PS3 (Research/Experimental) - NEC/TOKIN Capacitors Replacement - YLOD

RIP-Felix said:
If temps are under control without needing high fan%, then leave it alone. There is significant risk of damage from delidding. If you don't need to, then don't.

I believe the IHS acts as a stiffener to keep the RSX package coplaner and reduce stress on the BGA around the edge of the chip (where it's highest). If you need to delid I would reccomend reattaching the IHS with appropriate adhesives.

PS: stay away from ali express caps.
Click to expand...

As RIP-Felix has stated, adding adhesive to the ihs keeps the structure of the chip and an absorber of the heat - if anyone is interested I use this stuff https://www.aliexpress.com/item/1005004894286583.html - works a treat and is cheap to buy.

In terms of deliding, its only really needed if there is no heating coming out from the back of the ps3 or readings are getting over 75C.

70C and below is perfectly fine, especially for the 90nm chips - trying to achieve 65C or below is very hard for those type of size chips.
 
gmipf said:
@RIP-Felix I have ordered at jlcpcb now and they said this in the email:

Code: 
Hi Sir or madam,

Well got your order with many thanks~
Sorry to bother you, but there is an issue that we want to confirm with you before proceeding.
As shown below, do you need to do the edge plated process?
Sorry the edge plated process we can only do ENIG, can you accept it?
Could you kindly point out which area need to do edge plated process?

ENIG would be more expensive right? Does edge plated mean the copper goes down the edge? I just want as cheap as possible. Is without the edge plating still good? I can still use wires if soldering from edge isn't possible.

And pcbway is keep changing my order after the review:
Code: 
 PCBWay:Sides with castellated holes['0' was changed to '4'];
The prices goes up from 5$ to 27.26$ (50pcs)

Does your board has any holes on the edge?

EDIT: Oh ok, castellated holes mean those 4 half holes on the edge.

Can I proceed at jlcpcb with ENIG and without the castellated holes? Or better without ENIG and but the half edge holes option? Both options together are too expensive I think.
Click to expand...
They are complaining.

These are not castillated edges they are slots. They support slots without charging extra for edge plating or castellation. I did it this way to avoid those costs. So no, you cannot proceed without the "edge plating" as they are calling it. If you do, it will not have copper to wick the solder up to the top layer. It will be a massive PITA to work with and defeat the purpose.

The solution is to move the edge cuts outside the slot area and then they STFU, becaus you will be paying for the added area. That's the real issue here. They don't like that I placed the edge cust through the center of the slot. I did that so you wouldn't have to snip them out, and it saves on PCB real estate, costing less to order. The don't like this because they can't butt another persons PCB right up against mine. And since you've already ordered the PCB, they can't charge you extra for doing what they want to do. Which is to move the edge cuts outside the slots. It adds cost and they would loose profits.

So they make up a BS excuse and say pay us more for enig and edge plating of castellated edges.

OSH park had issues for a while. Recently they've been making them fine. I didn't have a complaint from JLCpcb or PCBway when other had used them, then I recomeneded people go with thenm instead because they didn't complain. Now they are too. I mean, FFS! The inconsistancy of these places is annoying. Sometime they don't say a word and it shows up perfect, no problem. Other times they say stuff like this. I don't get why they can't figure it out.

I have however addressed this in v0.7a, but am not ready to release it yet. I have been working on other projects lately. Too many projects to juggle ATM with work and hobbies. I may update the Tantalizer v0.6b to address this specific issue however. I just havent gotten to it yet.

There is a reason this is still a beta
 
Last edited:
@RIP-Felix Thanks for the info, here is another longer conversation with jlcpcb:

Code: 
Dear customer,

Castellated holes processing is 37.6USD.


So total should be 82.7USD.


Best Regards.


https://d33v4339jhl8k0.cloudfront.net/users/695412.305455.jpg
Tammy Huang
    
Web:     JLCPCB.COM 
Email:     [email protected]


