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

Tupakaveli said:
Probably ~24awg. I was also only using 2 wires.
Click to expand...

i can only advise on what ive used that worked but i would say something similar should suffice but if u have any electronics store where u buy ur wire it may benefit u more as it will come with more info.

Ive got this figure 8 wire that ive split in half so i get single wires. It doesn't tell me the gauge size since its figure 8? seems odd as it should still haha but its rated for low voltage use 50V @ 5A. consists of 50 strands @ 0.12mm

this hasnt melted for me im guessing its around 20-22awg?
 
Edwired said:
any process on your ps3 yet?
Click to expand...

Bad news.

I put all the new tantalum capacitors on both sides... made two bridge "18awg" (cell and rsx) and when it turns on it just beep once and shuts off completely (blackout - as if the power source was disarming).

I checked the tantalums to see if there were any shorted or some welds between positive and negative but nothing unusual.
 
wrx884 said:
i can only advise on what ive used that worked but i would say something similar should suffice but if u have any electronics store where u buy ur wire it may benefit u more as it will come with more info.

Ive got this figure 8 wire that ive split in half so i get single wires. It doesn't tell me the gauge size since its figure 8? seems odd as it should still haha but its rated for low voltage use 50V @ 5A. consists of 50 strands @ 0.12mm

this hasnt melted for me im guessing its around 20-22awg?
Click to expand...

Yeah I have some thicker 7.5A rated wire here, just haven't got around to it yet.
 
Alisson said:
Bad news.

I put all the new tantalum capacitors on both sides... made two bridge "18awg" (cell and rsx) and when it turns on it just beep once and shuts off completely (blackout - as if the power source was disarming).

I checked the tantalums to see if there were any shorted or some welds between positive and negative but nothing unusual.
Click to expand...

So it went from YLOD (pre modification) to one-beep -> black out (w/o even the red standby led lit after)?
I think it's time to post some photos on your mod.
 
jacobsson said:
So it went from YLOD (pre modification) to one-beep -> black out (w/o even the red standby led lit after)?
I think it's time to post some photos on your mod.
Click to expand...

Hello.

Exactly. The console is in standby I turn on and beep once and shut down completely without being in standby again.

Edit:
I noticed that 12v pins/connector on the board are shorted or low resistence?? using dmm, if i just plug that 5v cable from the power supply the console turns on it shows YLOD and gives three beeps, as it was before the tantalum.

12v fully shorted or it has 2.4ohms only ??

tan.jpg

That yellowish tantalum is not burnt, it's dirty.
Top side: 330uf
Underside: 470uf

Edit2:
Checked all thermistors and fuses are good.
 
Last edited:
Alisson said:
Hello.

Exactly. The console is in standby I turn on and beep once and shut down completely without being in standby again.

Edit:
I noticed that 12v pins/connector on the board are shorted or low resistence?? using dmm, if i just plug that 5v cable from the power supply the console turns on it shows YLOD and gives three beeps, as it was before the tantalum.

12v fully shorted or it has 2.4ohms only ??

View attachment 20796
That yellowish tantalum is not burnt, it's dirty.
Top side: 330uf
Underside: 470uf

Edit2:
Checked all thermistors and fuses are good.
Click to expand...

I can't spot anything that looks off to me. I think the soldering looks pretty clean, some marks here and there, but that shouldn't really matter as long as you didn't hit any traces.

Do you get different resistance readings on the NEC pads if you measure GND <-> GND and GND <-> Vcc?
 
Have you checked if any of the really small caps havent been broken off around the nec tokin or shorted with soldering tin. i say try cutting the wire off and do the bridge at the back to see if anything changed
 
jacobsson said:
I can't spot anything that looks off to me. I think the soldering looks really clean.
Do you get different resistance readings on the NEC pads if you measure GND <-> GND and GND <-> Vcc?
Click to expand...

