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

squeept said:
That's literally one of the ways I identify BGA issues. With the board in a jig, I place a small weight on top of the heatsink then check if it will work...
Click to expand...
But that would fix a broken electrical connection only temporally... after you remove the weight and/or the motherboard is unbent or/and the materials returns to his original geometry at ambient temperature... you should be back at step one, maybe it works for the next minutes but under heavy workloads is going to fail again

But they are reporting success while making a normal usage of the PS3 for the next hours/days/weeks/months so it looks the tokins replacement fixed it permanently
So... in that cases it was not a broken electrical connection

By looking at your screenshots of the osciloscope and the tests removing some of the tokins it looks the total capacitance is not the cause of the problem
So... not sure, but it seems we are missing something related with the tokins, i mentioned it before several times that is not correct to talk about them as simple capacitors because they does some more "magic"
Actually, if we want to be strict in the naming we should say that are "proadlizers", is the name given by his inventor, and there is no alternative for it, probably they have a patent and are unique in his class
I just call them "the tokins" because is a catchy name, and to simplify it :)

-----------
Btw... someone mentioned before why sony used them... this tokins are more expensive than the tantalum capacitors, so in my oppinion they did an additional effort to keep that power lines "clean"
The tokins increased the manufacturing costs of the motherboard, and initially seems like a good decission for the consumers
Probably the specs of the tokins original from factory was very good (a nice component), the problem is they seems to be prone to failure or/and have a short lifespan

Now the question is... they was aware of it ?, is another case of planned obsolescence ? :rolleyes:
 
Last edited:
Just wondering. Are NEC capacitors in RSX/CELL on PS3 motherboard in series or parallel?

Also are the Caps on opposite sides of motherboard in series or parallel to caps on other side?
 
sandungas said:
But that would fix a broken electrical connection only temporally... after you remove the weight and/or the motherboard is unbent or/and the materials returns to his original geometry at ambient temperature... you should be back at step one, maybe it works for the next minutes but under heavy workloads is going to fail again
Click to expand...

I have no way to measure or verify any of this, so this is pure speculation:

I just took one of the heatsink clamps and put some free weights on it until it flattened itself out beyond what looked normal. Took like 35 pounds. So you can theoretically add 35 pounds of pressure to the chips before you visually notice parts starting to yield. When you remove the heatsink screws from a factory sealed console, they are not tight. They only torque them down from the factory just enough for contact. I suspect when people are putting their systems back together, they're torquing them down as hard as they can or until they bottom out. Screws are like levers. Ever used a screw jack to lift a car?

The weights I use to test are 1kg (2.2lbs). I assume just a few extra degrees tightening on the clamp screws with barely any force during reassembly can add that much extra pressure "permanently." I know plenty of people had shimmed or x-clamp replaced systems that worked for years before they died again.

Here's a request for a video I'd like to see from "the other side" since I've put so much effort in to recording my results: someone who has fixed their YLOD console by replacing the caps, take it apart and set it up so you just have a bare motherboard completely out of the tray without heatsink clamps adding any pressure - like in my first videos with the non-BC model. Xbox 360 heatsinks are nice and light, then just point a box fan or something at it for cooling. Please show the replaced caps and all the hookups to the TV clearly. Boot it up and run GT6 like that for a minute or two.

I'd be much more willing to believe there is more to this story if I saw a video or two of that working. Mostly, I'd just be confused since we would have so much conflicting information.

edit:
lsWGPN5.png

I realized that nobody is going to have the jigs to keep the board flat to make this totally accurate. If you attempt it, please try to support the board from the middle so that it is not "sagging" near the chips, which would drastically increase warpage specifically in the area we're trying not to warp at all.
 
Last edited:
squeept said:
I have no way to measure or verify any of this, so this is pure speculation:

I just took one of the heatsink clamps and put some free weights on it until it flattened itself out beyond what looked normal. Took like 35 pounds. So you can theoretically add 35 pounds of pressure to the chips before you visually notice parts starting to yield. When you remove the heatsink screws from a factory sealed console, they are not tight. They only torque them down from the factory just enough for contact. I suspect when people are putting their systems back together, they're torquing them down as hard as they can or until they bottom out. Screws are like levers. Ever used a screw jack to lift a car?

The weights I use to test are 1kg (2.2lbs). I assume just a few extra degrees tightening on the clamp screws with barely any force during reassembly can add that much extra pressure "permanently." I know plenty of people had shimmed or x-clamp replaced systems that worked for years before they died again.
Click to expand...
I dont remember right now if what im going to say applyes to all PS3 models, but in PS3 slim the heatsink screws are designed to be screwed to the max. The screw "body" is like 2 cylinders, the big cylinder limits the height of that "gap" very accuratelly... in other words the screws are not designed to be "calibrated"
All the pressure depends of the bending of the clamp, some people adds more bending (i did in the PS3 i use normally) but as far i did read the people messing around with the tokins did not increased this bending (at least they did not mentioned it)

squeept said:
Here's a request for a video I'd like to see from "the other side" since I've put so much effort in to recording my results: someone who has fixed their YLOD console by replacing the caps, take it apart and set it up so you just have a bare motherboard completely out of the tray without heatsink clamps adding any pressure - like in my first videos with the non-BC model. Xbox 360 heatsinks are nice and light, then just point a box fan or something at it for cooling. Please show the replaced caps and all the hookups to the TV clearly. Boot it up and run GT6 like that for a minute or two.