How would you rate my reply?
Great    Okay    Not Good
{#HS:2197479781-845823#}
On Wed, Mar 29, 2023 at 8:21 AM HKT, <> wrote:
> What is the total cost if I add castellated holes?
>
> On Wed, Mar 29, 2023 at 8:18 AM HKT, JLCPCB Support <[email protected]> wrote:
>>
>> Dear customer,
>>
>>
>> It is for 30pcs~
>>
>>
>> Best Regards.
>>
>>
>> https://d33v4339jhl8k0.cloudfront.net/users/695412.305455.jpg
>> Tammy Huang
>>     
>> Web:     JLCPCB.COM 
>> Email:     [email protected]
>> On Wed, Mar 29, 2023 at 8:05 AM HKT, <> wrote:
>>> Is 45.1USD additional or total cost for 30 pcbs?
>>>
>>> On Wed, Mar 29, 2023 at 7:54 AM HKT, JLCPCB Support <[email protected]> wrote:
>>>>
>>>> Dear customer,
>>>>
>>>>
>>>> Edge plating is not available for HASL.
>>>>
>>>> The surface should be ENIG.
>>>>
>>>>
>>>> This is a special processing not open for clients.
>>>>
>>>> We need to negotiate with our factory for it first.
>>>>
>>>>
>>>> So should we add this to remark and send you supplement payment link?
>>>>
>>>>
>>>> Best Regards.
>>>>
>>>>
>>>> https://d33v4339jhl8k0.cloudfront.net/users/695412.305455.jpg
>>>> Tammy Huang
>>>>     
>>>> Web:     JLCPCB.COM 
>>>> Email:     [email protected]
>>>> On Wed, Mar 29, 2023 at 7:20 AM HKT, <> wrote:
>>>>> I don't think I need castellated holes, how much would it cost with edge plating and the HASL(with lead) surface finish instead of ENIG? And where can I find the edge plating option on the order page? Is it the Gold Fingers option?
>>>>>
>>>>> On Wed, Mar 29, 2023 at 7:13 AM HKT, JLCPCB Support <[email protected]> wrote:
>>>>>>
>>>>>> Dear customer,
>>>>>>
>>>>>> Yes the boards in the picture is with edge plating.
>>>>>>
>>>>>> It will take about 45.1USD to proceed edge plating.
>>>>>>
>>>>>>
>>>>>> Is it acceptable for you?
>>>>>>
>>>>>>
>>>>>> How about the Castellated Holes processing?
>>>>>>
>>>>>>
>>>>>> Best Regards.
>>>>>>
>>>>>>
>>>>>> https://d33v4339jhl8k0.cloudfront.net/users/695412.305455.jpg
>>>>>> Tammy Huang
>>>>>>     
>>>>>> Web:     JLCPCB.COM 
>>>>>> Email:     [email protected]
>>>>>> On Wed, Mar 29, 2023 at 6:01 AM HKT, <> wrote:
>>>>>>> Hello,
>>>>>>>
>>>>>>> how much would it cost extra for the option with edge plating and is edge plating possible without the castellated holes option? Since the board is only 0.6mm I think it should be possible to solder without the edge plating on top of another pcb. What would you suggest?
>>>>>>>
>>>>>>> I have added two photos of another hobbyist using the same board. I wanted to ask if those boards also have plated edges and castellated holes?
>>>>>>>
>>>>>>> Best regards
>>>>>>>
>>>>>>>
>>>>>>> On Wed, Mar 29, 2023 at 5:44 AM HKT, JLCPCB Support <[email protected]> wrote:
>>>>>>>>
>>>>>>>> Hi Sir or madam,
>>>>>>>>
>>>>>>>> Well got your order with many thanks~
>>>>>>>>
>>>>>>>>
>>>>>>>> Sorry to bother you, but there is an issue that we want to confirm with you before proceeding.
>>>>>>>>
>>>>>>>>
>>>>>>>> 1/ do you need edge plating? If yes, there'll be extra cost.
>>>>>>>>
>>>>>>>> 2/ you didn't select Castellated Holes processing but there are half holes/slots in your file.
>>>>>>>>
>>>>>>>> Please kindly confirm as following options and which option you need?
>>>>>>>>
>>>>>>>> A)Proceed those half cut holes as Castellated Holes processing and pay for the extra cost and those half holes would be plated well finally.
>>>>>>>>
>>>>>>>> B)Proceed those half cut holes without Castellated Holes processing and the half holes might not be plated well, the copper of those half holes maybe peeled off as below picture, but no need extra cost.
>>>>>>>>
>>>>>>>>
>>>>>>>> Your early reply will be highly appreciated!
>>>>>>>>
>>>>>>>> Thank you so much!
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> https://d33v4339jhl8k0.cloudfront.net/users/695412.305455.jpg
>>>>>>>> Tammy Huang
>>>>>>>>     
>>>>>>>> Web:     JLCPCB.COM 
>>>>>>>> Email:     [email protected]


I will wait for your new pcb version then I think.
 