GND <-> GND = 0.0 ohms (close)
GND <-> Positive NEC Pad (RSX/Cell) = 2.4ohms
GND <-> Positive Board Connector 12v = 2.4ohms
Positive Board Connector 12v <-> Positive RSX Nec Pad = 0.0 ohms (close)
Positive Board Connector 12v <-> Positive CELL Nec Pad = 5.0 ohms ??
Positive NEC Pad RSX <-> Positive NEC Pad CELL = 5.0ohms ??

Cell positive is not "closed" with 12v power connector like RSX.
RSX and CELL positives don't close, is that normal?

Edwired said:
Have you checked if any of the really small caps havent been broken off around the nec tokin or shorted with soldering tin. i say try cutting the wire off and do the bridge at the back to see if anything changed
Click to expand...

I checked all and they are perfect, especially those little ones near the NEC.

Strange is all the positive points of the board are giving 2.4ohms resistance between the ground. As I don´t have another motherboard to test I do not know if this is normal.
 
Last edited:
I just had a thought about the value of 4 470uf and 4 330uf you are too far under the normal range of 4800uf as on the board you put in 3200uf that might be the reason why it wont start that might be wrong as i was thinking of something else my bad.

I say is to double check if any of the tan caps are not shorting with v-in/v-out to gnd as there a small margin to cause a short. Try cutting the wire to see if the problem changes on the board.

Also give the area a clean with iso 99.9% to clear away any flux or residue that might be causing the short
 
Last edited:
Edwired said:
I just had a thought about the value of 4 470uf and 4 330uf you are too far under the normal range of 4800uf as on the board you put in 3200uf that might be the reason why it wont start that might be wrong as i was thinking of something else my bad.

I say is to double check if any of the tan caps are not shorting with v-in/v-out to gnd as there a small margin to cause a short. Try cutting the wire to see if the problem changes on the board.

Also give the area a clean with iso 99.9% to clear away any flux or residue that might be causing the short
Click to expand...
Hmmm, yeah that seems to be
@Alisson , you can solder more capacitors externally with wires just as a temporal fix (for testing mostly), in the way it was suggested by @jacobsson before

Incase it works, you know, is needed to do some solder job to add more tantalums and to reorganize them

Wait, is ok, every processor have 8 * 330 + 8 * 470... is more than enought
 
Last edited:
sandungas said:
Hmmm, yeah that seems to be
@Alisson , you can solder more capacitors externally with wires just as a temporal fix (for testing mostly), in the way it was suggested by @jacobsson before

Incase it works, you know, is needed to do some solder job to add more tantalums and to reorganize them

Wait, is ok, every processor have 8 * 330 + 8 * 470... is more than enought
Click to expand...
Yeah sorry for the mishap on my typo my brain shortcircuited when i was thinking and typing the same time my bad
 
sandungas said:
Hmmm, yeah that seems to be
@Alisson , you can solder more capacitors externally with wires just as a temporal fix (for testing mostly), in the way it was suggested by @jacobsson before

Incase it works, you know, is needed to do some solder job to add more tantalums and to reorganize them

Wait, is ok, every processor have 8 * 330 + 8 * 470... is more than enought
Click to expand...

Yes, 6400uf per chip (RSX & Cell)
 
Ok then it might well need 16 470uf tan caps on the bottom of the board where the rsc/cell are not present. As it seems clear the 330uf on the bottem arent enough it ok to have 330uf on nec tokin 900uf version

Judging that you have dia-001
*Phat Models: "470uF"

*DECR-1000 - "TMU-520"
*DECHA00A - "COK-001"
*CECHA - "COK-001"
*CECHB - "COK-001"
*CECHC - "COK-002"
*CECHE - "COK-002/W"
*CECHG - "SEM-001"
*CECHH - "DIA-001"
*CECHJ - "DIA-002"
*CECHK - "DIA-002"
*CECHL - "VER-001"
*CECHM - "DIA-001"
*CECHP - "VER-001"
*CECHQ - "VER-001"
*DECR-1400 - "DEB-001"

"Due to the DEB-001 being identical to a DYN-001 Board, only the RSX needs a capacitor repalcement, you can use either 470uF, although 330uF will work just fine"

*Slim Models: "330uF" - RSX Only

*CECH-20XX - "DYN-001
 
Man I'm mad now my brother came into my room looking for something and he managed to drop something on the cok-002 board and I didn't know till now I had a look and im gutted that on the back of the board it missing a bunch of real tiny caps on the rsx side
 
Edwired said:
Man I'm mad now my brother came into my room looking for something and he managed to drop something on the cok-002 board and I didn't know till now I had a look and im gutted that on the back of the board it missing a bunch of real tiny caps on the rsx side
Click to expand...

