PS3 CECHA00 with several SYSCON errors (3004, 1001,1002, 2120,3011)

[LSL]

Hello! I've been on hiatus from PS3 maintenance for almost 4 and a half years. During that period, due to the Covid19 lockdown, no imports and components running out, I took a break and am now coming back. I have a batch of BC PS3, most of them CECHA, two CECHB and two CECHE. I bought all of them second-hand and they are defective.

The PS3 in this topic was purchased in 2015 with a YLOD of approximately 2.5 seconds. My mistake was not documenting what I did in previous years, so I probably tried a reflow to see if it changed any behavior, since nothing changed, I left it saved until access to SYSCON was released.

Due to SYSCON errors, the device began to behave differently. With the inclusion of tantalum capacitors, when the PS3 was turned on, the fan would spin at maximum speed and about 5 seconds later it would turn off and display error codes 1001, 1002, 2120, 3001. It was at this time that I removed 2 NEC-Tokin devices from the RSX because one of them was deformed, and I also noticed that the HDMI sector had a short circuit with GND. This was when I stopped due to the lockdown.

Nowadays, from January 3, 2025, I analyzed this CECHA again and due to the hiatus I had to review and relearn how to access SYSCON and the surprise was that now it only had a single error, 3004, and even though I was one of the first to talk about this error (https://www.psx-place.com/threads/f...-and-error-reporting.30100/page-6#post-249825) I didn't know how to proceed because now I had a real error and at that time it was an error caused for studies.

After analyzing the voltages and signals, including tantalum capacitors (I had almost 15 capacitors), I only corrected error 3004 today in the simplest way possible. The NEC-TOKIN has a positive pole on both sides, but when I replaced it with tantalums I only soldered on one side (near the coils), so after including 2 capacitors on the other side of the place where the NEC-Tokin is (near the RSX), error 3004 disappeared.

With that, I returned to the status when I was in the COVID period, the PS3 turned on again with the fan at maximum and turned off 5 seconds later, but now with errors 1001, 1002, 2120 and sometimes 3011. Errors 1001 and 1002 are easy to solve, so now I'm trying to solve error 2120.

I checked the HDMI sector again and the HDMI_I2C_SCL and HDMI_I2C_SDA signals are shorted to GND, the DDCLK and DDData signals (HDMI port) are also shorted to GND, I will replace IC2502 (SIL9132CBU-TR) soon.

Now one point intrigues me, I saw stains on the RSX's DIE and CELL, could they just be oxidation stains caused by some thermal paste or is it an indication of a worse problem? I took measurements on the RSX in relation to Ohm and voltages and everything is ok.

 

[nickxab]

this RSX looks fried.

 

[LSL]

this RSX looks fried.

I hope not. I measured resistance (https://www.psx-place.com/threads/ps3-reball-cell-rsx.33388/#post-287587) and voltages on the RSX and they are all OK.

Errors 3004 and 1002 have not appeared since I installed the tantalums.

After I isolated the HDMI_I2C_SCL and HDMI_I2C_SDA signals, which is a communication between the HDMI IC and SYSCON, the PS3 no longer turns the fan in the last stage and soon returns error 2120.

I will have to change the HDMI IC to see if the PS3 returns more errors.

I think that if the RSX is fried It would have thermal errors, such as 1201 or 2031.

 

[birigui]

Good afternoon friend, finally someone with errors 3004.

I would like some help if possible, I have a PS3 like this with errors 3004, all voltages and resistances on the board are ok.

What you mentioned years ago is responsible for controlling the buck converters that generally supply the RSX's 1.2v voltage, well, I analyzed it for 2 weeks, comparing it with a Frank PS3 that I left functional, and I couldn't get to a conclusion about what was happening, causing the defect.

A big question I have is whether the nectokins were responsible for the error? Because when analyzing the vrm sector, the nectokins receive the voltage generated by the buck converters and feed the rsx, how can they be the ones causing the 3004, if the voltage does not reach them?

I confess that I have not yet tried to install tantalums in place of the necs, as in any case they do not seem to be the cause of the defect. the 1-2-3-32-31-30 pins that receive voltages, as you said a few years ago, need to be at the correct high and low logic levels, I saw that the syscon is the one sending these signals, but where does the syscon get it from? this information? to pass on?

 

[LSL]

Yes, NEC-Tokins cause error 3004. When replacing with tantalums, distribute the capacitors equally on the positive poles of the NEC. As I said above, I had installed the tantalums only on the side closest to the FETs and I still got error 3004. When I distributed them equally, the error stopped.

The IC6201 (NCP5318FTR2G) expects to receive the VID voltage defined by SYSCON. This voltage is precisely the signals that arrive on pins 3, 2, 1, 32, 31, 30. If the IC6201 does not receive the correct voltage, it does not activate pin 7 (PWRGD) at a high level (3.3v).

