PS3 CECHC04 - delayed YLOD even after replacing 3 NEC/TOKIN caps in the RSX area

N

Norrecito

Forum Noob
Hello,

I have recently picked up PS3 fat (CECHC04) that wouldn't boot. When trying to start, it went green then yellow after 3 seconds and finally blinking red.

I think this is called the yellow light of death (YLOD).

The HDD was missing however the warranty seal was still intact and I've found this wonderful guide here on PSX place so I have decided to order some tantalum capacitors (470uf, 6.3V, "D" size):
https://www.psx-place.com/threads/tutorial-research-nec-tokin-capacitors-replacement-ylod-fix.25260/

I've started by replacing the right NEC/TOKIN cap of the RSX area (with 4 tantalum capacitors) on the top of the board but ended up with the same delayed YLOD once I have assembled the console. I assumed that this means that more NEC/TOKIN caps needs to be replaced so I have replaced the second one on the RSX area and both caps for the CELL(?) in case that is the one not getting enough power. (So all NEC/TOKIN caps got replaced on the top side of the board) Even after that, I still had the same delayed YLOD. As a final attempt, I replaced one of the NEC/TOKIN caps of the RSX area on the bottom. (I only had 3 tantalum capacitors left so I have used only 3 for this one) but the issue persist.

According to what I've understand from the guide, replacing 2 of the NEC/TOKIN capacitors in the RSX area should supply more than enough current for the console to boot so I'm starting to think that the issue lies somewhere else in the system.

To summarize things: I had a console with a 3 seconds delayed YLOD and still have the very same YLOD after replacing 5 of the NEC/TOKIN capacitors. (3 on RSX area, 2 in CELL)

I've noticed two things:
-The diode mode readings (multimeter) of the positive and negative rails of the NEC/TOKIN capacitors show a very low value (like 005 and 007) which seems like a short to me but on the other hand, I don't know what readings I would get from a working motherboard, that is, I have nothing to compare it to. Wouldn't a short in this area cause an immediate YLOD instead of one that is delayed by 3 seconds?
-There is some corrosion on the top side of the shielding. I have cleaned up the rust(?) from the motherboard and it cleaned up pretty nicely and nothing seems like it was affected however I haven't checked it with a microscope just yet.

It's a "COK-002" motherboard.

I have also:
-Checked every fuse on the board (even thermal fuses) and none of them are blown.
Any comment that would help me to pinpoint this issue is much appreciated! :)

https://ibb.co/BLzK7MT
https://ibb.co/GnpLytr
https://ibb.co/fNkHp2T
https://ibb.co/mBnSC1D
https://ibb.co/MZMdNxs
https://ibb.co/fnB1VCT
https://ibb.co/qxn150w
https://ibb.co/T0QK7Cm
 
pinky said:
you might want to ask here: https://www.psx-place.com/threads/tutorial-research-nec-tokin-capacitors-replacement-ylod-fix.25260/

you're more likely to get a response er, at least more quickly probably. you can request this post/thread be merged with the thread above, then have my post deleted since it will be out of place then.
Click to expand...

I assumed that it is easier to follow this way. (As a separate topic) I mean if more than one person is a having a problem and all of them write to a single topic, it is kinda hard to follow but maybe I'm wrong.

Anyways thanks for the suggestion, I will definetely consider it if I don't get an answer :)
 
What we suggest is faulty finding YLOD thread is to use rs232 adapter to find errors, that 5 or 10 seconds can be a issue for rsx but not sure atm, must do some errors reports.
About measuring cpu/rsx power line use multimeter set to smallest gain for resistance and report that as well. A short there will give instant off, won't start 5 or 7 seconds.
 
pinky said:
yeah, I guess that's true. I hope you figure it out, my friend. @RIP-Felix @vyktormvmpay25 , I just mentioned a couple people well-versed in hardware problems with the ps3, so they'll know about this thread.
Click to expand...

Thanks pinky, very kind of you! :)

vyktormvmpay25 said:
What we suggest is faulty finding YLOD thread is to use rs232 adapter to find errors, that 5 or 10 seconds can be a issue for rsx but not sure atm, must do some errors reports.
About measuring cpu/rsx power line use multimeter set to smallest gain for resistance and report that as well. A short there will give instant off, won't start 5 or 7 seconds.
Click to expand...

