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

gillangit said:
hallo everyone,
im get errolog
> errlog
00000000
# CODE CLOCK
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF

i follow tutorial to change all cap in rsx like this:
ZcSZsNY.jpeg

note: this image not my settings.

my setting is
in back side :
3x 2.5 470uf
3x 2.5 470uf

in front side :
2x 2.5 470uf
2x 2.5 470uf

when i tested error is same in A0093004, please help me
Click to expand...
A0093004 = VDDC RSX Failure. NEC/TOKINs & related SMDs. Given that you have been working on that area the most likely reason is a short, or the bridge wires are not connected properly. Double check soldering, polarity, and confirm the RSX is not short to GND. Anything less than 1 ohms is bad. Can't use buzzer! It'll always buzz on low impedance lines like this. Measure ohms to ground. Short leads to know what your meter reads Short as. Compensate in your recorded measurment. Healthy range is 2 - 4 ohms.
 
TCLL said:
OK so I got another console which stopped working after a very clean delid. I was extremely careful as always. So I soldered my TTL and read the syscon log:

===================================
ERR 00: 00000000 A0403034 FFFFFFFF
ERR 01: 00000000 A0404411 FFFFFFFF
ERR 02: 00000000 A0403034 FFFFFFFF
ERR 03: 00000000 A0404411 FFFFFFFF
ERR 04: 00000000 A0403034 FFFFFFFF
ERR 05: 00000000 A0404411 FFFFFFFF
ERR 06: 00000000 A0403034 FFFFFFFF
ERR 07: 00000000 A0404411 FFFFFFFF
ERR 08: 00000000 A0403034 FFFFFFFF
ERR 09: 00000000 A0404411 FFFFFFFF
ERR 10: 00000000 A0802124 0B49F064
ERR 11: 00000000 A0801002 0B49F064
ERR 12: 00000000 A0802030 FFFFFFFF
ERR 13: 00000000 A0802130 FFFFFFFF
ERR 14: 00000000 A0802131 186B29C4
ERR 15: 00000000 FFFFFFFF FFFFFFFF
ERR 16: 00000000 FFFFFFFF FFFFFFFF
ERR 17: 00000000 FFFFFFFF FFFFFFFF
ERR 18: 00000000 FFFFFFFF FFFFFFFF
ERR 19: 00000000 FFFFFFFF FFFFFFFF
===================================

Looks like RSX? It was sold as broken so maybe it I just had luck it booted up before the delid just fine??

Here is some photos of the delid: https://imgur.com/a/jn8vDIj
Click to expand...
That's the intermittent nature of thermomechanical stress on BGA and Bump joints. One moment they can appear to be working. Such as after changing paste, reseating IHS, using hairdrier/heatgun or other methods of heating (that don't actually achieve a reflow), etc. These precarious connections can be physically held together for a while, until that pent up strain relaxes and the console reverts to it's non-functions state (YLOD).
 
Kalulo24 said:
I was able to get my PS3 to power on a couple times, each time it the light would stay green with the fan speed at max. I got another error code this time along with the previous two that I've mentioned: 00000000 A0902120 FFFFFFFF

This is the error log:
ERR 00: 00000000 A06114FF FFFFFFFF
ERR 01: 00000000 A0611701 FFFFFFFF
ERR 02: 00000000 A06114FF FFFFFFFF
ERR 03: 00000000 A0611701 FFFFFFFF
ERR 04: 00000000 A06114FF FFFFFFFF
ERR 05: 00000000 A0611701 FFFFFFFF
ERR 06: 00000000 A06114FF FFFFFFFF
ERR 07: 00000000 A0611701 FFFFFFFF
ERR 08: 00000000 A06114FF FFFFFFFF
ERR 09: 00000000 A0611701 FFFFFFFF
ERR 10: 00000000 A06114FF FFFFFFFF
ERR 11: 00000000 A0611701 FFFFFFFF
ERR 12: 00000000 A06114FF FFFFFFFF
ERR 13: 00000000 A0611701 FFFFFFFF
ERR 14: 00000000 A0902120 FFFFFFFF
ERR 15: 00000000 A06114FF FFFFFFFF
ERR 16: 00000000 A0611701 FFFFFFFF
ERR 17: 00000000 A06114FF FFFFFFFF
ERR 18: 00000000 A0611701 FFFFFFFF
ERR 19: 00000000 A0902120 FFFFFFFF
Click to expand...