I confess that I did not analyze pin 7 when it is defined whether the error is 1002 or 3004. It may be that the IC6201 has different behaviors on pin 7 that indicate when it is an error 1002 or 3004. Who knows, for one type of error the IC6201 sets pin 7 to a high level (3.3v) and then to a low level (0v), and for another error it always leaves it at a low level (0v), an issue to be analyzed in the future. It may also be that the RSX itself monitors the voltage and communicates with SYSCON, defining whether it is 1002 or 3004.

 

[birigui]

Sim, NEC-Tokins causam erro 3004. Ao substituir por tântalos, distribua os capacitores igualmente nos polos positivos do NEC. Como eu disse acima, eu tinha instalado os tântalos apenas no lado mais próximo dos FETs e ainda obtive o erro 3004. Quando os distribuímos igualmente, o erro parou.

O IC6201 (NCP5318FTR2G) espera receber a tensão VID definida pelo SYSCON. Essa tensão é justamente os sinais que chegam nos pinos 3, 2, 1, 32, 31, 30. Se o IC6201 não receber a tensão correta, ele não ativa o pino 7 (PWRGD) em um nível alto (3,3v).

Confesso que não analisei o pino 7 quando é definido se o erro é 1002 ou 3004. Pode ser que o IC6201 tenha comportamentos diferentes no pino 7 que indique quando é um erro 1002 ou 3004. Quem sabe, para um tipo de erro o IC6201 coloca o pino 7 em nível alto (3,3v) e depois em nível baixo (0v), e para outro erro ele sempre deixa em nível baixo (0v), questão a ser comprovada futuramente. Pode ser também que o próprio RSX monitore a tensão e se comunique com o SYSCON, definindo se é 1002 ou 3004.

I have no further attempts to make other than adding the tantalums and seeing if the symptom changes, I will start adding them 1 by 1. In fact, the sequence that the IC6201 (NCP5318FTR2G) receives from the syscon is different on a working cok002 that I have. I got stuck in the analysis, because the only thing different is the feedback and powergood pin, both 0V, I analyzed the resistive divider and it's ok. But it doesn't seem to make sense for tokens to be a problem, if the voltage isn't even released by the IC6201 (NCP5318FTR2G) to reach them. But as you said, syscon defines pins 3, 2, 1, 32, 31, 30. This definition may depend on the good functioning of the tokins.

Mod edit: Provide an English translation with your post.
Sorry again for not writing in English, I left the translator activated and it translated before I posted.

 

[LSL]

Yes, the sequences change, I don't know if it's only between different models (CECHE, CECHC, CECHA, CECHB) or if even on boards of the same model this sequence can change.

Change the NEC, if the IC6201 didn't receive anything on the feedback pin then it won't activate the PWRGD.

Post the comments in English, use Google Translate

I replaced the ICIC2502, the error 2120 stopped, but now I have a single error:

err_code:0xa0213011

Has anyone else gotten this error? On the page https://www.psdevwiki.com/ps3/Syscon_Error_Codes#3011 there is no specific explanation and it only says that it is the CELL.

>$ bringup
bringup
[SSM] state: 0000 -> 0101
Bringup Mode #0 (0xFF)
[SSM] ssmCb_OnStartingBePowOn() called.
[SSM] First Boot.
[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 -> 0302
[SSM] PowSeq Fail : Detected !
[SSM] state: 0302 -> 0700
[POWSEQ] AV Backend Letup
[SSM] Shutdown mode : syspm_stat=00000000/00000000
[ERROR]: 0xa0213011
Press Ctrl+C to exit
>$bringup   
[POWSEQ] PowerSeq_Letup called.
[SSM] state: 0700 -> 0600
(PowerOff State) (Fatal)
bringup
Do nothing. (FatalOff State)

 

[birigui]

Yes, the sequences change, I don't know if it's only between different models (CECHE, CECHC, CECHA, CECHB) or if even on boards of the same model this sequence can change.

Change the NEC, if the IC6201 didn't receive anything on the feedback pin then it won't activate the PWRGD.

Post the comments in English, use Google Translate

I replaced the ICIC2502, the error 2120 stopped, but now I have a single error:

err_code:0xa0213011

Has anyone else gotten this error? On the page https://www.psdevwiki.com/ps3/Syscon_Error_Codes#3011 there is no specific explanation and it only says that it is the CELL.

>$ bringup
bringup
[SSM] state: 0000 -> 0101
Bringup Mode #0 (0xFF)
[SSM] ssmCb_OnStartingBePowOn() called.
[SSM] First Boot.
[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 -> 0302
[SSM] PowSeq Fail : Detected !
[SSM] state: 0302 -> 0700
[POWSEQ] AV Backend Letup
[SSM] Shutdown mode : syspm_stat=00000000/00000000
[ERROR]: 0xa0213011
Press Ctrl+C to exit
>$bringup  
[POWSEQ] PowerSeq_Letup called.
[SSM] state: 0700 -> 0600
(PowerOff State) (Fatal)
bringup
Do nothing. (FatalOff State)

I've never seen this error, not even in my research... perhaps a BGA communication failure between cell and rsx?