Ouch! I guess taking high res images to watch on a computer screen and comparing them to a similar board from the internet is the way to go. Did he drop something heavy?
 
sandungas said:
The fact is in the electrical circuit all the tokins of a side of the motherboard are connected directly with the other tokins of the other side, is like this:
Power rail ---> 4 tokins (2 at top, and 2 at bottom) ---> RSX
Power rail ---> 4 tokins (2 at top, and 2 at bottom) ---> CELL

So forget about what is top and what is bottom... what matters here is what is CELL and what is RSX ;)

The only difference in between top and bottom is what some people have mentioned before... the tokins at bottom (opposite side of components) doesnt have an active cooling, so they overheats more along all his life, so are more prone to failure by overheats
But... we dont know if are dying by overheat (most probably but we are not sure)


The RSX goes first, his tokins are more prone to failure because the demanding voltage peaks of RSX varies a lot more than CELL
This can be deduce indirectly by looking at the temperature ingame... while you are playing the CELL has a more constant workload... but RSX varyes a lot
That variation in RSX heat is an indirect proof that his voltage requirements are varying too

But remember... RSX have 2 tokins at top of the motherboard and another 2 at bottom

The 2 RSX tokins at bottom (the ones that overheats more) goes first, then the other 2 tokins for RSX to complete his group of 4 tokins

After that you should make a test to see if the problem is fixed... then move to CELL and do the same


Is mandatory if you replace the 4 tokins for either CELL or RSX

The tokins have that bridge internally, if you keep at least 1 tokin (from the group of 4) then you dont need the bridge wire


It doesnt matters what you use, the point is you need to equal (or exceed a bit) the original capacitance of the group of 4 tokins

4 tokins of 0E108 makes a total of 4000 uf
4 tokins of 0E128 makes a total of 4800 uf


By destroying them carefully :D
In this thread there is a talk where we was talking about it, i agree with everything @jacobsson @wrx884 said, they explained it better than me :)
https://www.psx-place.com/threads/ps3-cechg-nlod.26640/#post-210783

Is like removing tiny SMD leds... if you try to remove it without damaging the led is a waste of time, when you get close with the solder iron the plastic transparent material of the led breaks, so is imposible
the best thing you can do is to burn the madrefocker with the solder iron, then cleanup the burned plastic and residues, and then drop a huge ball of solder tin in top of it, this is going to work for sure, and is safe because the motherboard can handle that amount of heat without problems


No, but mostly because you need a lot of them... in a PC motherboard you are not going to find so many

If you have problems to find the tantalum capacitors i guess you could use electrolitic capacitors (the standard ones)
This ones have a lot more capacitance, and are easy to find in PC motherboards, since pentinum 4 you are going to find a bunch around the CPU socket, and are rated up to 1500uf or so (very good for his size)
The problem incase of using this type of electrolitic capacitors is there is no room for them inside the "sandwich" of the metal shields that covers the PS3 motherboard, but well maybe you could make some mod in the metal shield (cuts, or holes, or whatever)
Click to expand...
Can i piggy back a electrolitic capacitor on tokin to testing purposes bro?
 
Yeah something heavy don't know what. It probaly gone past it to add injury the caps I ordered two weeks ago just arrived in ireland was looking forward to getting work done to the board. Now I gonna have to find another cechc03 ps3. As I got in touch with the seller who I was gonna buy the cechc03 ps3 told him I pay extra for postage and he gladly accepted so should have it late next week
 
You must log in or register to reply here.

Similar threads

Back
Top