TCLL said:
OK so I got another console which stopped working after a very clean delid. I was extremely careful as always. So I soldered my TTL and read the syscon log:

===================================
ERR 00: 00000000 A0403034 FFFFFFFF
ERR 01: 00000000 A0404411 FFFFFFFF
ERR 02: 00000000 A0403034 FFFFFFFF
ERR 03: 00000000 A0404411 FFFFFFFF
ERR 04: 00000000 A0403034 FFFFFFFF
ERR 05: 00000000 A0404411 FFFFFFFF
ERR 06: 00000000 A0403034 FFFFFFFF
ERR 07: 00000000 A0404411 FFFFFFFF
ERR 08: 00000000 A0403034 FFFFFFFF
ERR 09: 00000000 A0404411 FFFFFFFF
ERR 10: 00000000 A0802124 0B49F064
ERR 11: 00000000 A0801002 0B49F064
ERR 12: 00000000 A0802030 FFFFFFFF
ERR 13: 00000000 A0802130 FFFFFFFF
ERR 14: 00000000 A0802131 186B29C4
ERR 15: 00000000 FFFFFFFF FFFFFFFF
ERR 16: 00000000 FFFFFFFF FFFFFFFF
ERR 17: 00000000 FFFFFFFF FFFFFFFF
ERR 18: 00000000 FFFFFFFF FFFFFFFF
ERR 19: 00000000 FFFFFFFF FFFFFFFF
===================================

Looks like RSX? It was sold as broken so maybe it I just had luck it booted up before the delid just fine??

Here is some photos of the delid: https://imgur.com/a/jn8vDIj
Click to expand...

daday said:
Hi all;
I have buy some motherboards JTP-001, and I have not the historical problem of these motherboards. I've checked the syscon error then the codes of of them are looks below. What does means? Thanks
ERR 00: 00000000 FFFFFFFF FFFFFFFF
ERR 01: 00000000 FFFFFFFF FFFFFFFF
ERR 02: 00000000 FFFFFFFF FFFFFFFF
ERR 03: 00000000 FFFFFFFF FFFFFFFF
ERR 04: 00000000 FFFFFFFF FFFFFFFF
ERR 05: 00000000 FFFFFFFF FFFFFFFF
ERR 06: 00000000 FFFFFFFF FFFFFFFF
ERR 07: 00000000 FFFFFFFF FFFFFFFF
ERR 08: 00000000 FFFFFFFF FFFFFFFF
ERR 09: 00000000 FFFFFFFF FFFFFFFF
ERR 10: 00000000 FFFFFFFF FFFFFFFF
Click to expand...

miguelfcp said:
Hi everyone!

Every time I do a command I get this code: "F0000003". If turn on my Ps3 it automatically turns off (red light). What I need to do so I can get the codes when PS3 had YLOD?

The command "EEP GET 3961 01" returns "00000000 FF"

Thanks!
Click to expand...

gillangit said:
hallo everyone,
im get errolog
> errlog
00000000
# CODE CLOCK
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF
# A0093004 FFFFFFFF

i follow tutorial to change all cap in rsx like this:
ZcSZsNY.jpeg

note: this image not my settings.

my setting is
in back side :
3x 2.5 470uf
3x 2.5 470uf

in front side :
2x 2.5 470uf
2x 2.5 470uf

when i tested error is same in A0093004, please help me
Click to expand...

Everyone, please provide the Model number of the console and motherboards revision when you post your logs. It's a necessary piece of information we need to decide how to further help you. For example there are differences in the RSX, presence or absence of NEC/TOKINs, NAND or NOR, and SYSCON types. I also need it for my statistics when I scrape the forums looking for errorcodes and correlating them to their respective models of console.

Other notes are also quite helpful. Like was it dusty, warranty seal intact or broken, liquid damage, drop damage, Prefious rework and what they might have done. Provide pictures.