Hello everyone,i have 2 consoles with the same error 232102 ,i have already replace the capacitors with tantalyzers by @RIP-Felix (btw amazing work) .The consoles are with sem-001 and DIA-002 mobo ,before i was in 1002 syscon error ,after replaced necs i fix the 1002 and now i am in 2102.Both consoles has 4.2ohm and 3.7ohm in rsx nec pads and the console with sem001 mobo has only 13 days on .Any advice? :adoration:

PS-The state code is 23 and the full code is 232102
 
Last edited:
RIP-Felix said:
They are complaining.

These are not castillated edges they are slots. They support slots without charging extra for edge plating or castellation. I did it this way to avoid those costs. So no, you cannot proceed without the "edge plating" as they are calling it. If you do, it will not have copper to wick the solder up to the top layer. It will be a massive PITA to work with and defeat the purpose.

The solution is to move the edge cuts outside the slot area and then they STFU, becaus you will be paying for the added area. That's the real issue here. They don't like that I placed the edge cust through the center of the slot. I did that so you wouldn't have to snip them out, and it saves on PCB real estate, costing less to order. The don't like this because they can't butt another persons PCB right up against mine. And since you've already ordered the PCB, they can't charge you extra for doing what they want to do. Which is to move the edge cuts outside the slots. It adds cost and they would loose profits.

So they make up a BS excuse and say pay us more for enig and edge plating of castellated edges.

OSH park had issues for a while. Recently they've been making them fine. I didn't have a complaint from JLCpcb or PCBway when other had used them, then I recomeneded people go with thenm instead because they didn't complain. Now they are too. I mean, FFS! The inconsistancy of these places is annoying. Sometime they don't say a word and it shows up perfect, no problem. Other times they say stuff like this. I don't get why they can't figure it out.

I have however addressed this in v0.7a, but am not ready to release it yet. I have been working on other projects lately. Too many projects to juggle ATM with work and hobbies. I may update the Tantalizer v0.6b to address this specific issue however. I just havent gotten to it yet.

There is a reason this is still a beta
Click to expand...

PCBway does it without problems
 
Jimilias said:
Hello everyone,i have 2 consoles with the same error 232102 ,i have already replace the capacitors with tantalyzers by @RIP-Felix (btw amazing work) .The consoles are with sem-001 and DIA-002 mobo ,before i was in 1002 syscon error ,after replaced necs i fix the 1002 and now i am in 2102.Both consoles has 4.2ohm and 3.7ohm in rsx nec pads and the console with sem001 mobo has only 13 days on .Any advice? :adoration:

PS-The state code is 23 and the full code is 232102
Click to expand...

https://www.psdevwiki.com/ps3/Syscon_Error_Codes#2102_.28RSX.29

So its possible your new caps are shorting causing this error or a short somewhere on the RSX or CELL - which can only be identfied by removing the chips themselves.
 
Hello.
Decoupling capacitor replacement has been discussed for years, but most of it seems to be about power integrity.
But omitting MLCCs from decoupling capacitors in ICs operating at hundreds of Mega/Gigahertz is horrible from an EMC point of view.

When approximating the frequency characteristics of the Proadlizer with MLCCs, a large amount is required like the back side of the RTX3090, but later models without Proadlizers have only not so many MLCCs.

Can anyone measure the capacitance for reference?
 
