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

Just picked up another ps3 cechc03 in my country for €20 it have ylod and no hard drive since i have ordered 10 new original nec tokin i could try and use 8 on mine and 2 on the other ps3 thing is to figure out which needs replacing there was a video exlaining how to test nec tokin caps but it in foreign language
 
The deal fell through the guy who i was buying the cechc03 ps3 from wont answer my calls and messages so i might well find another seller for cechc03 model through ebay
 
None of the repair shops in Germany have caught up to this permanent fix. They all still list reballing and reflowing. :(

They wont yet until this fix is confirmed or there is a better way of testing these NEC tokin caps have in fact failed. only sure way of knowing is by removing them properly with out damaging them to know for sure which will be rather costly for a consumer. This method is a stab in the dark to try fix the issue and since we are all damaging them upon removal we dont know exactly which caps were at fault.
 
They wont yet until this fix is confirmed or there is a better way of testing these NEC tokin caps have in fact failed. only sure way of knowing is by removing them properly with out damaging them to know for sure which will be rather costly for a consumer. This method is a stab in the dark to try fix the issue and since we are all damaging them upon removal we dont know exactly which caps were at fault.

Quite frankly I don't care. I would gladly pay somebody to replace ALL of the caps preemptively without already having YLOD.
 
Quite frankly I don't care. I would gladly pay somebody to replace ALL of the caps preemptively without already having YLOD.

U dont really need to go find a console repairer to do this replacement. Anyone in the IT repair field should be able to do it for u, or anyone that repairs electronics such as TV's etc.... could even do it, replacing them is the same principle. you would only reply on the console repairer more so they get the thermal paste done right thats about it.
 
U dont really need to go find a console repairer to do this replacement. Anyone in the IT repair field should be able to do it for u, or anyone that repairs electronics such as TV's etc.... could even do it, replacing them is the same principle. you would only reply on the console repairer more so they get the thermal paste done right thats about it.

Thanks for the tip.
 
Super slims are not prone to this issue, so no.

Thanks for the info DoublesAdvocate, do you happen to known if there are any repair tutorials or info about the ps3 super slim not turning on? if I press power, the light goes green for 3 to 6 seconds then goes red light.. i tried replacing the power supply unit but still the same
 
Only way is to take it all apart and see if anything is blown on the motherboard as you need a dmm to check it is my guess
 
I was checking back in to do some data collection and start a spreadsheet, but some weird comments led me to some experiments that I hope will put this all to rest. This post will be full on drama. I will follow it with a separate post that is solely what I hope will put an end to this, so skip straight to that for results instead of my being defensive.


doubles 1: "I have no vested interest in reballing" reply

I repair electronics for a living, and have been doing it for 15 years. A grand on equipment is not a big deal. My instagram was often an outlet for frustration, so sometimes I used it to rant. I also change my mind when confronted with new information (which is going to be reallllllll important in my next post). That's something I hope everyone here would do. I've recently changed my mind about reflowing consoles that are almost 15 years old because the oxidation has made it a useless endeavor that won't survive the generous warranty periods I provide.

1b) Yeah, I've repaired something like 20 thousand game consoles in my time. A few hundred PS3 systems does not make it my bread and butter. I'm really only trying to nip this rumor in the bud before it becomes extra work that I have to do without being able to charge extra money for it. I don't repair consoles for people, I sell repaired consoles. I do some local contracts, but they are bulk and have a per console charge regardless of what is wrong. Again, I've provided you with my social media and eBay, you can look it up. I don't offer repair services. I sell the finished product, so I don't give a crap how it gets fixed. I just sell it fixed.

2b) Yep. Hundreds of PS3 systems are literally a drop in the bucket. You obviously looked at every single instagram post and current eBay listing but still purposely misconstrued that. I invite anyone else to look. I'm happy to show you a picture of my inbox of consoles right now, there are about 200, and less than 10 are PS3.

2.) First of all, that's just some dude's powerpoint, not an academic paper. Second of all, I think you've entirely misunderstood it. My understanding of reading that same slide is that they're saying you're not allowed to use a BGA package in an automotive application unless it can be shown to have no failures after 3000 heat cycles. Automotive standards are entirely different from the rest of the world. If you ever buy electronic components, look in the options on like Digikey or your favorite. There is a dropdown to pick automotive components. I feel like you're willfully misinterpreting what that sentence says, but I'm open to argument there.

3.) My bad. I am covering a lot of ground, so I've gone a little scatterbrained. There are defects from the factory that lay in wait, there are defects that form. All of them come to fruition through heat cycles - why I keep saying results of one day of playing Forza don't matter. I don't really understand your point of bringing this up, but I've been drinking.