I'd be much more willing to believe there is more to this story if I saw a video or two of that working.
Click to expand...
Not sure if someone could do that, most people that are doing this doesnt have spare PS3's to do test "for science"
Your tests was very good btw, thanks for that, you are throwing some light into this mistery :)

Not sure what could be made next to have some proof of what is happening
 
snkplkn said:
Just wondering. Are NEC capacitors in RSX/CELL on PS3 motherboard in series or parallel?

Also are the Caps on opposite sides of motherboard in series or parallel to caps on other side?
Click to expand...
In paralell, and keep in mind that are directly connected with the others at he other side of the motherboard

So... you can think in them as a single group (composed by 4 tokins) for RSX
And another single group (composed by 4 tokins) for CELL

Take a look at this post, i made some drawings, is basically like this for RSX (and the same for CELL, you have this circuit 2 times):
https://www.psx-place.com/threads/t...replacement-ylod-fix.25260/page-6#post-195624
DfMkPLq.jpg


And when you replace all the tokins of a group (either for CELL or RSX) it becomes this:
The yellow wire is mandatory incase of replacing all the 4 tokins of a group
eSjiK4K.jpg


That last image is electrically the same than this, all the tantalums in paralell:
vpDCJRx.jpg
 
sandungas said:
In paralell, and keep in mind that are directly connected with the others at he other side of the motherboard

So... you can think in them as a single group (composed by 4 tokins) for RSX
And another single group (composed by 4 tokins) for CELL

Take a look at this post, i made some drawings, is basically like this for RSX (and the same for CELL, you have this circuit 2 times):
https://www.psx-place.com/threads/t...replacement-ylod-fix.25260/page-6#post-195624
DfMkPLq.jpg


And when you replace all the tokins of a group (either for CELL or RSX) it becomes this:
The yellow wire is mandatory incase of replacing all the 4 tokins of a group
eSjiK4K.jpg


That last image is electrically the same than this, all the tantalums in paralell:
vpDCJRx.jpg
Click to expand...
Bro if we weld the tantalums in parallel with the nec/tokins the capacitance of the tokins will be added to the tantalums to work together ? or is better to get rid of the tokins for better results ?
 
SiegHart93 said:
Bro if we weld the tantalums in parallel with the nec/tokins the capacitance of the tokins will be added to the tantalums to work together ? or is better to get rid of the tokins for better results ?
Click to expand...

Caps in parallel add. At this point, I hope everyone can agree that it's not loss of capacitance that is the problem.

If you're still going to do the repair, then you should absolutely remove them all, because if there's an issue at all it's a short or something from one that's causing problems. Piggybacking more caps will not help that. Besides, you've already got it taken apart so at least be thorough and do it all at once.
 
I agree, is better to remove all the tokins of a group... lets say the 4 tokins for RSX first

Whatever was that additional features made by the tokins is not mandatory to exist, if the people that is reporting success are right they are a proof that the PS3 works normally without any of that "special magic" made by the tokins/proadlizers
 
I used to bend the clamps on the ps3 before theres a differences in the first gen ps3 as it was more stiffer then the later model where it can bend more than normal maybe different metal were used
 
I doubted the capacitors were to blame to start with and I'm more inclined to believe the RSX, if not properly cooled, eventually suffers internal transistor failure from overheating and/or inherent production defects. Pure speculation on my end as I do not have the equipment to actually verify. The only reason heatguns and reballing worked is because the heat coaxed the transistors to work again for a little while before failing again. Sony did not set an aggressive enough fan curve to prevent temps from going into the danger zone (80 degrees or more) and many of them simply cooked themselves. CPUs are fairly resistant to heat, but the GPU is always more sensitive especially 7xxx (which the RSX is based off), 8xxx, and 9xxx series Nvidia cores. The reason why probably the 21xx and later ps3s do not fail early as often is the 40/45nm chips do not produce nearly as much heat as the 90/65nm chips and would have rarely gotten into that same danger zone, plus of course an enhanced fabrication process that was fine tuned over time.

On top of all that, if the thermal compound was compromised, the heatsink and fan cannot expel heat that is bottled up in the CPU/GPU and instead radiates internally, damaging capacities, resistors, other SMD components. Not to mention, the greater the delta between running temp and off temp, the more "flexing" can occur, so I firmly believe in keeping the internal system as cool as possible to minimize this delta.
 
