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

[ElGris]

@pure3d2 You paid 36 bucks for 400 of them? I paid 19 to a chinese for only 100, what the hell..

Also, I don't recommend you the fishing line. It's a pain in the ass, on fats models is complicated 'cause the silicon is really hard. I even tried cutting wire, the one that is used to delid broken glass from a phone screen. Same story. I used a piece of metal that was really easy to use, search for the thread I posted today. If you're not sure about it, don't doubt, use the fishing line.

 

[pure3d2]

@pure3d2 You paid 36 bucks for 400 of them? I paid 19 to a chinese for only 100, what the hell..

Also, I don't recommend you the fishing line. It's a pain in the ass, on fats models is complicated 'cause the silicon is really hard. I even tried cutting wire, the one that is used to delid broken glass from a phone screen. Same story. I used a piece of metal that was really easy to use, search for the thread I posted today. If you're not sure about it, don't doubt, use the fishing line.

I did try the oil knife but it just felt really unsafe so I will give the fishing line a try. I ordered a roll of 0.1mm nylon fishing line and some key rings that I will wrap with fabric then see how it goes. I'd imagine holding the fishing line with bare hands or gloves would be painful so perhaps the key rings will help.

 

[ElGris]

I did try the oil knife but it just felt really unsafe so I will give the fishing line a try. I ordered a roll of 0.1mm nylon fishing line and some key rings that I will wrap with fabric then see how it goes. I'd imagine holding the fishing line with bare hands or gloves would be painful so perhaps the key rings will help.

I actually used a very malleable metal sheet. And about the fishing line, I used a 0.40mm one, but I also had problems with the cutting wire, trying to do some Agent 47 stuff to that CELL and failed, smh.
BTW, how did you removed thos NEC/T? Using a knife or heat gun (which)?

 

[pure3d2]

I actually used a very malleable metal sheet. And about the fishing line, I used a 0.40mm one, but I also had problems with the cutting wire, trying to do some Agent 47 stuff to that CELL and failed, smh.
BTW, how did you removed thos NEC/T? Using a knife or heat gun (which)?

For removing the NEC/TOKINs, I used my hot air station at 400C to soften the plastic and then used an oil knife to scrape it up. I used a large knife edge soldering tip (station set to 400C) to remove the metal contacts of the NEC/TOKINs. I added leaded solder and Amtech 559 flux paste (helped immensely). Then I used a desoldering braid and a 20mm spatula soldering tip to clean up the pads.

I did see your post about using the calendar metal strip. In one of your pictures, I noticed that there were 2 shiny orange spots on the XELL. Is that debris or did the substrate get scraped away by a different tool?

 

[ElGris]

@pure3d2 Damn, I need those adapters for my soldering iron. Which brand it is the station?

Yeap, those are the marks that I did before, with another metallic thing. I used that CELL damaged to test the calendar sheet I got and worked flawlessly. Later I realized it could be fixed it and did it. Was really happy, I must say lol.

 

[pure3d2]

@pure3d2 Damn, I need those adapters for my soldering iron. Which brand it is the station?

Yeap, those are the marks that I did before, with another metallic thing. I used that CELL damaged to test the calendar sheet I got and worked flawlessly. Later I realized it could be fixed it and did it. Was really happy, I must say lol.

The soldering station is a Hakko FX-951 and the spatula tip is a T15-1403. I need to find a calendar with a strip like that. I haven't seen those kinds of calendars since I was a kid. Most of the ones you get now are just paper with a reinforced hole punched at the top.

 

[TItans_Test_Team]

Thank you for the kind words.

As for the tantalums, these are the exact ones I bought: https://www.newark.com/panasonic/2r5tpe470mc/tantalum-capacitor-470uf-2-5v/dp/98W0352

They're in the same family as the other Panasonic ones others have posted. I'm not exactly sure why Newark sells these for $0.068 each. The Chinese sellers on eBay and AliExpress sell them for anywhere from $0.21 to $0.25 or more. I actually cancelled my AliExpress order because they wouldn't ship until Feb 4th (after Chinese New Year).

I bought 400 for under $36 shipped. If you're in the US, choose USPS as the shipping method and it'll come out to be $5.95 for shipping, which is cheaper than UPS or FedEx. Note: during checkout, they won't display the shipping charge if you choose USPS--you'll only see it once they process your order. It only took 3 days to arrive.

So I noticed that you went with the 2.5V caps instead of the recommended 6.3V caps.

Aside from the price and voltage is there really any difference the between the 6.3V and 2.5V caps?

the 2.5V caps are significantly cheaper than the 6.3 Caps but if they do the same job then they are the obvious win.