Lastly when retieving error codes. They are in hexidecimal. ERRLOG GET 00-09, then ERRLOG GET 0A-0F, then ERRLOG GET 10-19, Finally ERRLOG GET 1A-1F. That's 32 errors total. Not 20.

Lastly, it's advisable to use the search function of the forum. Put in your error code and it'll return many others that have had it before you. You'll very likely find that your question has already been answered many times before. This is a long thread after all. If you want to read some particularly helpful posts follow the link in my signature. There you find many resources, tutorials and FAQs.

Best of luck and happy sleuthing.
 
Last edited:
RIP-Felix said:
A0003001 is probably because you attempted a bringup with out the 12v from the PSU not connected.

PWR ON State CPU Overheat (A0801200) and Shutdown State SB Error (A0902203). CPU overheated and SYSCON couldn't complete shutdown sequence gracefully, forced PWR off to protect the console, and the SB complained about it. Solution = Repaste. If that doesn't work, then delid IHS. The TIM there is done. And before you ask incredulously. Yes, the CPU can overheat within a few seconds even if the IHS is attached and the paste is gone. I've seen in many times..
Click to expand...

Yes the 3001 error was because of PSU not being connected on the 2 prongs.

I tried the repaste, the fan didn't sound so loud for sure but returned the same errors.

The thing is I tried to delid before. The console was working but CPU was getting too hot only on idle (like more than 70ºC). Tried to delid and gave with heat gun to help the process but unfortunately I couldn't do anything. When I tried to boot up again got ylod. At first I thought that I destroyed any line on the circuit during the process and that's why I tried to debug the syscon too se what is happening. But the errors say other thing.
 
RIP-Felix said:
A0093004 = VDDC RSX Failure. NEC/TOKINs & related SMDs. Given that you have been working on that area the most likely reason is a short, or the bridge wires are not connected properly. Double check soldering, polarity, and confirm the RSX is not short to GND. Anything less than 1 ohms is bad. Can't use buzzer! It'll always buzz on low impedance lines like this. Measure ohms to ground. Short leads to know what your meter reads Short as. Compensate in your recorded measurment. Healthy range is 2 - 4 ohms.
Click to expand...
Thank You.
I was flattered by Felix's reply.

ok before I start all this I want to tell you a little, I have a high curiosity and want to learn about something, actually I have a PS4 that runs normally, and also a PC with specifications that can run the PS3 emulator (RPCS3) well, However, after I watched several YouTube videos including your YouTube video Felix, I was interested in learning a lot about this mechanism, I have some programming skills, microcontrollers, etc. So I deliberately bought a PS3 YLOD to use as study material. So whether it's good or bad, please guide me. can you tell me where can i learn to measure short gnd to rsx? I still don't understand the dotted value of 1ohm and whether it's in the 2 - 4 ohm range.
 
Last edited:
Kalulo24 said:
I was able to get my PS3 to power on a couple times, each time it the light would stay green with the fan speed at max. I got another error code this time along with the previous two that I've mentioned: 00000000 A0902120 FFFFFFFF

This is the error log:
ERR 00: 00000000 A06114FF FFFFFFFF
ERR 01: 00000000 A0611701 FFFFFFFF
ERR 02: 00000000 A06114FF FFFFFFFF
ERR 03: 00000000 A0611701 FFFFFFFF
ERR 04: 00000000 A06114FF FFFFFFFF
ERR 05: 00000000 A0611701 FFFFFFFF
ERR 06: 00000000 A06114FF FFFFFFFF
ERR 07: 00000000 A0611701 FFFFFFFF
ERR 08: 00000000 A06114FF FFFFFFFF
ERR 09: 00000000 A0611701 FFFFFFFF
ERR 10: 00000000 A06114FF FFFFFFFF
ERR 11: 00000000 A0611701 FFFFFFFF
ERR 12: 00000000 A06114FF FFFFFFFF
ERR 13: 00000000 A0611701 FFFFFFFF
ERR 14: 00000000 A0902120 FFFFFFFF
ERR 15: 00000000 A06114FF FFFFFFFF
ERR 16: 00000000 A0611701 FFFFFFFF
ERR 17: 00000000 A06114FF FFFFFFFF
ERR 18: 00000000 A0611701 FFFFFFFF
ERR 19: 00000000 A0902120 FFFFFFFF
Click to expand...

