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

[Fallen777]

If people are happy to waste their time, without a proper path to follow, then, yeh sure, good luck to them!

For me i would rather follow a path that has some logic and proper diagnosis before attempting this kind of endeavour.

So yes, there is always two sides to a coin.

well theres no stopping anyone from trying to attempt repairs knowing the consequences or not.

 

[Fallen777]

This whole thread makes me so worried for those poor PS3 motherboards, being butchered without proper diagnosis!

well whats the point of quoting me in this i told him the right thing to do by using syscon diagnosing.
And if you are the one who started that syscon error diagnosing thread then you know to points that are to be solder with usb uart right.
So can you help both us by posting those points of COK-002 and DIA-001 and DIA-002 motherborad.
i am planning to buy a PS3 MGS4 edition in ylod.IT has a 65mn CELL BE and 90mn RSX is it a bad choice.
Its a rare one.

 

[db260179]

OK, chill, sorry didnt relise you where pointing him in the right direction.

Its in that thread first post, stored in the github repo of mine - the odf or pdf guide.

well whats the point of quoting me in this i told him the right thing to do by using syscon diagnosing.
And if you are the one who started that syscon error diagnosing thread then you know to points that are to be solder with usb uart right.

 

[Fallen777]

OK, chill, sorry didnt relise you where pointing him in the right direction.

Its in that thread first post, stored in the github repo of mine - the odf or pdf guide.

Its OK . Even i have just found the syscon thread few days ago @sandungas refered it to me.So i thought it was the right thing to do refer @Purple to it too.
I just wanted to know if the CECH-H003 is gonna be worth it like if i change capacitors will it die again.
Aslo please take a lool at my other thread PS2 - No display . Sound Only.

 

[Purple]

I just tried to boot my ps3 before attempting the syscon. It was yloding before, now it seems like PSU just goes to protection mode like it has shortage some where. I tried few times and it yloded instead of straight up go to protection. So question is will I able to run syscon to gather anyinfo in the case. Or if i desolder all the tantalum caps (in case any of them actually causing the short) will ps3 without tokins YLOD or just do not boot?

 

[Fallen777]

I just tried to boot my ps3 before attempting the syscon. It was yloding before, now it seems like PSU just goes to protection mode like it has shortage some where. I tried few times and it yloded instead of straight up go to protection. So question is will I able to run syscon to gather anyinfo in the case. Or if i desolder all the tantalum caps (in case any of them actually causing the short) will ps3 without tokins YLOD or just do not boot?

Those capacitors (either NECS/TOKINS or Tantulums depending on what you have)are responsible to provide stable electric current to both CELL BE and RSX.
If any of these capacitors are dead then ylod occurs but Althrough i doubt if without any capacitors won't boot rather then ylod.
Lets see what others have to say about it.

 

[marciolsf]

I just tried to boot my ps3 before attempting the syscon. It was yloding before, now it seems like PSU just goes to protection mode like it has shortage some where. I tried few times and it yloded instead of straight up go to protection. So question is will I able to run syscon to gather anyinfo in the case. Or if i desolder all the tantalum caps (in case any of them actually causing the short) will ps3 without tokins YLOD or just do not boot?

Yeah, even if all you do is connect the syscon in external mode, you'll be able to get the latest error code with the ps3 in standby. Going into internal/diag mode is a bit more work, but you'll be able to get a bit more information.

 

[Christian Duy]

CHOOSING THE RIGHT "2.5V/330uF & 6.3V/470uF" Tantalum Capacitors
and the Importance of ESR and other factors for this method,

(I'm in no way an electronics engineer/expert, I only want to share my experience to help others)

I've opted for a full NEC Tokin replacement on my CECHA01, my plan is to preserve this unit as long as I can, I ordered this from Japan for around 30USD with a YLOD Condition (Delayed YLOD, so I have high hopes for this one). Upon receiving the unit, I opened it and replaced one Tokin chip for the RSX, and the system booted. I was very excited so I tested every game to fully stress the chips.