Hi to all, (sorry for my english)
Please help me to understand what's happening with my PS3 FAT CECHK03 (DIA-002 mobo). It disconnects/ shuts down after 20-25 minutes no meter what I do , watching movie or playing game or just seating idle in XMB , the temperature is ok around 50-67"C . I've installed PS3AdvancedToolSet and extracted syscon errors log a few times:
Firmware Version: 4.89 (build 50740)
Platform ID: CokE10
Product Code: 00 87
Product Sub Code: 00 07
Hardware Config: 4E00FFFF0107BCBF
Syscon Fimware Version: 0E69.0001000400040002 (EEPROM: 0001000400040002)
Bringup Count: 1461, Shutdown Count: 1340
Runtime: 71 Days, 18 Hours, 9 Minutes, 25 Seconds
Error Log
01: A0801601 Sat Dec 31 00:21:48 2005
02: A0801701 Sat Dec 31 00:21:48 2005
03: A0801601 Thu Jan 31 16:04:08 2013
04: A0801701 Thu Jan 31 16:04:08 2013
05: A08014FF Wed Jan 30 06:31:12 2013
06: A0801701 Wed Jan 30 06:31:12 2013
07: A0801601 Sat Jan 19 06:15:33 2013
08: A0801701 Sat Jan 19 06:15:33 2013
09: A08014FF Sat Jan 12 09:48:02 2013
10: A0801701 Sat Jan 12 09:48:02 2013
11: A08014FF Mon Dec 31 05:45:27 2012
12: A0801701 Mon Dec 31 05:45:27 2012
13: A0801601 Thu Oct 14 21:51:59 2010
14: A0801701 Thu Oct 14 21:51:59 2010
15: A08014FF Thu Apr 15 15:47:11 2010
16: A0801701 Thu Apr 15 15:47:11 2010
17: A0801701 Thu Mar 18 18:37:24 2010
18: A08014FF Thu Mar 18 18:37:24 2010
19: A0801802 Thu Mar 18 18:37:24 2010
20: A08014FF Mon Feb 1 01:37:33 2010
21: A0801701 Mon Feb 1 01:37:33 2010
22: A0801601 Sat Nov 25 06:27:06 2006
23: A0801701 Sat Nov 25 06:27:06 2006
24: A08014FF Thu Nov 9 03:43:26 2006
25: A0801701 Thu Nov 9 03:43:26 2006
26: A0801601 Sun Nov 5 05:42:27 2006
27: A0801701 Sun Nov 5 05:42:27 2006
28: A0801601 Sat Nov 4 05:23:02 2006
29: A0801701 Sat Nov 4 05:23:02 2006
30: A08014FF Thu Nov 2 01:03:44 2006
31: A0801701 Thu Nov 2 01:03:44 2006
32: FFFFFFFF Sun Oct 22 05:40:18 2006

Firmware Version: 4.89 (build 50740)
Platform ID: CokE10
Product Code: 00 87
Product Sub Code: 00 07
Hardware Config: 4E00FFFF0107BCBF
Syscon Fimware Version: 0E69.0001000400040002 (EEPROM: 0001000400040002)
Bringup Count: 1465, Shutdown Count: 1344
Runtime: 71 Days, 18 Hours, 59 Minutes, 28 Seconds
Error Log
01: A08014FF Sat Dec 31 21:31:08 2005
02: A0801701 Sat Dec 31 21:31:08 2005
03: A0801601 Sat Dec 31 00:21:48 2005
04: A0801701 Sat Dec 31 00:21:48 2005
05: A0801601 Thu Jan 31 16:04:08 2013
06: A0801701 Thu Jan 31 16:04:08 2013
07: A08014FF Wed Jan 30 06:31:12 2013
08: A0801701 Wed Jan 30 06:31:12 2013
09: A0801601 Sat Jan 19 06:15:33 2013
10: A0801701 Sat Jan 19 06:15:33 2013
11: A08014FF Sat Jan 12 09:48:02 2013
12: A0801701 Sat Jan 12 09:48:02 2013
13: A08014FF Mon Dec 31 05:45:27 2012
14: A0801701 Mon Dec 31 05:45:27 2012
15: A0801601 Thu Oct 14 21:51:59 2010
16: A0801701 Thu Oct 14 21:51:59 2010
17: A08014FF Thu Apr 15 15:47:11 2010
18: A0801701 Thu Apr 15 15:47:11 2010
19: A0801701 Thu Mar 18 18:37:24 2010
20: A08014FF Thu Mar 18 18:37:24 2010
21: A0801802 Thu Mar 18 18:37:24 2010
22: A08014FF Mon Feb 1 01:37:33 2010
23: A0801701 Mon Feb 1 01:37:33 2010
24: A0801601 Sat Nov 25 06:27:06 2006
25: A0801701 Sat Nov 25 06:27:06 2006
26: A08014FF Thu Nov 9 03:43:26 2006
27: A0801701 Thu Nov 9 03:43:26 2006
28: A0801601 Sun Nov 5 05:42:27 2006
29: A0801701 Sun Nov 5 05:42:27 2006
30: A0801601 Sat Nov 4 05:23:02 2006
31: A0801701 Sat Nov 4 05:23:02 2006
32: FFFFFFFF Thu Nov 2 01:03:44 2006

RSX chip was reballed long time ago, CELL wasn't ever reballed.
Yesterday i changed nec/tokins with tantalum capacitors (yellows ones 477J 16pcs instead of 4pcs nec/tokins)) all from backside of mobo(not from chips side) for both cell and rsx chips but anyway my console still disconects / freezing after 20-30 minutes :(
any help will be appreciated, Thanks.
 