Model Number: CECHK01
Board Revision: DIA-002

I've opened up the console before to replace the Blu-ray drive, that was about 2 years ago.
Console was manufactured July 2008.
Console has never been de-lided, reballed, or reflowed.
No CFW.
 
TCLL said:
Greetings everyone! I'm a newcomer to this community and I'm eager to discover some insights. Lately, I've delved into the world of acquiring broken PS3 consoles for the sheer enjoyment of fixing them up. I've successfully managed to restore functionality to most of them. However, there's this one particular console that's proving to be quite the puzzle.

I've extracted the syscon log twice, partly because I wanted to ensure accuracy after my initial attempt. Despite reaching out for help on Reddit, I unfortunately didn't receive the level of assistance I was hoping for. On the platform, I was advised that my Cell might have met its demise, and the only logical step would be to discard the entire board. Of course, I'm not entirely convinced yet, I'm seeking further opinions and expertise. Your insights would be immensely appreciated!

Result 1:
===================================
ERR 00: 00000000 A0901001 FFFFFFFF
ERR 01: 00000000 A0801701 FFFFFFFF
ERR 02: 00000000 A0801701 FFFFFFFF
ERR 03: 00000000 A0801701 FFFFFFFF
ERR 04: 00000000 A0801701 FFFFFFFF
ERR 05: 00000000 A0801701 FFFFFFFF
ERR 06: 00000000 A0801701 FFFFFFFF
ERR 07: 00000000 A0901001 FFFFFFFF
ERR 08: 00000000 A0801701 FFFFFFFF
ERR 09: 00000000 A0801701 FFFFFFFF
ERR 10: 00000000 A0801701 FFFFFFFF
ERR 11: 00000000 A0801701 FFFFFFFF
ERR 12: 00000000 A0801701 FFFFFFFF
ERR 13: 00000000 A0901001 FFFFFFFF
ERR 14: 00000000 A0801701 FFFFFFFF
ERR 15: 00000000 A0801701 FFFFFFFF
ERR 16: 00000000 A0801701 FFFFFFFF
ERR 17: 00000000 A0801701 FFFFFFFF
ERR 18: 00000000 A0801701 FFFFFFFF
ERR 19: 00000000 A0801701 FFFFFFFF
===================================

Result 2:
===================================
ERR 00: 00000000 A0901001 2C7F7875
ERR 01: 00000000 A0801701 2C7F7875
ERR 02: 00000000 A0901001 2C7F781D
ERR 03: 00000000 A0801701 2C7F781D
ERR 04: 00000000 A0901001 2C7F774D
ERR 05: 00000000 A0801701 2C7F774D
ERR 06: 00000000 A0801701 2C7F7735
ERR 07: 00000000 A0801701 2C7A82F8
ERR 08: 00000000 A0901001 2C7A82E7
ERR 09: 00000000 A0801701 2C7A82E7
ERR 10: 00000000 A0801701 2C4EAE65
ERR 11: 00000000 A0801701 2C4EAE5E
ERR 12: 00000000 A0801701 2C4EAE52
ERR 13: 00000000 A0801701 2C4EAE44
ERR 14: 00000000 A0801701 2C4EAC2E
ERR 15: 00000000 A0801701 2C4EAC24
ERR 16: 00000000 A0801701 2C4EAC0B
ERR 17: 00000000 A0801601 2C4EABC2
ERR 18: 00000000 A0801701 2C4EABC2
ERR 19: 00000000 A0801001 2B115D27
===================================
Click to expand...


Quick update on the situation:

I tried a reflow, and the console did boot up, but it was far from stable during Gran Turismo 6 sessions. With a 4.90 overclocked CFW, I managed to get about 3 minutes of gameplay before the console abruptly shut down and beeped.
Thinking that maybe my console couldn't handle the overclock, I switched back to the stock firmware. However, even with the stock firmware, GT 6 would only run for about 10 minutes before facing the same shutdown issue.
After looking into the sysconlog, I noticed many 1001/1002 errors. So I decided to take action and replaced the capacitors, hoping that this might finally resolve the problem.