Sounds good!

I'm trying to locate a guide that tells what hardware and software I need in order to read the error codes. So far I only seem to find the software related things.

By the way, that YLOD is delayed by 3 seconds and not 5-10. Does this change anything?
 
Last edited by a moderator:
I've just found an hour long Youtube video about the topic (reading error codes with SYSCON) and managed to order everything that I need from a local store. If it goes well, I will get it on Tuesday.
 
Hello,

The package has arrived and I even managed to fetch the error log.

Code: 
> AUTH
Auth successful
> ERRLOG GET 00
00000000 A0403034 FFFFFFFF
> ERRLOG GET 01
00000000 A0404412 FFFFFFFF
> ERRLOG GET 02
00000000 A0403034 FFFFFFFF
> ERRLOG GET 03
00000000 A0404412 FFFFFFFF
> ERRLOG GET 04
00000000 A0403034 FFFFFFFF
> ERRLOG GET 05
00000000 A0404412 FFFFFFFF
> ERRLOG GET 06
00000000 A0403034 FFFFFFFF
> ERRLOG GET 07
00000000 A0404412 FFFFFFFF
> ERRLOG GET 08
00000000 A0403034 FFFFFFFF
> ERRLOG GET 09
00000000 A0404412 FFFFFFFF
> ERRLOG GET 0A
00000000 A0403034 FFFFFFFF
> ERRLOG GET 0B
00000000 A0404412 FFFFFFFF
> ERRLOG GET 0C
00000000 A0403034 FFFFFFFF
> ERRLOG GET 0D
00000000 A0404412 FFFFFFFF
> ERRLOG GET 0E
00000000 A0403034 FFFFFFFF
> ERRLOG GET 0F
00000000 A0404412 FFFFFFFF
> ERRLOG GET 10
00000000 A0403034 FFFFFFFF
> ERRLOG GET 11
00000000 A0404412 FFFFFFFF
> ERRLOG GET 12
00000000 A0403034 FFFFFFFF
> ERRLOG GET 13
00000000 A0404412 FFFFFFFF
> ERRLOG GET 14
00000000 A0403034 FFFFFFFF
> ERRLOG GET 15
00000000 A0404412 FFFFFFFF
> ERRLOG GET 16
00000000 A0403034 FFFFFFFF
> ERRLOG GET 17
00000000 A0404412 FFFFFFFF
> ERRLOG GET 18
00000000 A0403034 FFFFFFFF
> ERRLOG GET 19
00000000 A0404412 FFFFFFFF
> ERRLOG GET 1A
00000000 A0403034 FFFFFFFF
> ERRLOG GET 1B
00000000 A0404412 FFFFFFFF
> ERRLOG GET 1C
00000000 A0403034 FFFFFFFF
> ERRLOG GET 1D
00000000 A0404412 FFFFFFFF
> ERRLOG GET 1E
00000000 A0403034 FFFFFFFF
> ERRLOG GET 1F
00000000 FFFFFFFF FFFFFFFF
>

I did some reading on "A0403034" and I'm afraid I already know the answer: it's a cracked solder joint under the RSX. Am I right? :)
 
Yes I am.

I've reflowed the RSX and the console now turns on. So far so good :) Haven't tested any games yet but I definetely need to clean that disc drive because it took ages until the game which was left inside the system came out.

Reading the SYSCON error log was very useful indeed, thanks for the help! :)
 
Just în case you get similar errors, reball both, I'm not sure about rsx 90 nm how much will last (depending on heating profile while soldering). If you still want to recover board with a dead rsx can add a modchip and slim /superslim rsx 40 nm new model.
 
Norrecito said:
Yes I am.

I've reflowed the RSX and the console now turns on. So far so good :) Haven't tested any games yet but I definetely need to clean that disc drive because it took ages until the game which was left inside the system came out.

Reading the SYSCON error log was very useful indeed, thanks for the help! :)
Click to expand...
Hi, I am kind of encountering the same issue. For reflow, are you heating the motherboard with a hot air gun or put in oven? What temperature and time did you set? Thanks.
 
tckoo1411 said:
Hi, I am kind of encountering the same issue. For reflow, are you heating the motherboard with a hot air gun or put in oven? What temperature and time did you set? Thanks.
Click to expand...