Ghector78 said:
Please help me to understand what's happening with my PS3 FAT CECHK03 (DIA-002 mobo). It
Click to expand...

Hi, this is my first post at this blog, have been learning for the past 12 months and I finally managed to get my hands into a non-powering-up PS3. Sharing details below (hopefully I managed to "correctly quote" the thread):

SEALED PS3 FAT CECHB01 with RLOD
Just opened and managed to get access to both SYSCON EXTERNAL and INTERNAL
EXTERNAL ERROR LOG (commands executed one by one):
> ERRLOG GET 00
00000000 A0801002 FFFFFFFF
> ERRLOG GET 01
00000000 A0801002 FFFFFFFF
> ERRLOG GET 02
00000000 A0801002 FFFFFFFF
> ERRLOG GET 03
00000000 A0101001 FFFFFFFF
> ERRLOG GET 04
00000000 A0801002 FFFFFFFF
> ERRLOG GET 05
00000000 A0801002 FFFFFFFF
> ERRLOG GET 06
00000000 A0801002 FFFFFFFF
> ERRLOG GET 07
00000000 A0801002 FFFFFFFF
> ERRLOG GET 08
00000000 A0801002 FFFFFFFF
> ERRLOG GET 09
00000000 A0801002 FFFFFFFF
> ERRLOG GET 0A
00000000 A0801002 FFFFFFFF
> ERRLOG GET 0B
00000000 A0801002 FFFFFFFF
> ERRLOG GET 0C
00000000 A0801002 FFFFFFFF
> ERRLOG GET 0D
00000000 A0801002 FFFFFFFF
> ERRLOG GET 0E
00000000 A0801002 FFFFFFFF
> ERRLOG GET 0F
00000000 A0801002 FFFFFFFF
> ERRLOG GET 10
00000000 A0801002 0B4886DD
> ERRLOG GET 11
00000000 A0801002 FFFFFFFF
> ERRLOG GET 12
00000000 A0801002 FFFFFFFF
> ERRLOG GET 13
00000000 A0801002 FFFFFFFF
> ERRLOG GET 14
00000000 A0801002 FFFFFFFF
> ERRLOG GET 15
00000000 A0801002 FFFFFFFF
> ERRLOG GET 16
00000000 A0801002 FFFFFFFF
> ERRLOG GET 17
00000000 A0801002 FFFFFFFF
> ERRLOG GET 18
00000000 A0801002 0B488698
> ERRLOG GET 19
00000000 A0801002 0B4886BF
> ERRLOG GET 1B
00000000 A0801002 0B4886DD
> ERRLOG GET 1C
00000000 A0801002 0B48875C
> ERRLOG GET 1D
00000000 A0801002 FFFFFFFF
> ERRLOG GET 1E
00000000 A0801002 FFFFFFFF
SYSCON INTERNAL COMMANDS/LOG:
> becount
becount
Bringup : 2199 times
Shutdown: 1545 times
Power-on: 119day 18hour 56min 51sec
[mullion]$
> powerstate
powerstate
ATA Power : OFF
PCI Power : OFF
RSX Power : OFF
XDR Power : OFF
Eurus Power : OFF
SB Power : OFF
RSX Thermal Sensor : UNAVAILABLE
BE Thermal Sensor : UNAVAILABLE
[mullion]$
> errlog
errlog
ofst[ 40]:err_code:0xffffffff, clock:0xffffffff
ofst[ 44]:err_code:0xa0801002, clock:0xffffffff
ofst[ 48]:err_code:0xa0801002, clock:0x0b48875c 2005/12/31 00:03:40
ofst[ 52]:err_code:0xa0801002, clock:0x0b4886dd 2005/12/31 00:01:33
ofst[ 56]:err_code:0xa0801002, clock:0x0b488694 2005/12/31 00:00:20
ofst[ 60]:err_code:0xa0801002, clock:0x0b4886bf 2005/12/31 00:01:03
ofst[ 64]:err_code:0xa0801002, clock:0x0b488698 2005/12/31 00:00:24
ofst[ 68]:err_code:0xa0801002, clock:0xffffffff
ofst[ 72]:err_code:0xa0801002, clock:0xffffffff
ofst[ 76]:err_code:0xa0801002, clock:0xffffffff
ofst[ 80]:err_code:0xa0801002, clock:0xffffffff
ofst[ 84]:err_code:0xa0801002, clock:0xffffffff
ofst[ 88]:err_code:0xa0801002, clock:0xffffffff
ofst[ 92]:err_code:0xa0801002, clock:0xffffffff
ofst[ 96]:err_code:0xa0801002, clock:0x0b4886dd 2005/12/31 00:01:33
ofst[100]:err_code:0xa0801002, clock:0xffffffff
ofst[104]:err_code:0xa0801002, clock:0xffffffff
ofst[108]:err_code:0xa0801002, clock:0xffffffff
ofst[112]:err_code:0xa0801002, clock:0xffffffff
ofst[116]:err_code:0xa0801002, clock:0xffffffff
ofst[120]:err_code:0xa0801002, clock:0xffffffff
ofst[124]:err_code:0xa0801002, clock:0xffffffff
ofst[ 0]:err_code:0xa0801002, clock:0xffffffff
ofst[ 4]:err_code:0xa0801002, clock:0xffffffff
ofst[ 8]:err_code:0xa0801002, clock:0xffffffff
ofst[ 12]:err_code:0xa0801002, clock:0xffffffff
ofst[ 16]:err_code:0xa0801002, clock:0xffffffff
ofst[ 20]:err_code:0xa0101001, clock:0xffffffff
ofst[ 24]:err_code:0xa0801002, clock:0xffffffff
ofst[ 28]:err_code:0xa0801002, clock:0xffffffff
ofst[ 32]:err_code:0xa0801002, clock:0xffffffff
ofst[ 36]:err_code:0xa0801002, clock:0xffffffff
[mullion]$