4.) I literally said I have no idea, but you're still making a point to attack this. I don't know what to tell you. I have very, very little data for slims so I'm not going to argue about them. I want hundreds of data points.

5.) k

6.) Rule of 10's is a pretty widely used term in the industry for back of the envelope calculations regarding capacitors, but it can be used for other guesswork. I thought I was clear on that, but again I'm covering a ton of ground here and I'm scatterbrained. There is no real citable science to it, it's just a tool for guesswork.

7.) You were SOOOOO close to understanding how I'm in business after 15 years, while everyone else closes up shop and sells their inventory to me after they go out of business. I love DS Lites. I can sell them for $25 plus shipping all day every day. I get them in bulk 20 at a time for around 5 bucks each, they usually need around 5 bucks in parts, and I pay around 3 bucks in fees to sell them. $12 bucks profit is HORRIBLE, right? BUT! I can have one repaired, cleaned, tested, and listed in about 20 minutes (I'm happy to provide a video speed run, I'll go for the world record). Are you REALLY going to look down on $36 an hour?

I buy anything when the price is right. I'll repair LeapFrogs all day if the money is there. I bought some Super Slims in 2016 (you conveniently left out the date on the picture). Some of my contracts have sent one from time to time, too. You did it, you caught the tater.

8a. but not numbered) You might want to look up what computer engineering is again, because not only have I now been mocked for being the only person to show up to an argument about capacitors with a capacitance/ESR meter, you're now mocking me for having a relevant education. Most of the core curriculum to any CPE degree is literally the design of circuit boards and the components that go on them.

8b. but not numbered) Yeah, it's a home based business. I have made no attempt to hide that, I OFFERED you my instagram. I even bought a bigger house about 5 years ago just so I could have an 800 square foot dedicated space for my workshop with a finished area and my main workbench sitting next to a giant double door full of sunshine and happiness right next to my garden. I don't really see how that's relevant except that you're trying to be petty or jealous about me chilling in a stack of video games at home all day as a job.

8c. but not numbered) I love my ACHI station. I bought it because it was a great value, it was well documented, and it is made almost entirely out of "off the shelf" components, so it is infinitely repairable. You've apparently never dealt with a rework station, but they.... uh.... get really hot. It's kind of in the job description. And as we've been discussing, that leads to failure. Since there's nothing proprietary in it, any failure is easy to diagnose and cheap to repair. I'm super happy with it. And since I didn't waste money on pointless upgrades like spending out the ass to get auto part placement, split vision, touchscreens, and other bells and whistles, I can spend that money on other fun toys. Like my FLIR camera that I used to verify completely even heat (literally the only parameter that matters in a rework station) or getting authentic Omega thermcouples to verify accurate temps.

I spend money where it matters. Again, you clearly looked at every single instagram post I ever made, but while criticizing my equipment you purposely left out explaining where I do spend money. For example, my JBC soldering station that I'm willing to bet each single tip costs more than the average entire soldering station being used in this thread. My work spaces are insured for 100K if you'd like a tour to try to make fun of the rest of my equipment.

8d but not numbered) Geez, you're really reaching here. I built a fume hood, fvck me right? I work with this crap for a living, so every little fume needs vented or I'll be dead before 50. I have sensitive ears, so the Hakko FA-430 unit that everyone uses that runs at 50 decibels was right out. The JBC FAE1 system runs at 55 decibels. Again, not an option. I have at least 2 different ventilation systems running at all times, sometimes 4. I spent $1200 on Panasonic bathroom whisper fans that run at 20 decibels each. Sue me for being practical, mildly autistic, and not caring about how good it looks. I also never have to pay for filters, either.

9 but not numbered) Yes, the TOKIN massacre. As I've done hundreds of PS3 systems, it's bound to actually be an issue once in awhile. Another way to look at it is by actually including the date of the post: a year ago. As in, I've been gathering data on a known issue for over a year and have entirely dismissed it for the time being.

10 but not numbered) I apologize for showing up to the thread already angry, but there were 20 some pages of someone essentially mocking my viewpoint before I even arrived. I'm childish and I make no excuses for it, and it's half the reason I decided to work for myself. If you can't see past that when I'm providing actual evidence, you're being equally childish. You can see my receding hairline on instagram and the date on my degree. Do the math.

11 maybe? Finally, again, I don't actually offer repair services. This is why I'm saying I don't care how these get fixed, just that they get fixed. I only do full service complete refurbishing, then give systems a complete warranty. Buy then sell. Anything else and you have to deal with lying customers and being blamed for unrelated issues, so I just avoid it entirely. Anyone here that has fixed a computer for someone has heard a sentence that starts with "ever since you fixed...." and I want no part of that garbage.

