PS3 Frankenstein PHAT PS3: CECHA with 40nm RSX

Hello,
I wanted to know if with the syscon it is possible to know in which fimrware is the console? If so with what code?
is it also possible to find the console ID?
THANKS.
 
Due to YLOD CECHA console, I reballed CELL and changed RSX to 40nm with ORBIS mod. Now console boots, but it indicates GLOD. When I try to put console to safe mod, i can hear second beep. The problem is there is no signal. I cannot see any error codes via syscon UART. I would like to confirm that does console have firmware above 3.55. Can someone guide me how I can read current fw revision? Southbridge UART? @RIP-Felix maybe you know the way?
 
OneFix said:
Due to YLOD CECHA console, I reballed CELL and changed RSX to 40nm with ORBIS mod. Now console boots, but it indicates GLOD. When I try to put console to safe mod, i can hear second beep. The problem is there is no signal. I cannot see any error codes via syscon UART. I would like to confirm that does console have firmware above 3.55. Can someone guide me how I can read current fw revision? Southbridge UART? @RIP-Felix maybe you know the way?
Click to expand...

UPDATE: Here is bringup log. Booting sequence is normal without HDD? Syscon firmware is v1.0.0_k1. So conclusion is that the syscon firmware could be the problem, because I can see 40nm support added in v1.5.1_k2 version?
Code: 
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.
WmDiscOpr_ChuckOffEvent
[SSM] state: 0101 -> 0201
[POWSEQ] AV Backend Setup
[SSM] state: 0201 -> 0102
[SSM] state: 0102 -> 0202
[SSM] state: 0202 -> 0103
[SSM] state: 0103 -> 0203
[SSM] ssmCb_BeforeBeOn() called.
[SSM] state: 0203 -> 0104
Psbd_SbTransMode_Half:0x20e2

> version
version
v1.0.0_k1
[mullion]$
>
>
 
Last edited:
Can someone help me out. I damaged the R2153, I have the JTP-001, can I salvage one from that board? I cannot find the schematics for that board. The other alternative is to take a resistor from a PS2 GH-023 board that I have around and of which I have the circuit diagram, would there be a problem with that?
 
Last edited:
ClarkyAU said:
Looking for someone within Australia that can carry out this mod/fix for me.

Is there a list of known shops/people??
Click to expand...
I believe ACE Console Repairs is based in Australia

I cant speak on behalf of their work, I've just seen their videos

Sent from my SM-G781U using Tapatalk
 
Passing by to thank those involved in all the work of the tutorial, I got a sealed ceche01 - cok002, and after following the tutorial, the result was a success. I'm from Brazil, and I really appreciate the effort in making a very well explained tutorial.
 
This is going even better for those who fail to reball first time there is second chance. Vrams on rsx are one of case rsx fail to understand better what is happening. I've recovered 7 of 7 40nm rsx now by replace Vrams on rsx. How? Simply one manual cnc mf70 and cut down to 01mm each ram. Now I still scrap that rest of epoxy resin with fiberglass pcb and mg chemicals paint remover(3 hours of scraping) . Still need to think buy one good 80 w laser engraver to burn it automatically.
Now about rams most of them I will test this they are all in xbox 360 on gpu side for those who want to test.
On all my test I bought new sets from Aliexpress new ic set of 7 dollars.
I still wonder what Sony tried to hide that time.
I seen something like test pads on edge of rsx under epoxy/paint mask.
Test left for later 65nm first attempt didn't succes (rsx passed from 3034 /4401 to special glod on normal state, remove, scraped rams second reball still special reball, swapped with another 40nm rsx scraped vram working flawless).
Another test I wish is old 90nm rsx with scrap vram and see if those vrams from 40nm will work. Theoretical should if not I've seen on old xbox 360 same tipe of vram, those models without hdmi port.
Anyway if anyone manage to test all that please let me know. For those who want to test/swap rsx in slims not sure if that was documented, probably most guys friends with Computer Booter found early.
Address of eepcsum is on 7FE
After we add rsx flex IO on 182 and rsx id on 194 we write csum swapped at 7FE (reversed bits as we get on eepcsum). It's always at 7fe.
Not sure if Felix had time to integrate this stage in his tutorial but may help other techs.
Example:
https://s.go.ro/90kk3k6z
 
Last edited:
Code: 
[QUOTE="botakompong, post: 291774, member: 1772"]thanks for friends who have helped, I try to explain how to change the voltage in cok-001 (1.2volt) to 0.95 volts
1. Unplug Q6200 in cok-001
2. replace with Q6200 from another machine (jpt-001, mpx-001, kte-001)
3. lift pins 2,3 and 4
4. ground pins 2 and 3 ( pin 4 nc)
5. finished[/QUOTE]

Would you happen to know how to increase the voltage on the slim versions?
 
florianX12 said:
Hello,
I wanted to know if with the syscon it is possible to know in which fimrware is the console? If so with what code?
is it also possible to find the console ID?
THANKS.
Click to expand...
Not with syscon uart, that I know of anyway.