Last of Us - YLODed
Gran Turismo 6 - YLODed
Beyond Two Souls - YLODed on game data installation screen

CPU/GPU Temps were around 68-70c around 35% fan speed (So temps aren't the issue)

From my investigation, some games YLODed on their loading screens even without stress to the RSX Chip yet, so I thought the CELL Processor needed some cap replacements too, after replacement, every game worked flawlessly. But I didn't stop right there, since I wanted this unit to last very long, it was time to get a full replacement of the NEC Tokin. I wouldn't recommend this to other people since it can add more hassle. (But it can be worth it, if done right)

So I ordered more "470uF - 6.3V" Caps, since that was the capacitors the main OP recommended.

After doing the full replacement, I noticed that the system won't boot anymore, it got stuck to YLOD again, this time it wasn't delayed, I got worried but with enough observation, I tried adding in more capacitors, 1 more to each NEC pads. so totaling to 40x 470uF Caps.

after doing all that, the system still was YLODing, but this time its delayed, so this means I was on the right track, (Do note I did not forget to add jumper wires, I only added two on one side, 1x for CELL and 1x RSX, i made sure that the gauge is thick enough to avoid overcurrent, more on this later..)

I added another 1x Cap to each token, this time totaling to 48 Caps, It was a gamble up to this point since I am adding more capacitance than the intended values.

But to my surprise, the system worked well! it played every game I mentioned very well with no YLODs. From this point on, I was prepared to be settled and assembled the unit all in great perfect condition, or is it...

A week later..

So I fired up the PS3 to play some sweet native PS2 games thru an HDMI outputted port (because hey why not, i shouldve just gone with slim if i wouldn't right?)

I noticed the system started YLODing again, and I was really worried, because it might've been the over-added capacitance, I did use 48 Caps lol. So I went investigating and redid the whole thing again, I added more jumper wires to each side, to evenly distribute the current, also to avoid burning the board with huge amperage. After finishing, still the same results. this time it was much worse than last week.

since I had an extra 52 "470uf - 6.3V" Caps from the bundled of 100pcs I ordered, i tried replacing them all again, and it works again just like the first timed I replaced it. BUT something was bothering me, what was the cause of the YLOD?

So I went online to investigate, first I tried checking the PDF spec sheet of the NEC Tokin 0E128, as it says right here, there are few variants of this chip.

As you've noticed here, I encircled the ones with the closest spec to the original PS3's NEC Token - 1200uF / 2.5V

So I started wondering why the OP recommended a 6.3V instead of the original 2.5V one? Upon checking previous posts on this thread, some say it shouldn't matter, since it would only limit itself on the voltage given by the board, so around 2.5V on a 6.3V capacitor.

How about capacitance? Surely more capacitance could ruin the board right? I don't think so either since I replaced the NECs with the proper amount of capacitors and the system won't boot, so I added more totaling to 48 Caps.

I tried considering other factors, I compared the spec sheet of my ordered caps to the original NECs, and there was definitely something wrong..

I noticed the ESR values (which is really crucial in this application since voltage ripple happens all the time when it comes to CPU/GPU chips) are way off in my ordered caps compared to the original NEC Tokens.

the NEC Token has "1.5mOhms - 2.0 mOhms" on its rated ESR, compared that to 1 sample of the cap i ordered, and BAM! 100mOhms per each cap.

Connecting them in parallel should reduce the ESR right? Yes, but the problem is each NEC Pad is supposed to have 1.5-2.0mOhms, and I have 100mOhms of each cap? so connecting them together would result in 100mOhms/4 caps = 25mOhms, still way far off the original intended values. I even added two more and it resulted in 16.7mOhms still far off, remember we need 1.5-2.0! Lesser the better! it avoids the strain of pushing the current to the caps from small voltage adjustments needed by the chips

Sample pic of the caps I first ordered (best to avoid this one):
TCJX477M006R0100