Finally i fixed my PS3 CECHE01 by replacing just 1 nec/tokin on the rsx (bottom side of the board) and 1 nec/tokin on the cell (bottom side of the board). Thanks again to all the brothers of the thread :like:, my ps3 had ylod first even when booting and playing less demanding and high demanding games but i tested the tokins by using the heat gun and the process did the trick and my ps3 booted flaweslly and playing not demanding games wasn't a problem but again with the most demanding like the last of us it ylods. Then i managed to get to my hands 8 tantalums capacitors of 330uf 2.5v and replaced the tokins i said above.

The tantalums even fixed a weird issue that was annoying me on that ps3, every time i wanted to turn it off it just 3 beeps and then the blinking red light (no ylod), so the tantalums are a must to revive your dead ps3 ;)

Now im gonna buy a ylod CECHA01 to revive with tantalums and hope for the best :D

Thanks again
 
squeept said:
I have no way to measure or verify any of this, so this is pure speculation:

I just took one of the heatsink clamps and put some free weights on it until it flattened itself out beyond what looked normal. Took like 35 pounds. So you can theoretically add 35 pounds of pressure to the chips before you visually notice parts starting to yield. When you remove the heatsink screws from a factory sealed console, they are not tight. They only torque them down from the factory just enough for contact. I suspect when people are putting their systems back together, they're torquing them down as hard as they can or until they bottom out. Screws are like levers. Ever used a screw jack to lift a car?

The weights I use to test are 1kg (2.2lbs). I assume just a few extra degrees tightening on the clamp screws with barely any force during reassembly can add that much extra pressure "permanently." I know plenty of people had shimmed or x-clamp replaced systems that worked for years before they died again.

Here's a request for a video I'd like to see from "the other side" since I've put so much effort in to recording my results: someone who has fixed their YLOD console by replacing the caps, take it apart and set it up so you just have a bare motherboard completely out of the tray without heatsink clamps adding any pressure - like in my first videos with the non-BC model. Xbox 360 heatsinks are nice and light, then just point a box fan or something at it for cooling. Please show the replaced caps and all the hookups to the TV clearly. Boot it up and run GT6 like that for a minute or two.

I'd be much more willing to believe there is more to this story if I saw a video or two of that working. Mostly, I'd just be confused since we would have so much conflicting information.

l.
Click to expand...

I pretty much do exactly as u b4 i reball so i think we might be on the same page somewhat. I have the tools and machines for this test, if and when i finally get a console that turns off in the same way this threads suggest i will use this as a guinea pig. unfortunately all the consoles i have attempted this repair on so far haven't worked and i dont have a known candidate yet so i cant say how soon i can get this for ya but i will just so we can have some sorter result to keep this thread going.
 
jacobsson said:
@Naked_Snake1995
Just an idea:

I was thinking we could set up a table witha couple of columns in the first post showing all users in this thread that (successfully and unsucessfully) made this fix and with which tokin replaced, date and date of last working state.
Click to expand...
You mean like a pool, or just a thread reply, sure ill start [emoji23]

Model : CECHC04 - PAL (60GB)

Board: COK-002

Symptoms:Random Shutdown, intensive games, crashes.

NEC Repair Date: 15th June 2019

Cap Used: Tantalum 470uF - 6.3V - 2xRSX/1xCELL (Side-B) - 10 NEC Original

Observations: 330uF Tantalums did work although it didn't last long, GT6 and TLoU would crash only with one RSX Replacement or Original NECs on a fully healthy C04 "Previous Owned System", after the replacement the system holds, regardless of the title executed.

Life-Span Runtime: 156d

Current Status: Fully Working

Sent from my G8141 using Tapatalk
 
Edwired said:
Im currently looking around online for a few ylod and spotted a few cechc04 what the difference from that to cechc03?
Click to expand...
CECHC04 is the PAL PS3 from the European continent and the CECHC03 is from the UK and ireland, they are just the same only difference is region code. I recommend u to get a CECHA01 that is the best model of the ps3 out there with full backwards compatibility
 
I know but they are asking mad money for the CECHA01 nowadays but question if i got CECHC04 would that affect games region?
 
off topic CECHA01 have full retro CECHC04/3/E01 have GS only but this system have one vantage which CECHA01 don't CCHC0x in ps2 have more quality graphics than CECHA01 but don't have the same compatibility i go to give two examples driver 3 don't run well on ps3 60gb european but run well on ps3 60gb CECHA01 Burnout 3 in european 60gb have better quality image than CECHA01 in same game
 
Just to shime in, the NTSC-J CECHA00 should be equivalent to the NTSC-U CECHA01. Also, you should be able to configure to use GXEMU over PS2EMU for burnout 3 with a full BC system somehow right?
 
Last edited:
jacobsson said:
Just to shime in, the NTSC-J CECHA00 should be equivalent to the NTSC-U CECHA01. Also, you should be able to configure to use GXEMU over PS2EMU for burnout 3 with a full BC system somehow right?
Click to expand...
Is the same.
I never tried but what i tested was burnout 3 in European ps3 60gb have better quality image than 60gb Japanese i have one 60gb European and one 20gb Japanese.
 
You must log in or register to reply here.

Similar threads

Back
Top