Calling what I did a "reflow" is most likely an overstatement because I lack the proper equipment that a skilled technician would have used. I was way too excited that I could read the SYSCON error log and find the root cause of the issue by searching for forum posts.

Anyways I will write down what I did (in case it helps someone) just please DON'T use it as a guide on how to do reflows :)

I have removed all the thermal pads near the RSX and cleaned the thermal paste from both RSX and CELL. I only own a cheap hot air rework station which I set to 350C and 6 out of 7 air flow. I have warmed the board from the underside for 2 minutes, flipped it and gave another 2 minutes to the top side (I put an old, broken lock under the board to keep it even). Then I have applied flux to the sides of the RSX and tilt the board so it could get under it. Finally, I have concentrated only on the RSX for 2 minutes and kept circulating the bigger area afterwards for a minute or two to help it cool down a little slower. I haven't touched the board for an hour, cleaned all the flux and reassembled the console and the YLOD is gone (for now).

I don't think that it will be a long lasting repair (as it was not a proper reflow) but at least it proves that the issue is the RSX and reading those error codes from SYSCON is a very useful thing to do.

I'm only planning to use this console for PS2 games...and will report back once I'm back to the YLOD :)
 
Norrecito said:
Calling what I did a "reflow" is most likely an overstatement because I lack the proper equipment that a skilled technician would have used. I was way too excited that I could read the SYSCON error log and find the root cause of the issue by searching for forum posts.

Anyways I will write down what I did (in case it helps someone) just please DON'T use it as a guide on how to do reflows :)

I have removed all the thermal pads near the RSX and cleaned the thermal paste from both RSX and CELL. I only own a cheap hot air rework station which I set to 350C and 6 out of 7 air flow. I have warmed the board from the underside for 2 minutes, flipped it and gave another 2 minutes to the top side (I put an old, broken lock under the board to keep it even). Then I have applied flux to the sides of the RSX and tilt the board so it could get under it. Finally, I have concentrated only on the RSX for 2 minutes and kept circulating the bigger area afterwards for a minute or two to help it cool down a little slower. I haven't touched the board for an hour, cleaned all the flux and reassembled the console and the YLOD is gone (for now).

I don't think that it will be a long lasting repair (as it was not a proper reflow) but at least it proves that the issue is the RSX and reading those error codes from SYSCON is a very useful thing to do.

I'm only planning to use this console for PS2 games...and will report back once I'm back to the YLOD :)
Click to expand...
Just to understand, have you delid cpu and rsx? If so never reflow with high temperatures without ihs on ic may destroy those. If you want to delid and will fail, just add silicone back, let it cure and reball, remove that silicone and add new thermal paste. That thermal silicone adhesive should be easy remove with ipa alcohol. I'll put a photo of it, at least is easy to find it in Europe.
 
vyktormvmpay25 said:
Just to understand, have you delid cpu and rsx? If so never reflow with high temperatures without ihs on ic may destroy those. If you want to delid and will fail, just add silicone back, let it cure and reball, remove that silicone and add new thermal paste. That thermal silicone adhesive should be easy remove with ipa alcohol. I'll put a photo of it, at least is easy to find it in Europe.
Click to expand...

Nope, I didn't even try to delid any of the ICs. I heard too many stories where people accidentally destroyed the ICs during delid attempts.

I consider myself lucky that I haven't killed anything with my "reflow" and the system actually boots :)

Reballing is out of the scope now as I lack both the knowledge and the equipment needed for it.
 
I did a very brief testing and everything seems to be fine so far, with one exception:
After a few minutes, the fan kicks in and the console gets quite loud. Not sure why this happens as I have cleaned the old thermal paste and applied Artic MX-4 on both RSX and CELL but on the otherhand, this is my first ps3 fat so maybe this is normal.

Anyway, I will continue testing with PS2 games to see how long my repair lasts.
 
Delid will give stability with temperatures. Think under ihs you still have old thermal paste. In webman an 68 temperature constant în game is maintained by 40 % of fan speed automatically. Now if an hour of gaming will run isn't enough, usually 3~4 maximum 6 months of use after reflow even after delid. Now last resort will be delid if you get those temperature panics.
For cpu sit back in place use that AG TermoGlue.
 
You must log in or register to reply here.

Similar threads

Back
Top