But the SB UART can tell you which SDK is currently installed. Problem is that the console has to be able to reach the bootloader before it'll spit any SB diagnostics out.
 
OneFix said:
Due to YLOD CECHA console, I reballed CELL and changed RSX to 40nm with ORBIS mod. Now console boots, but it indicates GLOD. When I try to put console to safe mod, i can hear second beep. The problem is there is no signal. I cannot see any error codes via syscon UART. I would like to confirm that does console have firmware above 3.55. Can someone guide me how I can read current fw revision? Southbridge UART? @RIP-Felix maybe you know the way?
Click to expand...
The SB UART will show a rsx rom abort message if the FW is too low. I'm not certain about the orbis modchip, it may not matter. We speculated it might spoof the 90nm rsx ID and still work below the rsx rom revision limit, but AFAIK no one has been able to confirm that.

Be sure you did the resistor mod correctly, otherwise you can get a GLOD.

SB UART Tutorial:
UART_All (COK-00x).jpg

  1. Follow SYSCON UART Tutorial as normal.
  2. Choose the correct command to enable SB UART, depending on your SYSCON version.
    • Older SYSCONs, like 201GB = w 7202 02
    • Newer SYSCONs, like 304GB = w 4202 02
  3. Use putty to UART over serial @ 115200 Baud rate.
  4. Turn the console on and the SB log should just start pouring out. Except when…
    • the console doesn't get far enough into the bootloader for the Southbridge to initialize. For example, if you have a Green Light of Death (GLOD) use the bringup command in SYSCON UART to start the console and check the log. The console should get past the power on sequence and start the bootloader. But in some cases, an "XDR Not initialized" failure can cause a GLOD that does not get far enough for the Southbridge to start. That error can indicate a dead XDR Ram module or broken CPU traces leading to one of them. This is one situation where the SB will not output a log.
(Optional)
Thanks to @M4j0r on the PSX place for the following:
To get more messages from the Hypervisor:
w 42CF 03
w 42F0 03
w 42F1 03
Click to expand...

After that, boot the system and attempt the Southbridge UART installation again. The SB UART log will now be longer and include some information from various Hypervisor services. The PS3 will be slower because of all the log outputs.
To disable the Hypervisor log, use:
w 42CF FF
w 42F0 FF
w 42F1 FF
Click to expand...
 
Last edited:
OneFix said:
Code: 
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.
WmDiscOpr_ChuckOffEvent

...
>
Click to expand...
That WmDiscOpr_ChuckOffEvent makes me think there's an issue with your blueray drive. Not sure, I've not seen that particular message. Kudos on posting the bringup though! This is exactly the kind of useful information that the bringup debug can show us. As for what it meas, well that's a bit of a guess ATM. We are still relatively new at using these messages to diagnose the console. So let up know if you sort the issue out, and maybe we'll be able to associate the fix with this bringup message!!!
 
quake said:
Code: 
[QUOTE="botakompong, post: 291774, member: 1772"]thanks for friends who have helped, I try to explain how to change the voltage in cok-001 (1.2volt) to 0.95 volts
1. Unplug Q6200 in cok-001
2. replace with Q6200 from another machine (jpt-001, mpx-001, kte-001)
3. lift pins 2,3 and 4
4. ground pins 2 and 3 ( pin 4 nc)
5. finished[/QUOTE]

Would you happen to know how to increase the voltage on the slim versions?
Click to expand...
Can you explain what you mean? What voltage are you wanting to increase? The frankenstein mod would involve reducing VDDR voltage, not increasing it.

If you meant, the core voltage (VDDC), it's automatically adjusted on the fly with VID tables, stored on the the buck controller itself, and is requested dynamically by the processor itself. You shouldn't increase it manually, as it will be adjusted automatically. The only reason I can think of for why you would want to is for overclocking it (RSX), but that's not an issue AFAIK. I'm running the most aggressive overclock and it works fine without any voltage adjustments. If for some reason you wanted to increase it you could do so by replacing the shunt resistors, like with current limited graphics cards. Again, not recommended.
 
