Question about 85c caps vs. 105c caps

[Wester547]

Re: Question about 85c caps vs. 105c caps

Well, I'm gonna try replacing the CMOS battery first, but thanks for the advice. I was also wondering... if anyone had any info about Hitano caps? I was wondering, knowing they're not too common a brand (but I haven't heard good things about them), if there's an inherent defect in their capacitors since I've heard of older Toshiba laptops going down with them and one of the laptops I have is a Toshiba M55-S3293 (I read on these forums that an M60 and M65 died with those capacitors).

[mariushm]

Re: Question about 85c caps vs. 105c caps

You mean http://www.hitano.com.tw/home.html ?

[Wester547]

Re: Question about 85c caps vs. 105c caps

Well, I read they were Hitano capacitors (from Taiwan?), but I suppose those would be the ones. I think they're 470uf, and may look turquoise/gray.

[PCBONEZ]

Re: Question about 85c caps vs. 105c caps

Hitano are mediocre.
I've personally never had trouble with them but other people have.
[For a while I mistakenly thought they were Japanese because of the name, then I saw some complaints and looked them up.]
Better to use Jap caps.

The CMOS battery has nothing to do with spontaneous start-ups from standby. (aka 'Soft-Off')
It simply won't cause that because in Standby the battery is not the CMOS's power source.


Is uncommon but bad small caps can cause spontaneous start-ups and kill a CMOS battery.
[Although I think your battery is just old and used up. - A coincidence...].
.

[Wester547]

Re: Question about 85c caps vs. 105c caps

Do you know if Hitano caps are made with Taiwan/Chinese aluminum foil (which might expound upon why some are noting that they don't last too long)? Also, there are two Nichicon VR(M)s on the board rated at 85C, and they sit next to the RAM modules so I imagine that they could be a candidate for issue.

[PCBONEZ]

Re: Question about 85c caps vs. 105c caps

Do you know if Hitano caps are made with Taiwan/Chinese aluminum foil (which might expound upon why some are noting that they don't last too long)?

There's no way to find out short of asking them and even then how do you know they tell the truth?
[In general] Jap companies are king at QA. . CH & TW companies are king at pinching pennies.
Very unlikely they pay the cost to import the better Al from Japan, the US or the UK.

Also, there are two Nichicon VR(M)s on the board rated at 85C, and they sit next to the RAM modules so I imagine that they could be a candidate for issue.

All VR are 85C and any of them could have the problem.
There are MOSFETs all over a board and MOSFETs get hot.
So do chipsets..
So do video cards..
So do MOSFETs on other add-in cards.
.
I'm not about to Guess which ones might be the culprit on some setup I haven't even seen.
I don't even do that for VR if it's right in front of me.
Unexplained problem + VR = Replace the VR.
Small caps are cheap. Replacing them one at a time is a huge waste of time.
.
Replacements for VR don't cost much at all so if you are going to do any then just do all of them and be done with that problem.
Guessing (Easter Egging) just sets you up to order caps twice or more.
The shipping will probably be more than the cost of -those- caps to do the whole board plus it would add a week or two to the project.
.

[Wester547]

Re: Question about 85c caps vs. 105c caps

Thanks for the info.

Do you know if KZG capacitors die any quicker or slower than capacitors with P50/51 electrolyte or Chinese/Taiwan aluminum, or is all this way too dependent upon usage and heat levels to give an answer?

[PCBONEZ]

Re: Question about 85c caps vs. 105c caps

Thanks for the info.

Do you know if KZG capacitors die any quicker or slower than capacitors with P50/51 electrolyte or Chinese/Taiwan aluminum, or is all this way too dependent upon usage and heat levels to give an answer?

KZG are slower and it also seems to be heat dependent.
KZG near the chipset, a MOSFET or the CPU seem to go first.
.
A lot of brands got accused of having P50/51 when the real problem was probably low quality Aluminum.
Caps that actually had P50/51 often failed in 3 to 6 months if the system was run 24/7.
.
KZG isn't the only UCC series with that problem.
KZG, KZJ, TMV, TMZ, (and another TM_) all behave the same way and presumably use the same or a very similar electrolyte.
The TMx caps were special order and only sold direct to manufactures.
Asus used a lot of them (meaning TMx) and AOpen at least a few.
.