Based on the Syscon Error Codes link, 1002 is related to 1002 (Power RSX)

Components Involved:
RSX (IC2001 on COK-001)
NEC/TOKIN Proadlizers (C6229/C6230/C6231/C6232 on COK-001)
Other nearby components of the power block

Can I get any deeper to confirm that this issue is related to the NEC TOKIN only? at the end of the code error description states that 1002 is the fingerprint of bad tokins.

I have some of those handy and I really want to bring it back to life, should I move forward and replace the NEC TONKINS?
 
Hello. I want to share my experience on replacing Nec tokin with PS3 FAT CECHL00
Syscon code 0801002 0802124
Replacement Token Cell-10x470uf4v RSX-10x470uf4v
The errors have disappeared, everything works smoothly.
I would like to note that the replacement should be carried out with capacitors of the same nominal value, within (+-~5000uf) on each of the processors.
Cell and RSX are very sensitive to the capacity of tantalum capacitors, initially I set different ratings and did not calculate the total capacity very accurately (to be more precise, it was overestimated), as a result I received errors 0801002 0802124 and shutdown when starting the gameplay.
As soon as everything is soldered to Cell-10x470uf4v RSX-10x470uf4v, it works like clockwork.
Be precise and attentive.
 
Hello everyone.....I would like to know for sure, from this error log, that the Nec capacitors are the only issue. And if so, just to replace them and play in peace. Thnx in advance......
By the way, I bought original ones, NEC tokins, for my PS3 slim 20XXX, but my friend who has the repair shop for many years, told me that no one has the tools to replace them, in my country.....it's some kind of special machinery for soldering those things.
Few years ago, I've replaced the thermal paste completely, under and on the plates.

Firmware Version: 4.87 (build 50723)
Platform ID: CokG11
Product Code: 00 84
Product Sub Code: 00 09
Hardware Config: 4E00FFFF0203BC3C
Syscon Fimware Version: 0832.00010002083E0832 (EEPROM: 00010002083E0832)

Bringup Count: 6783, Shutdown Count: 6389
Runtime: 611 Days, 0 Hours, 43 Minutes, 2 Seconds