But why did it work in my case the first time?
I considered the heating of the caps changed its tolerances, in this case, testing the PS3 just right after soldering the caps in, and in a week after, the caps have cooled down resulting in a different tolerance on capacitance and even in its ESR thus the YLOD. (I can even bet all these caps i ordered are in good condition, just having different ESR Values) (ESR falls as the temperature increases)

Due to having higher ESR, this could lead the ps3 thinking the system is shorted, when the capacitors are not charged, it has a short amount of time of a zero resistance, until enough current pushed through then the resistance rises as the caps is charged. In this case, the ps3 won't continue to push more voltage/current since the caps are still in zero resistance state, thinking something might be shorted inside, thus instant YLOD.

I have found a lot more better caps with much less ESR values, up to just 10mOhms each, if divided by 4 capacitors, would result in 2.5mOhms! way much closer to the original.

So in order to fix this problem I have, I decided to order this one specifically:

https://www.mouser.ph/ProductDetail/Panasonic/2R5TPE330M9?qs=OE1iw1LrrPHigwp/AIDFFQ==

I ordered 40x 2.5V/330uF/9mOhms ESR Panasonic Caps, note I preferably want my voltages close to stock as possible.

These are decent Panasonic caps, which are designed to withstand high temps, this should be useful since the ps3 heatsink and the board are basically hot sandwich.

My chips should arrive on monday, hoping these will be the final fix on my system. Will provide more updates in order to help and guide others to order the right tantalum capacitors.


MY RECOMMENDATIONS!

I will be sharing a site which you can use as a reference if you ordered the right caps for your unit.

Link: https://www.mouser.ph/Passive-Compo...ors/Tantalum-Capacitors-Polymer-SMD/_/N-75hr2

(Note: It's up to the user to either order a "6.3V 470uF or a 2.5V 330uF", as long as the ESR is up to 9-10mOhms each)

Apologies for the long post, I wanted to share this to help people avoid ordering the wrong caps, since there are a lot of other factors of choosing the right one. Hope this helps!

im open for corrections if I said something wrong on this post!

 

[reapers2007]

so i have a cechj system which if left on xmb or watching movies will run all day, played black ops 1 campaign for an hour no problems, thought id play modern warfare 3 online and about 30mins in system shutdown red flashing light, played gt5 some times its good some times it red lights system never gets hot max temp 67c is this a nec/tokin issue or something else

 

[coleb87]

Hey guys, quick question - I'm in the process of removing the old caps.
On the actual solder pads is some sort of adhesive material that is very thin, yet very very difficult to remove. It's making desoldering extremely difficult! Anyone know how to remove this?

Sent from my SM-G965U using Tapatalk

 

[marciolsf]

Hey guys, quick question - I'm in the process of removing the old caps.
On the actual solder pads is some sort of adhesive material that is very thin, yet very very difficult to remove. It's making desoldering extremely difficult! Anyone know how to remove this?

Sent from my SM-G965U using Tapatalk

With a lot of swearing... it really helped

 

[marciolsf]

All joking aside, I destroyed most of the surface of the chip using a pick and and old iron tip. Then I just fed it a ton of solder, and slowly worked through removing the remaining bits. Slowly and carefully does it.

 

[coleb87]

All joking aside, I destroyed most of the surface of the chip using a pick and and old iron tip. Then I just fed it a ton of solder, and slowly worked through removing the remaining bits. Slowly and carefully does it.

Yes!!! It's a total PITA.

I got the chips off, looked at the soldering pads, said "huh, those sure don't look shiny - the hells that stuff!?"

Tried to desolder - lo and behold the adhesive caused the solder on my iron to either:
1) not flow at all
2) flow only to the bare corner, and SOLIDIFY despite my iron being directly on it (adhesive acting as a heat sink)

I suspect this "adhesive" acts as a pretty decent heat insulator (which as you may guess..... makes desoldering suck!)

Sent from my SM-G965U using Tapatalk

 

[marciolsf]

Yes!!! It's a total PITA.