[Wester547]

Re: Question about 85c caps vs. 105c caps

If P50/51 fail that quickly, do they possibly fail quicker than Nichicon's shoddy run of HN/HM capacitors in the early 2000's, or is it again all relative of usage and heat?

[PCBONEZ]

Re: Question about 85c caps vs. 105c caps

Yes the P50/51 fail that quick.
That's why I said earlier that you aren't likely to see any, at least that aren't bloated.

The HN/HM issue isn't related and isn't that fast. [Just happened at about the same time..]
Sometimes you'll still see HN/HM from 2003-ish that haven't failed yet.
Those are usually in well cooled cases with low hours on the system.
Vast quantities failed in the Dell clam shell case systems at a year or two old.
Those mini systems don't cool very well.
.

[Wester547]

Re: Question about 85c caps vs. 105c caps

But I would imagine that the HN/HM issue would take a couple of years (of notable running time) or so to surface, and that the Taiwan aluminum problem might take a bit longer (though again, as you said, it's probably about how many power on hours and heat the system has been subjected to)?

[Scenic]

Re: Question about 85c caps vs. 105c caps

The TMx caps were special order and only sold direct to manufactures.
Asus used a lot of them (meaning TMx) and AOpen at least a few.
.

You can add a bunch of older ASRocks to that list too. I've got an ASRock K7VT4-4X where all the 6.3V 1000uF caps are TMZ's. None of them bad.. yet.
Seen a bunch of these back then and didn't replace them..
TMV's on the other hand.. most of the ones I've seen were bad already (popped the rubber bung out as they don't have a vent (4V 680uF))
The ones on my Abit motherboard are still OK (visibly at least). They're in parallel with chemicon polys though, so that might be part of the reason why they're not dead yet


edit: wait a second.. while I was looking at the 2nd photo again, I noticed a "Z" on the brown cap in the (blurry) foreground. Guess what? More TMZ's :O
Different style this time though. They look exactly like KZG's at first, which is what I thought they were the whole time

[Wester547]

Re: Question about 85c caps vs. 105c caps

^ The D845PEBT2 of mine also had PS Polymers from Chemicon, along with Nichicon HDs, which is probably why, in tow with the VRs, it isn't gone yet. But I think with great cooling that the board can still last a while.

[PCBONEZ]

Re: Question about 85c caps vs. 105c caps

But I would imagine that the HN/HM issue would take a couple of years (of notable running time) or so to surface, and that the Taiwan aluminum problem might take a bit longer (though again, as you said, it's probably about how many power on hours and heat the system has been subjected to)?

Why would you imagine that?
Different problems. - Different characteristics.
.
The HN/HM were supposedly overfilled so chemistry had nothing to do with it.
No room for expansion causes leaks and then they dry-out.
(There was a second HN/HM problem in 2004 I think but that was short lived.)
.
.
There are easily 100's and probably 1000's of cap manufacturers in CH and TW [some of which are very small] so any exception you can think of [good or bad] probably exists somewhere.
.
The Al used in TW comes from Chinese mining operations.
The impurities are natural - trace.
.
It's safe to assume most CH & TW caps use Al made from Chinese ore and so have those impurities but that doesn't mean they are all using the same quality end product by the time it's foil.
The best foil made from CH ore is worse than even average Japanese made cap foil because of the impurities and QA in manufacturing but not all foil made from CH ore is the same.
There are more than one producer of cap foils in CH & TW.
Companies that buy their foils don't always buy from the same place.
That means the caps they make aren't always going to be the same quality.

"Bad Cap Brand" doesn't mean the caps are ~guaranteed~ to go bad.
It means they are inconsistent and there is just no way to know if they will last or not.

That said, there are a few brands that historically are so horrible they get replaced on sight.
Which brands those are depends on who you talk to based on their own experience.
It's also vintage and device related because the most commonly used caps for an application changes over time.
Someone that usually does P-3 boards will have a different 'replace on sight list' than someone that does P-4 boards or PSUs or screens because they see different caps.
.

