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

[RIP-Felix]

Also, these will work too. PANASONIC EEFSX0D331E4. They're 330ohms, so you would need to use 4 to replace each tokin, but they're much cheaper than the tantalum. you can get 32x $8.

I feel it necessary to point out that these are not quite as good as the tantalum caps. But they are similar to what SONY switched to on the slim models. They are similar to the Tokins. But the tokins used a hybrid solid silver / electrolyte paste, instead of full polymer. So these are more reliable. They just don't reject as much of the high frequency noise as do either the tokins or tantalums.

But they work and are cheap.

 

[Computer Booter]

Also, these will work too. PANASONIC EEFSX0D331E4. They're 330ohms, so you would need to use 4 to replace each tokin, but they're much cheaper than the tantalum. you can get 32x $8.

Wouldn't these EEFSX0D331E4's actually be better because they have 4.5mohm rating instead of the 2R5TPE470M7's 7mohm rating?

When I DO order these, it will be in bulk so I want to order the "best" ones. I don't mind installing more of them at this point. With the grinding pen I bought it's a breeze now.

EDIT: OH I just read your most recent post where you say that the Tantalum 2R5TPE470M7's are better... so I guess that answers that?

 

[Workz_777]

@RIP-Felix ...i need your help, OSH Park just refunded my order for the Tantalizers 0.3b version. They said there was an error manufacturing them. They said version 0.5 has better manufacturing results, but i can't find this one. Please could you upload the file here for version 0.5, or link to the OSH Park one? It would be very much appreciated.

 

[RIP-Felix]

@RIP-Felix ...i need your help, OSH Park just refunded my order for the Tantalizers 0.3b version. They said there was an error manufacturing them. They said version 0.5 has better manufacturing results, but i can't find this one. Please could you upload the file here for version 0.5, or link to the OSH Park one? It would be very much appreciated.

You are now the second person to have this problem. I've been in contact with OSH park to see what the issue was and have been working on revisions to fix it.

I just ordered a test batch of 0.5a. In fact your 0.3b may have been on the same panel, given the timing. They said they could only ship 5 of the 9 I ordered. So aparently some of mine failed too. But they didn't provide me any details so I can fix the issue. I sent them an e-mail asking why they're failing and what I can do to prevent it. Until I know more, I can't fix it.

I haven't recieved them yet. I'm concerned they may have mixed the orders up. Perhaps I should silkscreen the revision on them to avoid confusion...IDK.

This is anoying! My test batch of 0.3b were perfect and I like that design best. The revisions I've since made have been sacrifices to try and improve manufaturing reliability.

I don't even know if it's just me or their router needs a new drill bit. Why all of a suddon are they failing?

 

[Workz_777]

You are now the second person to have this problem. I've been in contact with OSH park to see what the issue was and have been working on revisions to fix it.

I just ordered a test batch of 0.5a. In fact your 0.3b may have been on the same panel, given the timing. They said they could only ship 5 of the 9 I ordered. So aparently some of mine failed too. But they didn't provide me any details so I can fix the issue. I sent them an e-mail asking why they're failing and what I can do to prevent it. Until I know more, I can't fix it.

I haven't recieved them yet. I'm concerned they may have mixed the orders up. Perhaps I should silkscreen the revision on them to avoid confusion...IDK.

This is anoying! My test batch of 0.3b were perfect and I like that design best. The revisions I've since made have been sacrifices to try and improve manufaturing reliability.

I don't even know if it's just me or their router needs a new drill bit. Why all of a suddon are they failing?

Thanks Felix, yeah last time i ordered a few and it was fine, this time when i ordered more there was a problem. It's annoying because we wait almost 1 month to find this out. Tuesday they emailed me to say the total amount i had ordered was shipped. Then a day later (Wednesday) they tell me nope there was a problem and i have been refunded, huh?

Here is what they emailed me yesterday in case it helps:-

From OSH Park:- "I issued a refund to cover the missing boards from this order, because we pretty much always have manufacturing issues with the 0.3b version of this board. If I were to remake this order, most would probably fail. I'm not sure what revision they're at currently, but I've seen 0.5 seems to work better, but still isn't a lot of room for tabs. No tabs, or tabs that are too small, means they fall out of the panel during manufacturing. The substrate being thinner only exacerbates this problem. If there's a newer revision, you're probably better off ordering the newest one available."

 

[RIP-Felix]

From OSH Park:- "I issued a refund to cover the missing boards from this order, because we pretty much always have manufacturing issues with the 0.3b version of this board. If I were to remake this order, most would probably fail. I'm not sure what revision they're at currently, but I've seen 0.5 seems to work better, but still isn't a lot of room for tabs. No tabs, or tabs that are too small, means they fall out of the panel during manufacturing. The substrate being thinner only exacerbates this problem. If there's a newer revision, you're probably better off ordering the newest one available."

Yes that does help, thanks. And that's basically the same thing they told me. In 0.5a I removed the long GND slots on the side, instead making 2 smaller slots with room in between for tabs. And I shortened the + rail slots so now they don't go all the way to across (a feature which I really liked about 0.3b). So now the corners have more support. I was hoping those 2 changes would be enough.