I got the chips off, looked at the soldering pads, said "huh, those sure don't look shiny - the hells that stuff!?"

Tried to desolder - lo and behold the adhesive caused the solder on my iron to either:
1) not flow at all
2) flow only to the bare corner, and SOLIDIFY despite my iron being directly on it (adhesive acting as a heat sink)

I suspect this "adhesive" acts as a pretty decent heat insulator (which as you may guess..... makes desoldering suck!)

Sent from my SM-G965U using Tapatalk

That's exactly where I was! That is actually a bunch of layered strips. My process was:

• Clean the surface well with ipa
• Apply a bunch of flux
• Let the iron sit on top of the topmost strip, and as soon as the solder starts to go molten (it'll take a few seconds, so no fear), use a pick and gently try to work it underneath the top strip.
• Keeping the solder molten (it might take more flux), repeat the process until you extract the strip. Applying fresh solder will help with this. There might be more than one strip... If you get the solder molten again, you might feel it.

Extra tip: don't keep the tip at the same spot!! Move it back and forth along the strip -- it'll prevent heat spots and help the whole strip go molten.

 

[wrx884]

short answer is ur soldering iron isnt good enough to heat the larger grounding areas which is acting like a large heat soak. as above mentioned its the only other way around it but if u still have issues its ur iron and u need more heat to soak into the board for the solder to liquefy better, Make sure u use plenty of flux also. U can use a hot air wand or heat gun to heat up that grounding area that will help with the heat soak issue just dont blast it for too long in one area.

 

[MYS]

Yes!!! It's a total PITA.

I got the chips off, looked at the soldering pads, said "huh, those sure don't look shiny - the hells that stuff!?"

Tried to desolder - lo and behold the adhesive caused the solder on my iron to either:
1) not flow at all
2) flow only to the bare corner, and SOLIDIFY despite my iron being directly on it (adhesive acting as a heat sink)

I suspect this "adhesive" acts as a pretty decent heat insulator (which as you may guess..... makes desoldering suck!)

Sent from my SM-G965U using Tapatalk

Use a penknife and "peel" it off, there is a small "step" which you can feel when you pass your penknife from the edge moving inwards. This is the part you need to push slightly and "peel" off the thin layer of metal. It can be removed easily, just make sure your penknife is sharp.

 

[vinhtq115]

Can I replace the Nec/Tokins with 470uF 2.5V tantalums (2R5TPE470M9)?

 

[Purple]

Yeah, even if all you do is connect the syscon in external mode, you'll be able to get the latest error code with the ps3 in standby. Going into internal/diag mode is a bit more work, but you'll be able to get a bit more information.

I tried to run the syscon. but all i get is "answer length"

 

[Purple]

Here is how I connected the syscon. Grounded with grey cable to body.
When i do that and connect TXD and RXD, the converter ground LED goes ON even without pluging it to usb.

Pins on the converter from top to bottom based on the picture below are (+5v, Gnd, rxd, txt, 3.3v)

so in my case Blue is connected to TxD, Purple to Rxd, Green to Diag

Also my ps3 acting like this now. So i suspect something shorted and psu straight up goes to protection.

 

[jimftr]

at the moment I don't have the console at home yet but it should be a similar thing

Would this not be the ideal solution? Rather than removing the nec tokin, simply adding tantalums to piggyback onto them?

I would assume if you replace all 8 NEC/tokins, mlcc's would be an obvious important requirement for long term use.

However, surely the original NEC chips provide the functions that mlcc's would, and adding tantalums would reduce the stress and strain on the original NEC's? Therefore significantly reducing chance of failure?

It would also mean you wouldn't need to butcher your board by removing them.

Can anyone who is much smarter than me explain why this method wouldn't be the obvious choice for the less experienced diy'er?

I could replace half the NEC tokin's (2 per each side) for tantalums, but if this method shown here is just as effective, then it makes more sense to me, but please, if I'm wrong, please educate me, my knowledge on this is quite poor so it would be great to hear what someone else thinks.