[PCBONEZ]

Re: Question about 85c caps vs. 105c caps

You can add a bunch of older ASRocks to that list too. I've got an ASRock K7VT4-4X where all the 6.3V 1000uF caps are TMZ's. None of them bad.. yet.
Seen a bunch of these back then and didn't replace them..
TMV's on the other hand.. most of the ones I've seen were bad already (popped the rubber bung out as they don't have a vent (4V 680uF))
The ones on my Abit motherboard are still OK (visibly at least). They're in parallel with chemicon polys though, so that might be part of the reason why they're not dead yet


edit: wait a second.. while I was looking at the 2nd photo again, I noticed a "Z" on the brown cap in the (blurry) foreground. Guess what? More TMZ's :O
Different style this time though. They look exactly like KZG's at first, which is what I thought they were the whole time

I've seen a number of TMZ not bloat but with a little leakage trail out the bottom.
.
There are three TMx I know of but the third one is pretty rare and I've only seen them one online.
Keep forgetting what series it is.
.
[Edit:] Oh...-- See the pic Bonez!
Yes TMZ, TMV, TMJ
From what I can tell each one only comes in one size.
Like: TMZ I have only seen in 6.3v 1000uF.
.
.

[Wester547]

Re: Question about 85c caps vs. 105c caps

I don't necessarily imagine that, because you're right, how effectual a capacitor is depends on how hard it is worked. I was just curious, but thanks again for the information. Another question... could bad PSU caps cause ripple current without any sort of sign (the system seeming absolutely fine, you would never know it)?

[Scenic]

Re: Question about 85c caps vs. 105c caps

From what I can tell each one only comes in one size.
Like: TMZ I have only seen in 6.3v 1000uF.

Now that you mention it.. yeah.. I've only ever seen 6.3V 1000uF TMZ in 8x11.5 .. no other values. Same goes for those 4V 680uF TMJ..

Weird thing is.. there are supposedly tons of different values and sizes out there.. lookie what I found:


What makes the whole situation even more complicated is that the only "common" TMZ's (6.3V 1000uF in 8mm) happen to be the only one that's missing in the PDF..

WTF Chemicon..!?

edit: according to info embedded in the PDF, it's from april 2004. In the PDF, the caps are described as having "white color letter on dark brown sleeve", so the ones on the ASRock board I posted a photo of must be even older than that, as they're black with golden letters

this is weird..

[PCBONEZ]

Re: Question about 85c caps vs. 105c caps

could bad PSU caps cause ripple current without any sort of sign (the system seeming absolutely fine, you would never know it)?

PSU's always have -some- ripple so edit that to excessive ripple.

And to that... Frequently.

It's usually in that condition for months before something actually breaks and the user takes notice.
.
Excessive Ripple can be the stealth root cause of things like bad RAM or even corrupted hard drives.
IC chips don't work well with noise in their DC power.
If the chip takes an input signal and converts it to an output signal then noisy power can mean the output signal is full of random glitches. [errors]
If the IC is a RAM chip it will probably fail Memtest.
If the IC is in a drive interface it can be corrupting files on their way to the drive.
.
Even a slow system can be caused by ripple.
If the system is catching and correcting all those errors then that amount of it's resources is not available for other things.
.

[PCBONEZ]

Re: Question about 85c caps vs. 105c caps

Now that you mention it.. yeah.. I've only ever seen 6.3V 1000uF TMZ in 8x11.5 .. no other values. Same goes for those 4V 680uF TMJ..

Weird thing is.. there are supposedly tons of different values and sizes out there.. lookie what I found:


What makes the whole situation even more complicated is that the only "common" TMZ's (6.3V 1000uF in 8mm) happen to be the only one that's missing in the PDF..

WTF Chemicon..!?

edit: according to info embedded in the PDF, it's from april 2004. In the PDF, the caps are described as having "white color letter on dark brown sleeve", so the ones on the ASRock board I posted a photo of must be even older than that, as they're black with golden letters

this is weird..

That is a TMZ series but I don't think it's Chemicon.

Chemicon shows their logo at the top of the page and the catalog number at the bottom in their data sheets.
That sheet has neither.
Also Chemicon Lytic part numbers begin with an E. The ones in that sheet don't.
There is also no 6.3v 1000uF 8mm shown.
.
.

[PCBONEZ]

Re: Question about 85c caps vs. 105c caps

I found it.
That's a Samsung TMZ data sheet.

I've measured the ESR on Chemicon TMZ and those Samsung TMZ aren't even close.
Chemicon TMZ ESR is between KZG and KZJ.
.