyone seen a 3034 caused temporarily by a major overheat?

 

[feng_ye]

I recently got an interesting CECHB00 from Japan. Was listed as a working console, but when I received it that was only kind of true.

Noticed the console had been opened, it powered on, booted to XMB then after about 45sec it YLOD. I decided to pull it apart and get syscon errors and found an absolutely atrocious attempt at replacing the back tokins with capacitors. but thankfully no shorts or major damage to the board. Pulled the board out and found the entire cooler shroud was melted/warped, this lines up with what I found in the syscon errors later.

Pulled syscon errors to find 1001 and 1002 being the most recent, but the backlog also showed signs of a major thermal event (hence the melted cooler shroud) and 3034 errors. Found it interesting to see south bridge thermal error 1214, followed up by 3034 shortly after. The following codes show the console got really hot, throwing pretty much every error for thermal possible.

I decided to throw some proper tokin replacement boards in, swap the entire cooler for a non melted one and even undervolted it a bit. So far its working flawlessly, I'm going to keep using it and see if the 3034 comes back! Has anyone seen a 3034 caused temporarily by a major overheat?

That melted plastics is probably from a "oven reflow" trick. The temperature sensor will shut the console down way before such temperature is reached.

And unfortunately without a proper fix it's almost guaranteed 3034 will come back

 

[oiamlaffin]

That melted plastics is probably from a "oven reflow" trick. The temperature sensor will shut the console down way before such temperature is reached.

And unfortunately without a proper fix it's almost guaranteed 3034 will come back

I didn't even consider someone would try to put the entire board and cooler into the oven as a unit, that is absolutely absurd and beyond lazy, even for a half assed attempt.

I guess time will tell!

 

[DoublesAdvocate]

I didn't even consider someone would try to put the entire board and cooler into the oven as a unit, that is absolutely absurd and beyond lazy, even for a half assed attempt.

I guess time will tell!

Is it just the cooler that's warped or the rear panel just below the serial number as well?
It's more likely that instead of putting the entire console in the oven they instead stuck a hairdryer or heatgun in the back and just left it on for several minutes.

 

[oiamlaffin]

Is it just the cooler that's warped or the rear panel just below the serial number as well?
It's more likely that instead of putting the entire console in the oven they instead stuck a hairdryer or heatgun in the back and just left it on for several minutes.

Only the cooler, the lower shell and back panel are in great shape, I would expect to see massive warping on them too if heat was introduced from the outside. It is equally warped all the way around too, CMOS battery holder looks totally fine as well as the fan.

 

[marianito1984]

Hello, I have a short circuit in a line going to the Wi-Fi module. It goes directly to a pin on the sysconn chip. I isolated it, but compared to another motherboard, the value is very low, and I'm getting an error as if it were a Wi-Fi IC failure. After replacing the entire module along with its BIOS, I noticed this short circuit. My question is, is it possible to create a backup like on a PS4 and install it on another sysconn chip? Thanks.

 

[marianito1984]

Hi, can I dump data on a Slim 3000 series chip? I have a short circuit on a line going to the Wi-Fi. Can I read the data and transfer it to another chip?

 

[oiamlaffin]

Been working on a CECHA00 recently, 72 days run time, 1001 and 1002 errors in syscon.

Pulled the console apart, cleaned it up, delidded both cell and RSX, both came off easily and clean, no damage to substrates or die for either. Cleaned everything up with 99% iso, reassembled and was met with GLOD.

Pulling syscon still only shows 1001 and 1002. Since syscon was still hooked up I decided to undervolt the Cell just to see what would happen and I actually got video output from the console! Would show the initial PS3 screen, controller would sync, then the video output would corrupt with artifacts and the system would become un-responsive. Decided to replace the cell tokins and the console has now returned back to GLOD.

Guessing that even though I was super careful with the delid, something got damaged, but I have never seen anyone undervolt a console with GLOD just to see what would happen. Guess this is another board for the pile of "maybe repairable"

 

[Ganjo]

I recently got an interesting CECHB00 from Japan. Was listed as a working console, but when I received it that was only kind of true.

Noticed the console had been opened, it powered on, booted to XMB then after about 45sec it YLOD. I decided to pull it apart and get syscon errors and found an absolutely atrocious attempt at replacing the back tokins with capacitors. but thankfully no shorts or major damage to the board. Pulled the board out and found the entire cooler shroud was melted/warped, this lines up with what I found in the syscon errors later.

Pulled syscon errors to find 1001 and 1002 being the most recent, but the backlog also showed signs of a major thermal event (hence the melted cooler shroud) and 3034 errors. Found it interesting to see south bridge thermal error 1214, followed up by 3034 shortly after. The following codes show the console got really hot, throwing pretty much every error for thermal possible.

I decided to throw some proper tokin replacement boards in, swap the entire cooler for a non melted one and even undervolted it a bit. So far its working flawlessly, I'm going to keep using it and see if the 3034 comes back! Has anyone seen a 3034 caused temporarily by a major overheat?

403034 does not always mean a BGA defect. Improper voltages can also can 403034, so a overheating system could theoretically throw a 403034 if something got too hot and output an improper voltage. From the look of it and all of the error codes its very likely someone wrapped this in a towel and tried to fix it by overheating it.

 

[oiamlaffin]

403034 does not always mean a BGA defect. Improper voltages can also can 403034, so a overheating system could theoretically throw a 403034 if something got too hot and output an improper voltage. From the look of it and all of the error codes its very likely someone wrapped this in a towel and tried to fix it by overheating it.

I had considered this too, hilarious that anyone used to think that would fix anything. ABS plastic can deform around 90-100c if exposed for too long.

When I read through syscon errors I noticed 1214 could be related to the 3034, southbridge hitting thermal limits, shutting down maybe triggered the 3034. So far the console is still running fine!