I have been waiting for them to arrive to before I release 0.5b. But now I'm thinking I'll have to make another revision and wait. I'm debating whether or not I should close 0.3b so people don't order them.

 

[Workz_777]

Yes that does help, thanks. And that's basically the same thing they told me. In 0.5a I removed the long GND slots on the side, instead making 2 smaller slots with room in between for tabs. And I shortened the + rail slots so now they don't go all the way to across (a feature which I really liked about 0.3b). So now the corners have more support. I was hoping those 2 changes would be enough.

I have been waiting for them to arrive to before I release 0.5b. But now I'm thinking I'll have to make another revision and wait. I'm debating whether or not I should close 0.3b so people don't order them.

Awww that's a shame, i like the 0.3b design, i have installed a set of 8 x Tantalizers (0.3b) with all the MLCCs, they look really nice!

 

[mekmitasdigoat]

Hey @RIP-Felix I have a question. So for the tantalizer you recommended several X5R MLCCs for the array. I am wondering if there's any reason to pick X5R instead of X7R.

Sorry I'm not a pro on those electronic components so this might be a dumb question, but from what I searched and saw online, seems the X7R has a wider working temperature range up to 125C where the X5R ranges up to 85C. Given that the CPU or RSX on the default fan curve could get to this temperature somehow, is it better to use X7R MLCCs to give some tolerance?

 

[RIP-Felix]

Hey @RIP-Felix I have a question. So for the tantalizer you recommended several X5R MLCCs for the array. I am wondering if there's any reason to pick X5R instead of X7R.

Sorry I'm not a pro on those electronic components so this might be a dumb question, but from what I searched and saw online, seems the X7R has a wider working temperature range up to 125C where the X5R ranges up to 85C. Given that the CPU or RSX on the default fan curve could get to this temperature somehow, is it better to use X7R MLCCs to give some tolerance?

I modeled the simulation using X5R because that's all the K-sim had. I linked to some caps, but they're just examples. I don't necessarily recommend those specifically. X7R have a wider temp specification, sot that's not a bad choice.

As far as the MLCC selection goes it's up to you to experiment.

 

[MarcoAG11]

Yes that does help, thanks. And that's basically the same thing they told me. In 0.5a I removed the long GND slots on the side, instead making 2 smaller slots with room in between for tabs. And I shortened the + rail slots so now they don't go all the way to across (a feature which I really liked about 0.3b). So now the corners have more support. I was hoping those 2 changes would be enough.

I have been waiting for them to arrive to before I release 0.5b. But now I'm thinking I'll have to make another revision and wait. I'm debating whether or not I should close 0.3b so people don't order them.

Hey there!

So, I've been scanning this forum for a bit now without ever posting, and I figure I should go ahead, make an account and contribute at least in some small way.

I also ordered some of those tantalizers about a month ago and had them come back at me with a rejection letter. When I asked them why I've seen them in the wild they told me that someone there modified them to work, and then they sent me a zip file with the version they got to work. I didn't bother to make an account because just as I got that email, you had posted that you were working on a new beta. As it seems that there is still trouble, maybe this file can help you? This is what they told me:

"I talked with a colleague and we suspect the ones that make it through manufacturing had some human intervention to make them work. Else it's just dumb luck.

Said colleague modified the files to help them be more manufacturable, as I'm just a shipping guy
If you have any way to let the author of the project know, removing the drill slots should be all they need to do to make it work just fine.
I've attached the modified files without drill slots."

Let me know if you would like me to forward you the email. I don't know if it's possible to DM in this forum, or how I can get you the file, but I'm more than happy to get it to you.

 

[bguerville]

Hey there!

So, I've been scanning this forum for a bit now without ever posting, and I figure I should go ahead, make an account and contribute at least in some small way.
...
Let me know if you would like me to forward you the email. I don't know if it's possible to DM in this forum, or how I can get you the file, but I'm more than happy to get it to you.

FYI
To counter spamming, a new member's account has limited permissions & posts are submitted to moderation team approval.
Those restrictions will be lifted automatically after a few approved posts & you will be able to use all the direct messaging features.

 

[Djudju]

There is nothing wrong to excede the Factory Value, in fact for the Phats you need more than 1200uF, for each NEC/TOKIN, in fact your PS3 thanks you, i have 4x 470uF for each NEC/TOKIN,making a total of 1880uF per-NEC,never had instabilities or shutdown afterwards, the most important thing is not to go below the original NEC/TOKIN Spec of 1200uF, going lower that value,can give you a YLOD again,as one of the Chips are not being feed properly,but thats up to you which value you want to use,if its 330 or 470uF.

Hi,
I have a question for you experts. I replaced the 2 nec tokins OE108 rsx (6 tantalum 470uf per nec) on top of the motherboard (these are the ones recommended to be changed first) because it wouldn't turn on. Now it starts very well. The only problem is that most games, after 5-6 min, the console turns off with 3 beeps. FYI, it's a phat CECHL...
Should the nec tokin CPU also be changed? Thanks very much

 

[truemaster]

is the fan working? does it power on soon after is shutdown?