Here is the result: https://imgur.com/a/GNDJSHW

Console works now without issues and I can play GT6 with oc CFW :)


Does that suggest the NEC Tokins were always broken but regained some capacitance after heat was applied and that's why my console worked just enough?
 
Last edited:
TCLL said:
Quick update on the situation:

I tried a reflow, and the console did boot up, but it was far from stable during Gran Turismo 6 sessions. With a 4.90 overclocked CFW, I managed to get about 3 minutes of gameplay before the console abruptly shut down and beeped.
Thinking that maybe my console couldn't handle the overclock, I switched back to the stock firmware. However, even with the stock firmware, GT 6 would only run for about 10 minutes before facing the same shutdown issue.
After looking into the sysconlog, I noticed many 1001/1002 errors. So I decided to take action and replaced the capacitors, hoping that this might finally resolve the problem.

Here is the result: https://imgur.com/a/GNDJSHW

Console works now without issues and I can play GT6 with oc CFW :)


Does that suggest the NEC Tokins were always broken but regained some capacitance after heat was applied and that's why my console worked just enough?
Click to expand...
Send motherboard revision, always.

Anyway, you scared me with that first image, for a second i thought you only soldered one capacitor.

1701 and 1601 issues relate to issues with the CELL-BE. Thats where Reddit was correct
1601 is a LiveLock condition (Livelock is similar to deadlock but the state of the process constantly changes, however this changing does not resolve the deadlock.) The reason for this LiveLock is most likely the RSX failing due to a BGA/Bump failure and the CELL livelocking due to this.

1701 is a condition that is nearly always the result of a CELL error such as in this case 1601 and does not indicate any specific hardware failure (apart from the hard drive maybe?)

1001 is most likely in this case from you turning the console on and off to read the SYSCON and can probably be safely ignored.

1002 probably signifies bad NEC/Tokin capacitors

Here's how i see it play out:
RSX dies due to bad BGA/Bumps. CELL LiveLocks and throws 1601 which causes 1701 to be emitted. SYSCON reads this (Specifics on the Dev wiki) and shuts down the console immediately.
I would imagine that your replacement of the NEC/Tokin capacitors fixed one part of the issue (1002 error code) and due to the application of heat by the reflow this caused the RSX's BGA/Bumps to reseat. Unless the 1002 error indicated something else apart from the NEC/Tokins, or it somehow was a false alarm

To find out if that NEC/Tokin fixed worked. Come back here if it fails, Especially if you get a 3034/4xxx error code in the SYSCON.
If it does fail then chances are you are gonna have to "frankenstien" a 65nm/40nm RSX or CELL-BE (depending on the SYSCON code) on the motherboard
 
Last edited:
Kalulo24 said:
Model Number: CECHK01
Board Revision: DIA-002

I've opened up the console before to replace the Blu-ray drive, that was about 2 years ago.
Console was manufactured July 2008.
Console has never been de-lided, reballed, or reflowed.
No CFW.
Click to expand...
Similar issue to @TCLL i would imagine.
You are getting a GLOD with 1701 issues (BE ATTENTION) and 14FF issues (CHECK STOP). What this means in theory is that you have a RSX with solder ball that hasn't quite broken completely but it cracked enough that it doesn't pass through a good enough signal.
The CECHK model is old enough to likely have a 65nm RSX before the midlife revision and thus may have symptoms of "BumpGate"

Were you playing your console when this happened or did you just come back to it later, turning it on and finding that it YLODed after you turned it off?
 
Tomatarto said:
Send motherboard revision, always.

Anyway, you scared me with that first image, for a second i thought you only soldered one capacitor.

1701 and 1601 issues relate to issues with the CELL-BE. Thats where Reddit was correct
1601 is a LiveLock condition (Livelock is similar to deadlock but the state of the process constantly changes, however this changing does not resolve the deadlock.) The reason for this LiveLock i would imagine would be most likely the RSX failing due to a BGA/Bump failure. We can imagine person 1 being the CELL-BE and person 2 being the RSX. It could also just be the CELL failing, they are not invincible after all!