Error Log
01: A0801002 Sun Mar 26 16:17:25 2023
02: A0801002 Sun Mar 26 16:04:47 2023
03: A0801002 Sat Mar 25 20:55:41 2023
04: A0801002 Sat Mar 25 17:59:58 2023
05: A0801002 Sat Mar 25 17:49:43 2023
06: A0801002 Sat Mar 25 17:49:34 2023
07: A0801002 Sat Mar 25 17:49:24 2023
08: A0801002 Sat Mar 25 17:49:15 2023
09: A0801002 Sat Mar 25 17:48:58 2023
10: A0801002 Sat Mar 25 17:48:48 2023
11: A0801002 Thu Mar 23 19:31:38 2023
12: A0801002 Thu Mar 23 19:30:47 2023
13: A0801002 Thu Mar 23 19:30:00 2023
14: A0801002 Thu Mar 23 16:06:04 2023
15: A0801002 Thu Mar 23 15:56:18 2023
16: A0802022 Thu Mar 23 15:55:44 2023
17: A0801002 Thu Mar 23 15:25:15 2023
18: A0801002 Thu Mar 23 15:24:14 2023
19: A0801002 Thu Mar 23 15:22:51 2023
20: A0801002 Thu Mar 23 15:22:41 2023
21: A0801002 Thu Mar 23 15:22:30 2023
22: A0801002 Thu Mar 23 15:22:20 2023
23: A0801002 Thu Mar 23 04:07:11 2023
24: A0801002 Thu Mar 23 04:06:09 2023
25: A0801002 Thu Mar 23 04:05:59 2023
26: A0801002 Thu Mar 23 04:05:50 2023
27: A0801002 Thu Mar 23 04:05:01 2023
28: A0801002 Thu Mar 23 04:03:12 2023
29: A0801002 Thu Mar 23 04:02:58 2023
30: A0801002 Wed Mar 22 15:30:17 2023
31: A0801002 Wed Mar 22 15:29:32 2023
32: FFFFFFFF Fri Dec 31 23:59:59 1999
 
I have a COK-002 board on which I've replaced the caps. The console was working for a short time, then shut down and it wouldn't power on again. After disassembly, I measure short on CPU/RSX side of the caps and on all tantal caps (C1501, etc.) around RAM. Any ideas?
 
Gus21 said:
PCBway does it without problems
Click to expand...

Here is what the support is telling me:

Code: 
 Hi,

Thanks for your reply.
There are castelled holes/slots in the file, pls check attached picture and advise, thanks.
 2023-04-05
-----------------------------------

Best Regards


Zoe

Customer Service Executive
Skype:Zoe PCBWay | Tel:86-571-85317609 | WhatsAPP: Zoe Zhang| www.pcbway.com/p



PCBWay.com | Follow us Facebook  and Twitter .

Adr: Room 1105-1108, No.2 Building, Est Mall Business Center,No.698 Changbang Road, Xiacheng District, 310004, Hangzhou, China
------------------ 原始邮件 ------------------


 Date:2023/4/5 3:21:58

Hi,

yes I have a problem. The edges were changed to castellated holes but those are supposed to be edge connectors without extra cost.

Kind regards


Am 04.04.23 um 04:27 schrieb PCBWay Online Services Team:
> Hi,
>
> Thanks for your order on PCBWay.
>
> It is been several days since your order passed review but you haven't made payment.
>
> Is there any problem?
>
> Looking forward to your reply.
>  2023-04-04
> -----------------------------------
>
> Best Regards
>
>
> Zoe
 

Attachments

  • W563848ASP6.png
    W563848ASP6.png
    35 KB · Views: 82
Old Tokin must be removed (mechanically without using a soldering iron), new ones must be soldered by pre-soldering thin wires to the Tokin contacts, so you can get close with a soldering iron when soldering
And you need to make jumpers from a thick wire between the extreme conclusions of Tokin
kLwVZQQ7wwz9iw

kLwVZQQ7wwz9iw
 
Last edited:
zdenko said:
Been looking at the schematic a bit more again.
Fuse 6001 seems to protect a bunch of ancillaries that get voltage supplied by the Q6002 Mosfet. Q6002 seems to be the 5V supply for USB, Cardreader, HDD and Bluray plus a variety of other IC's which then O/P 3.3, 1.8 and 1.5 Voltages.

So when my fuses arrive I will not be plugging in the HDD, Bluray drive, or any USB devices when powering up.

Have not ordered fuses yet.
Should I be buying Slow, Fast or Very Fast Blow?
My gut feel is Fast.
Click to expand...

LOL. Very long time no post.
Honestly no time, but today I had time and nothing else to do so thought I'd open up the old PS3 again.

I replaced blown fuse F6001.
Guess what, it blew on first power up.