 

[RIP-Felix]

I decided to stop sharing the Tantalizer v0.3b.

v0.5b is slightly more manufacturable, but 3 of the 9 I ordered from the test batch failed So I'm going back to the drawing board and will release a new version.

...until then the Tantalizer is on hiatus.

 

[marciolsf]

Hi,
I have a question for you experts. I replaced the 2 nec tokins OE108 rsx (6 tantalum 470uf per nec) on top of the motherboard (these are the ones recommended to be changed first) because it wouldn't turn on. Now it starts very well. The only problem is that most games, after 5-6 min, the console turns off with 3 beeps. FYI, it's a phat CECHL...
Should the nec tokin CPU also be changed? Thanks very much

you might be overheating, or it might be some other issue! Since you're able to boot, if you have cfw/hen, you can get the error log via PS3 advanced toolset (https://github.com/bucanero/psl1ghtv2_ports/releases), which is much easier than using the dongle…

EDIT: I listed the wrong tool, it's not webman mod, it's PS3 Advanced toolset. My bad!

 

[RIP-Felix]

you might be overheating, or it might be some other issue! Since you're able to boot, if you have cfw/hen, you can get the error log via webman mod, which is much easier than using the dongle…

Wait...webMAN mod can do the SYSCON errorlogs now?

I did not know that. I am only aware of PS3 advanced toolset

 

[marciolsf]

Wait...webMAN mod can do the SYSCON errorlogs now?

I did not know that. I am only aware of PS3 advanced toolset

lol, maybe I'm misremembering! That's probably why I couldn't find that feature on wmm yesterday, I was thinking maybe i was missing a module or something.

 

[RIP-Felix]

I would like it to be added to webman, that would be convenient

 

[marciolsf]

I edited my previous post to correct the tool name! I also went ahead and did a syscon error log dump of my 2101a slim, since it's been behaving oddly lately. Temperatures haven't been super high (RSX on low to mid 60's with wmm temperatures set to auto #2), but XMB has been blinking out for a 1-3 seconds off and on.

So here's the log output! Seems like I'm having video issues, with lots of 2021/2022. @022 is a DEV error, and we don't have a code for 2021 on the wiki yet.

Spoiler: slim error codes

Firmware Version: 4.88 (build 50731)
Platform ID: CokH11
Product Code: 00 84
Product Sub Code: 00 0A
Hardware Config: 4E00FFFF0603BC3C
Syscon Fimware Version: 08A0.0000000000000000 (EEPROM: 0000000000000000)
Bringup Count: 1311, Shutdown Count: 1297
Runtime: 66 Days, 20 Hours, 6 Minutes, 56 Seconds
Error Log
01: A0802022 Fri Mar 25 22:01:42 2022
02: A0802022 Fri Mar 25 22:00:23 2022
03: A0802022 Tue Dec 28 21:01:08 2021
04: A0802022 Tue Dec 28 20:55:43 2021
05: A0802022 Mon Dec 27 23:58:49 2021
06: A0802022 Mon Dec 27 23:57:55 2021
07: A0802022 Mon Dec 27 23:56:32 2021
08: A0802022 Mon Dec 27 23:55:07 2021
09: A0802022 Mon Dec 27 23:52:33 2021
10: A0802022 Mon Dec 27 23:52:31 2021
11: A0802022 Mon Dec 27 23:52:23 2021
12: A0802022 Mon Dec 27 23:51:10 2021
13: A0802022 Mon Dec 27 23:48:50 2021
14: A0802022 Mon Dec 27 23:44:32 2021
15: A0802022 Mon Dec 27 23:41:12 2021
16: A0802022 Mon Dec 27 23:23:09 2021
17: A0802022 Mon Dec 27 23:23:07 2021
18: A0802022 Mon Dec 27 23:22:59 2021
19: A0802022 Mon Dec 27 23:10:28 2021
20: A0802022 Mon Dec 27 23:10:26 2021
21: A0802022 Mon Dec 27 23:10:18 2021
22: A0802031 Sat Oct 30 20:22:51 2021
23: A0802022 Tue Oct 19 04:57:24 2021
24: A0802022 Tue Oct 19 04:52:58 2021
25: A0802022 Thu Oct 14 20:16:58 2021
26: A0802022 Thu Oct 14 20:11:44 2021
27: A0802022 Thu Feb 11 10:35:58 2021
28: A0802022 Thu Feb 11 10:34:56 2021
29: A0802022 Sun Feb 7 10:59:20 2021
30: A0802022 Sun Feb 7 10:57:04 2021
31: A0802022 Sun Feb 7 10:00:21 2021
32: FFFFFFFF Fri Dec 31 23:59:59 1999

 

[marciolsf]

I would like it to be added to webman, that would be convenient

I agree! I wonder if @aldostools thinks that's doable.