1701 is a condition that is nearly always the result of a CELL error such as in this case 1601 and does not indicate any specific hardware failure (apart from the hard drive maybe?)

1001 is most likely in this case from you turning the console on and off to read the SYSCON and can probably be safely ignored.

1002 probably signifies bad NEC/Tokin capacitors

Here's how i see it play out:
RSX dies due to bad BGA/Bumps. CELL LiveLocks and throws 1601 which causes 1701 to be emitted. SYSCON reads this (Specifics on the Dev wiki) and shuts down the console immediately.
I would imagine that your replacement of the NEC/Tokin capacitors fixed one part of the issue (1002 error code) and due to the application of heat this caused the RSX's BGA/Bumps to reseat.

To find out if that NEC/Tokin fixed worked. Come back here if it fails, Especially if you get a 3034/4xxx error code in the SYSCON.
If it does fail then chances are you are gonna have to "frankenstien" a 65nm/40nm RSX or CELL-BE (depending on the SYSCON code) on the motherboard
Click to expand...


Thanks for the explanation! The strange part is after the reflow (before I replaced the caps) I was able to play Crysis 3 and Tomb Raider without any issues and the 1601/1701 errors completely disappeared. Gran Turismo 6 (heavier game probably) would make my console crash with 1001/1002 and even faster with the OC firmware. I was really hoping it's just the caps or RSX, a Cell issue would be really bad. I will keep the console for now and see what happens, everything seems to be fine currently.
 
RIP-Felix said:
mb_ohm_test_points-jpg.36295

VDDC (core voltage) should read in the 2-4 ohms range. Your 3004 is saying RSX_VDDC is the issue. Likely short.
Click to expand...
Ok i will try after holyday.
Type: CECHL
Board: Ver-001

Another questions. If i remove rsx from the mainboard. What the syscon error showup?
Thanks my bro.


Sent from my SM-A720F using Tapatalk
 
Tomatarto said:
Similar issue to @TCLL i would imagine.
You are getting a GLOD with 1701 issues (BE ATTENTION) and 14FF issues (CHECK STOP). What this means in theory is that you have a RSX with solder ball that hasn't quite broken completely but it cracked enough that it doesn't pass through a good enough signal.
The CECHK model is old enough to likely have a 65nm RSX before the midlife revision and thus may have symptoms of "BumpGate"

Were you playing your console when this happened or did you just come back to it later, turning it on and finding that it YLODed after you turned it off?
Click to expand...

It got YLOD while uploading save data to online storage in the middle of the night
 
TCLL said:
Thanks for the explanation! The strange part is after the reflow (before I replaced the caps) I was able to play Crysis 3 and Tomb Raider without any issues and the 1601/1701 errors completely disappeared. Gran Turismo 6 (heavier game probably) would make my console crash with 1001/1002 and even faster with the OC firmware. I was really hoping it's just the caps or RSX, a Cell issue would be really bad. I will keep the console for now and see what happens, everything seems to be fine currently.
Click to expand...
You might have to ask RIP-Felix as he has more knowledge on this subject than i do

My explanation is:

The CELL has not failed, the error codes it produces are due to the BGA/Bumps on the RSX failing.

1002 errors do sometimes have a tendency to YLOD during intense games such as Gran Turismo 6 which is evidence of capacitors that are failing due to wear and tear caused by heat or old age. these 1002 errors typically go hand in hand with 3004 errors

The solder that is used for the space between the die and the interposer is called the bumps. The solder that makes up the bumps is lead free and thus has a melting temperature of 350 degrees or higher. The solder in between the motherboard and the interposer is called the BGA (ball grid array) and uses high lead solder which does have a lower melting temperature.
When you heat gunned the board you reflowed the BGA which was the area in which the crack happened. You got lucky as that was where the crack did happen. if it happened in the bumps then you most certainly could not get it high enough to melt it, and if you did you would damage the chip.