RIP-Felix said:
Can you explain what you mean? What voltage are you wanting to increase? The frankenstein mod would involve reducing VDDR voltage, not increasing it.

If you meant, the core voltage (VDDC), it's automatically adjusted on the fly with VID tables, stored on the the buck controller itself, and is requested dynamically by the processor itself. You shouldn't increase it manually, as it will be adjusted automatically. The only reason I can think of for why you would want to is for overclocking it (RSX), but that's not an issue AFAIK. I'm running the most aggressive overclock and it works fine without any voltage adjustments. If for some reason you wanted to increase it you could do so by replacing the shunt resistors, like with current limited graphics cards. Again, not recommended.
Click to expand...

Hey man! Great work on the ps3 story video! That was fascinating to stuff!

Yes, I would be interested in a something like a shunt mod for the rsx to increase core or mem clocks with an overclocked cfw on the slim
However I was under the impression that it may be possible to increase voltage in software as well?
 
Not that I'm aware of, but I'm by no means an expert on that. From what I know, the vid is requested by the processor based on load using a 6 digit code. The code is litterally 6 VID pins which are pulled to high or low logic levels. The buck controllers read the VID pins states and adjust the voltage acording to a table that can be found in its data sheet.

What I don't know, is how the Cell/RSX decides what voltage it needs. That's going to be on the firmware side of things. So someone more knowlegable about it could weigh in. For the cell, we have IBMs Hardware Installation Guide (HID), which probably has a section on how that works. I've read it several times, but do not recall the specifics. It's alot to take in and I didn't understand most of it. Although, every time I read it I pick out more and more. It may give you insight into how it works and a potential strategy for adjusting it.

But there is no such guide publically available for the RSX. It's a black box of sony "NOPE!"
 
Hello everyone, I followed the topic and used the tutorial to change an RSX from a DYN-001 motherboard that uses the CXD2991EGB which was dead, I replaced it with a CXD5302, so far everything was correct, getting YLOD when turned on, then I did the changing the syscom with the EC code. I used the bringup command, the console turned on and showed an image normally asking for HD, I turned it off using the power on/off button, I turned it on another 2x to 3x working normally, still with UART connected to the computer, finally I put the HD in and cleared the error log with command "errlog clear", I turned it on using the bringup command, the device turned on normally starting the HD system, I turned it off using the power button and reassembled the entire device, and now the device doesn't turn on at all, as soon as I plug in the power cable it gives 3 beeps and the standby keeps blinking, without even touching the power button, where is my error? Could anyone help me solve this problem and bring this console back to life?
 
Double check the power ribbon cable didnt get pinched or bent/inserted incorectly. I had something similar that turned out to be some bent pins on that ribbon.

If it's not that and the errorlog shows 3034 is back, your reflow failed. Understand that you are not in the clear after a reball/reflow until the console has been throrughly tested over many thermal cycles. The strain from heat can take a while to relax. So it can work fine untile all of a suddon it doesn't. But since this is occuring in standby, I doubt that's what's happeneing. Sounds more like a short.
 
RIP-Felix said:
Double check the power ribbon cable didnt get pinched or bent/inserted incorectly. I had something similar that turned out to be some bent pins on that ribbon.

If it's not that and the errorlog shows 3034 is back, your reflow failed. Understand that you are not in the clear after a reball/reflow until the console has been throrughly tested over many thermal cycles. The strain from heat can take a while to relax. So it can work fine untile all of a suddon it doesn't. But since this is occuring in standby, I doubt that's what's happeneing. Sounds more like a short.
Click to expand...

I reviewed everything I did, and redid the entire sycon process again for the ID of the new CXD, while the red LED kept flashing, finally after all the checks and unplugging it from the socket, when I plugged it in again it returned to normal, It is working and is being tested on the bench, thank you for your help, it was most likely something that went unnoticed on the sycon part
 
I'll wait around to see if someone and increase RSX vddc via software first.

https://www.psdevwiki.com/ps3/RSX
https://www.psdevwiki.com/ps3/CXD5300AGB
What is RSX_FBVDDQ ? Is that gpu core voltage? Is RSX really running at 1.8v?!



Seeing how the chip performed in desktop cards:
https://hwbot.org/hardware/videocards?from=videocard_743#key=geforce_7_series

Seems the stock voltage for 7xxx was around 1.3v and people voltmodded upto 1.5-1.6v ?
https://www.overclock.net/threads/max-voltage-7600-gt.272685/
 
You must log in or register to reply here.

Similar threads

Back
Top