For an idea of what I charge/make on a backwards compatible PS3: I buy them with warranty seals for 60 or 70, without for 30 or 40. I sell them for around 400. About a third of them are dead beyond repair, with an average of 2 hours of work before discovering that. It takes me a week of on/off work to finish them fully stress tested. It takes about 8 hours of hands on work to restore them to my standards. You checked my listings, you see that I even take the fans apart and regrease the bearings. I am thorough. I do 10 heat cycles, at least 1 overnight. I give a 6 month warranty, and I am VERY generous when they are out of warranty. Since you already stalked my eBay listings, you can see I sold 2 CECHA01 in the last 10 days.

Congratulations to anyone that finished.
 
"They wont yet until this fix is confirmed or there is a better way of testing these NEC tokin caps have in fact failed. only sure way of knowing is by removing them properly with out damaging them to know for sure which will be rather costly for a consumer. This method is a stab in the dark to try fix the issue and since we are all damaging them upon removal we dont know exactly which caps were at fault."

I already did that multiple times and was entirely ignored.




Anyway, something about this whole thing hasn't made sense to me from the beginning and I just didn't put my finger on it until I saw the comment from someone about "charging up" the TOKINs, and nobody felt the need / understood how to correct them. Then I looked back at the original post. Dude literally says the TOKINs need to provide current. That's so wrong that while trying to think of how to explain how wrong, I realized that there was a really simple way to prove my point. (Also easily reproducible for anyone that thinks I'm lying.)

They're high speed low ESR decoupling caps that charge and discharge small amounts thousands of times a second. They don't "supply" power at all. They are filters that smooth transient voltage from switching power supplies, and handle small power spikes from other devices on the same power rail. In other words, when things are functioning as designed in a well designed circuit, they are BARELY needed.

You can actually entirely remove some filter caps from circuits, and not notice a difference in standard operating conditions.

So, If I'm right: not only can the TOKINs be WAAAY WAAAAY WAAAAAAAAY out of spec, like literally dead before it makes a difference.... I could probably just rip them off the board and things will be fine. Everyone arguing against me is telling me that if they're like 1% out of spec, the system will start throwing up Christmas trees.

Well, just today, I was tearing down a non-BC system that had a dying flash and was missing the drive. The perfect candidate. No YLOD, no BGA issues, no TOKIN issues.


I make it all the way up to 5 COMPLETELY MISSING TOKIN caps before it has the slightest problem. Still even boots missing 5, but only crashes when it finally tries to display graphics.

That's over 60 times the amount they want you to believe is problematic, and it still doesn't even go YLOD.

You're inadvertently doing "the towel trick" and getting lucky.
 
Last edited:
"They wont yet until this fix is confirmed or there is a better way of testing these NEC tokin caps have in fact failed. only sure way of knowing is by removing them properly with out damaging them to know for sure which will be rather costly for a consumer. This method is a stab in the dark to try fix the issue and since we are all damaging them upon removal we dont know exactly which caps were at fault."

I already did that multiple times and was entirely ignored.




Anyway, something about this whole thing hasn't made sense to me from the beginning and I just didn't put my finger on it until I saw the comment from someone about "charging up" the TOKINs, and nobody felt the need / understood how to correct them. Then I looked back at the original post. Dude literally says the TOKINs need to provide current. That's so wrong that while trying to think of how to explain how wrong, I realized that there was a really simple way to prove my point. (Also easily reproducible for anyone that thinks I'm lying.)

They're high speed low ESR decoupling caps that charge and discharge small amounts thousands of times a second. They don't "supply" power at all. They are filters that smooth transient voltage from switching power supplies, and handle small power spikes from other devices on the same power rail. In other words, when things are functioning as designed in a well designed circuit, they are BARELY needed.

You can actually entirely remove some filter caps from circuits, and not notice a difference in standard operating conditions.

So, If I'm right: not only can the TOKINs be WAAAY WAAAAY WAAAAAAAAY out of spec, like literally dead before it makes a difference.... I could probably just rip them off the board and things will be fine. Everyone arguing against me is telling me that if they're like 1% out of spec, the system will start throwing up Christmas trees.

Well, just today, I was tearing down a non-BC system that had a dying flash and was missing the drive. The perfect candidate. No YLOD, no BGA issues, no TOKIN issues.


I make it all the way up to 5 COMPLETELY MISSING TOKIN caps before it has the slightest problem. Still even boots missing 5, but only crashes when it finally tries to display graphics.

That's over 60 times the amount they want you to believe is problematic, and it still doesn't even go YLOD.