It seems like the caps did go bad and replacing them was a good idea which would resolve the 1002 errors

At the end of the day if it works then it works and keep playing and come back here if it fails and certainly give us the SYSCON dump. ESPECIALLY if it has 3034 and 4xxx codes.
 
Last edited:
TCLL said:
Does that suggest the NEC Tokins were always broken but regained some capacitance after heat was applied and that's why my console worked just enough?
Click to expand...
No. It changes the impeedance equation and factors into some quite techinal math. But not capacitance.
 
Tomatarto said:
Send motherboard revision, always.

Anyway, you scared me with that first image, for a second i thought you only soldered one capacitor.

1701 and 1601 issues relate to issues with the CELL-BE. Thats where Reddit was correct
1601 is a LiveLock condition (Livelock is similar to deadlock but the state of the process constantly changes, however this changing does not resolve the deadlock.) The reason for this LiveLock is most likely the RSX failing due to a BGA/Bump failure and the CELL livelocking due to this.

1701 is a condition that is nearly always the result of a CELL error such as in this case 1601 and does not indicate any specific hardware failure (apart from the hard drive maybe?)

1001 is most likely in this case from you turning the console on and off to read the SYSCON and can probably be safely ignored.

1002 probably signifies bad NEC/Tokin capacitors

Here's how i see it play out:
RSX dies due to bad BGA/Bumps. CELL LiveLocks and throws 1601 which causes 1701 to be emitted. SYSCON reads this (Specifics on the Dev wiki) and shuts down the console immediately.
I would imagine that your replacement of the NEC/Tokin capacitors fixed one part of the issue (1002 error code) and due to the application of heat by the reflow this caused the RSX's BGA/Bumps to reseat. Unless the 1002 error indicated something else apart from the NEC/Tokins, or it somehow was a false alarm

To find out if that NEC/Tokin fixed worked. Come back here if it fails, Especially if you get a 3034/4xxx error code in the SYSCON.
If it does fail then chances are you are gonna have to "frankenstien" a 65nm/40nm RSX or CELL-BE (depending on the SYSCON code) on the motherboard
Click to expand...
+1

Well said.
 
The Heat applied will cause a false repair. So be prepared for it to fail again (nand backup, saves, erk).

The impeedance equation changing can extend a teetering Bump defect. Less voltage ripple can improve it's stability in this precarious state. For awhile before it declines again and dies. Hair dries do the same thing. Can last for months, then weeks on 2nd attempt, the days, and finally no work anymore.
 
Tomatarto said:
Similar issue to @TCLL
CECHK model is old enough to likely have a 65nm RSX before the midlife revision and thus may have symptoms of "BumpGate"
Click to expand...
I have since confirmed that all 65nm revision have fixed underfill. While that doesn't mean they are free from bumpgate materials (namely a lack of the polyamide stress layer, which I cant confirm exists or not), I can say there is not sufficient evidence of an increased RSX failure rate among 65nm models. Which I interpret to mean the decreased TDP of the 65nm RSX, but same load (ps3 games), and addition of more durable underfill is adequate to not exploit the defect even it it exists.

IMO, the 65nm is in the clear.
 
TCLL said:
Thanks for the explanation! The strange part is after the reflow (before I replaced the caps) I was able to play Crysis 3 and Tomb Raider without any issues and the 1601/1701 errors completely disappeared. Gran Turismo 6 (heavier game probably) would make my console crash with 1001/1002 and even faster with the OC firmware. I was really hoping it's just the caps or RSX, a Cell issue would be really bad. I will keep the console for now and see what happens, everything seems to be fine currently.
Click to expand...
1601/1701 can occure from solder, but also from normal software crashes. Even Bluray drive and several non RSX/CELL related ICs. They do not always mean a 3034 is coming. They just often preceed one.

The 1001/1002 in game behavior you describe and correctly diagnosed is the classic NEC/Tokin behavior. So that is likely one issue solved. Now I would have done the caps differently, but to each their own.

Just keep an eye on it. Keep saves and stuff backed up and hope for the best. 1st diagnostic and repair step was done correctly. So we wait now to see if it is going to last. GL and good work!
 
You must log in or register to reply here.

Similar threads

Back
Top