 

[demetris]

Again, 2.5v are fine cpu and gpu hardly reach 1.6v tops

Sent from the Island of Venus

 

[wrx884]

6.3V is just the voltage rating. If u read the article properly it tells u 2.5V is recommended but 6.3V will work as thats what seems to be what some members can only get their hands on, no other reason really.

 

[ElGris]

Again, 2.5v are fine cpu and gpu hardly reach 1.6v tops

Sent from the Island of Venus

Now that you mention it, are you sure it's safe to use only 2.5v caps? I've saw the schematic a little bit and I found low voltajes on the RSX, but some of them are 3.3v, and for some lines like 1.5v, they use 6.3v caps, for example, a lot more. According to the datasheet, the original NEC/T are 2.5v, but for some reason they died so quickly too. I think I can use in newer models the ones with 2.5v, and fat models 6.3v, just in case.

 

[demetris]

Am sure. I don't know any cpu that works that high(3.3v)

Sent from the Island of Venus

 

[Edwired]

Maybe sony thought it was a good idea to add protection against overvoltage

 

[pure3d2]

Yeap those are the ones i am using and never failed me. I went for 330 though as to be usable on both models(fat+slim)

Sent from the Island of Venus

Why isn't the 470 compatible with all models but the 330 is?

 

[demetris]

The NEC/TOKIN Capacitors have the serial number 0E128 - 1200uF "Phat Models" and 0E108 "Slim Models" - 1000uF Capacitance respectively. For the slims i play it safe and i add 1x100uf ceramic capacitor if needed (photos attached) because sometimes capacitance is out of spec +/-

Sent from the Island of Venus

 

[hoyohoo]

Hey guys, ok, so, I had some partial success. I ordered 20 tantalum caps von the manufacturer KEMET off of ebay and I soldered a total of 12 of them and could revive the ps3 for the first time ever. It worked for like 2-3 Minutes, before, yet again, doing the YLOD and turning off.
Now it won't turn on again, not for a second.
So, either the tape I put on top the caps melted and now something is shorting to ground, or the caps aren't sufficient, as I really need more caps (however, I don't know where to fit them, they are quite big).
Ooor, the RSX is so done that it already killed the new caps.


What should I do next?

 

[ElGris]

@demetris Are you adding those in paralel along with the tantalums? BTW, the paper writing has a typo, and it says nano, not micro.

Hey guys, ok, so, I had some partial success. I ordered 20 tantalum caps von the manufacturer KEMET off of ebay and I soldered a total of 12 of them and could revive the ps3 for the first time ever. It worked for like 2-3 Minutes, before, yet again, doing the YLOD and turning off.
Now it won't turn on again, not for a second.
So, either the tape I put on top the caps melted and now something is shorting to ground, or the caps aren't sufficient, as I really need more caps (however, I don't know where to fit them, they are quite big).
Ooor, the RSX is so done that it already killed the new caps.
View attachment 23644 View attachment 23645

What should I do next?

You need more brute force. I use at least 24 per console. 16 on the RSX, and the rest on CELL. If the console is still trying to boot, you didn't messed up, but if the console just shutdown the next yyou turn it on, then you have a short. I don't think so, if you some electrical tape over the caps, there's no risk. Add more on the RSX, and listen to the fan behaviour, 'cause if launchs at high speed in just a few seconds, you'll have to delid before even try to use it. The important thing is that the PS3 is still trying to boot.

 

[marciolsf]

Hey guys, ok, so, I had some partial success. I ordered 20 tantalum caps von the manufacturer KEMET off of ebay and I soldered a total of 12 of them and could revive the ps3 for the first time ever. It worked for like 2-3 Minutes, before, yet again, doing the YLOD and turning off.
Now it won't turn on again, not for a second.
So, either the tape I put on top the caps melted and now something is shorting to ground, or the caps aren't sufficient, as I really need more caps (however, I don't know where to fit them, they are quite big).
Ooor, the RSX is so done that it already killed the new caps.
View attachment 23644 View attachment 23645

What should I do next?

Yeah, those are huge! I can tell they're 470uf, but what's the voltage?

 

[demetris]

Yes i do and is μ which is the m in English and is the correct letter but since the English alphabet dont have the μ its replaced by the lookalike u

Sent from the Island of Venus

 

[TItans_Test_Team]

the 6.3V are way taller than the 2.5V which makes fitment in with the shields difficult.

 

[marciolsf]

the 6.3V are way taller than the 2.5V which makes fitment in with the shields difficult.

That's what I've been noticing. I have to buy another batch of tantalums to finish this project, but I'm thinking of buying 2.5V ones, hoping they'll be easier to install. Should I replace all my 6.3v with 2.5v though? Or do I have to worry about mixing voltages if I don't, even when then have the same 470uf capacitance? It's been a very long time since I've had to calculate voltage on serialized components.