I did find this post from ZakeSLM.
PS3 - PS3 NAND Chip Transplant | PSX-Place

Very interesting that he had a blow F6001 as well, and mentioned mosfet Q6002 being the culprit in the end.
I have put a meter across the various legs of the mosfet and I do have shorts which could all mean nothing considering it is still soldered to the board.
I can see 2 other mosfets that are the same and have checked those across the same legs and do not have shorts.
So maybe I have actually found my real problem after not looking for 15 months.

I also measure a short across Capacitors C6019, C6020 and C6023 and looking at the circuit it could technically be one of those as well.
Only way to tell would be by starting to remove those components, which I'm not to keen on as they are all packed very tighly together, and I don't have a steady enough hand for that.

Really, really wishing I did not pull a Tokin off in mid 2021.

I am thinking of trying ACE console repairs in QLD Australia.
Seems they got onto the bandwagon of SYSCON and also do all the Frankenstein mods listed on this forum.
Looks like finally there is a proper place in Australia, not just a bloke with an oven.
 
Need help

Hi everyone, I need help, I bought a PS3 CECHG01 from a local site where I live.
On the b side of the pcb I removed the nec tokins from the rsx and put tantalums.
The PS3 turns on and works.
But when I tried to turn it on other times I noticed the following, something strange happened that I can't understand, when I put the power cable on the PS3 it turns on by itself with the fan at maximum and no image appears on the TV.
I turn off the PS3 on the AC inlet and remove the cable and plug it back in, and that way I can turn on the PS3 normally and it works perfect.
This is happening all the time, I need help.
Can someone help me ?
 
Last edited:
maekawa said:
Hello.
Decoupling capacitor replacement has been discussed for years, but most of it seems to be about power integrity.
But omitting MLCCs from decoupling capacitors in ICs operating at hundreds of Mega/Gigahertz is horrible from an EMC point of view.

When approximating the frequency characteristics of the Proadlizer with MLCCs, a large amount is required like the back side of the RTX3090, but later models without Proadlizers have only not so many MLCCs.

Can anyone measure the capacitance for reference?
Click to expand...

That's why I made the PS3 Tantalizer.
 
Yeah. With, your point about the width of the frequency response is correct as EMC aspect.

What I'd like to ask here is the possibility that the PS3's EMC decoupling capacitors don't have to be "broad" as long as I observed the slim model with the Proadlizer removed.
In other words, it is about the possibility that sufficient performance can be obtained with additional only 1uF or 10uF but not 0.1, 0.01...
what about this?

Of course, we are not Sony's EMC engineers, so keeping the frequency response the same as the original design is always the best choice.
 
maekawa said:
Yeah. With, your point about the width of the frequency response is correct as EMC aspect.

What I'd like to ask here is the possibility that the PS3's EMC decoupling capacitors don't have to be "broad" as long as I observed the slim model with the Proadlizer removed.
In other words, it is about the possibility that sufficient performance can be obtained with additional only 1uF or 10uF but not 0.1, 0.01...
what about this?

Of course, we are not Sony's EMC engineers, so keeping the frequency response the same as the original design is always the best choice.
Click to expand...
They MLCC's are not necessary at all to allow the console to function. As was done before we ever attempted the use MLCCs. It was mainly to emulate what SONY had done on slims. But we have a different voltage regulation module here. The IOR buck phases create a different switching noise/ripple environment compared the one SONY used the slim array of AlPol/MLCC for. What's important is to measure the ripple and be sure it's attenuated. The general rule for that is ripple and noise should be less than 1% of the output voltage. So 1% of 1.0 - 1.3 ish volts that the CELL/RSX tend to use is 10-13mV. When I've tested tokins I've seen ripple as high 100mVpp not cause a YLOD. But usually if they are showing that high, a voltage spike will come along and cause 1001 or 1002 at randome or under load. Witnessed that once.

BTW, this is what bad tokins look like...
bad-tokins-jpg.35397

And this is what good tokins look like...
working-tokins-jpg.35396


You'll notice the CPU Vpp is around 40mV, not under 10, but what I usually consider unacceptably high is anything greater than 50mVpp under load. And that's not the easiest thing to achieve on some particularly noisy consoles.

Usually an array of 12x 470uF TaPol is enough without the MLCCs. The MLCCs help a bit.
 
You must log in or register to reply here.

Similar threads

Back
Top