You're inadvertently doing "the towel trick" and getting lucky.
What you trying to prove here as some of us find a reason to fix something here and you had to go even the extra lenght to say all that i just fell asleep after the first 3 lines
 
@squeept What happens if you short one NEC out (simulating it's gone to crap making a short instead of decoupling), will it give the infamous YLOD pattern? Just curious.

Also, what happens if you start pulling the bottom ones too? For the sake of your argument you might be able to pull all of them w/o YLOD, THIS would be epic.

I think we all could use each others help and findings here since there are evidence pointing towards NECs going to crap, which for some users has been resolved in conjunction with replacing them.
 
Last edited:
"They wont yet until this fix is confirmed or there is a better way of testing these NEC tokin caps have in fact failed. only sure way of knowing is by removing them properly with out damaging them to know for sure which will be rather costly for a consumer. This method is a stab in the dark to try fix the issue and since we are all damaging them upon removal we dont know exactly which caps were at fault."

I already did that multiple times and was entirely ignored.

I make it all the way up to 5 COMPLETELY MISSING TOKIN caps before it has the slightest problem. Still even boots missing 5, but only crashes when it finally tries to display graphics.

That's over 60 times the amount they want you to believe is problematic, and it still doesn't even go YLOD.

You're inadvertently doing "the towel trick" and getting lucky.

well to be frank u have been ignored because uve gone about this the wrong way, if u came to this thread with a different approach im sure the responses would be different, but instead your personality got the better and made u out to be a D**K, sorry, but true.

Im still sitting on the fence at this stage still so i can see both views to work out solutions to these repairs, i come from the same repair back round so i can afford to do that. So far i haven't had any luck with this recent find so i cant back up this repair yet but still it isnt a repair that so far u cant rule out.

Since u pulled the 4 caps from the board and the 5th didnt work, we need to see this mother board (or similar) that works and put it under load, so far ur videos show no load on the BGA's yet so id imagine YLOD will come more premature so . lets prove this?

BTW who drinks lite beer? hahaha thats a bit soft lol


What you trying to prove here as some of us find a reason to fix something here and you had to go even the extra lenght to say all that i just fell asleep after the first 3 lines

come on mate, i know the guys got off on the wrong foot here but u gotta at least show some respect since he has put in more effort than most to show what he has so far. u dont need to bring urself down to that same level.
 
Last edited:
Edwired - you are everything that's wrong with the world.

Jacobsson - as I covered in my previous rants, that is not the failure mode of a polymer capacitor. The self healing properties cause any internal resistive path to be immediately burned off until open. They fail open, not short.

I am mildly curious what happens if I straight up short it just for shits and giggles, and I don't care about this board, so maybe I'll get back to you in a bit. I'm guessing that I hit the button and the lights go out altogether without fanfare until unplugged.

wrx884 - This board has a failing NOR, so I doubt I can get it to a point where it will play games. It would obviously make sense from everything I've laid out if failing TOKINs were responsible for some systems that were having random shutdowns due to instability under load, but the entire theory I'm debunking is about YLOD from the initial boot, and I think that what I posted by itself thoroughly debunks that.

Everyone is claiming it's due to lost capacitance, and I can't think of a greater loss of capacitance than losing the capacitors altogether.

If the goalposts are going to be moved so far at once, I'm happy to buy any meter or test equipment I don't already have to close down whatever avenue an entirely new and different argument is gonna go down, but someone is going to need to articulate exactly what characteristic of the caps is causing problems.
 
Last edited:
Edwired - you are everything that's wrong with the world.

Jacobsson - as I covered in my previous rants, that is not the failure mode of a polymer capacitor. The self healing properties cause any internal resistive path to be immediately burned off until open. They fail open, not short.

I googled around for support of this claim and I found that do tend to short, for example see this reference from Vishay.com:
http://www.vishay.com/docs/42106/faqpolymertantalumcaps.pdf
The answer to the very last question is actually claiming short is a "common failure" for polymer tantalums. I'm far from an expert so I can read this wrong.

Not trying to break your balls or anything, just saying they tend to fail differently.
 
Hmm, couldn't tell you. I can't find anything that tells me exactly what kind of polymer the tokins are, and it looks like every one has a different failure mode. All I can say is that I pulled a dozen from YLOD consoles and they all had 0.0 ESR and read open for normal resistance, exactly as they should. I'll go short one out after breakfast.

edit: immediate YLOD in under a second when totally shorted with a jumper wire, I assume there's no need for a video. I'll be happy to go back to pulling them and showing the meter readings all day to prove that isn't what's happening, though.
 
Last edited:
Back
Top