KZG's still bad?

[PeteS in CA]

Re: KZG's still bad?

2000 hour KZG caps aren't expected to last. We're just annoyed when these 2000 hour caps fail a lot sooner than other brands' 2000 hour caps. Either KZG is overrating or other brands are underrating.

Nichicon HM and HN are rated for 2000 hours as well. Keep in mind that that is 2000 hours minimum at rated ripple current and at rated temperature continually. If a computer mfr has components continually running that hot and with equivalent stress it won't be just the lytic caps that will fail in a few months.

[Behemot]

Re: KZG's still bad?

Well, not really. Most FETs and diode bridges could operate under these temperatures for long time. If these are heating the caps, they can have 105 °C but still have to be able to survive it for almost a year (year is ~5,5 hour every day). We all know that KZGs will do this for less than 1000 hours

[PeteS in CA]

Re: KZG's still bad?

Don't forget, though, that a 105C case temp on a semiconductor means an even hotter junction temperature, and that's what matters. In 1980, Boschert's design standard was not to have junction temps above 100C (or maybe it was 90C). I think companies allow higher nowadays.

[brethin]

Re: KZG's still bad?

Yo guys, I got Lenovo M57 microcomputer system in my hands. It is basically full of Chemi-Cons KZG, 1000 uF/6,3 V and 1500 uF/10 V. The big question is are they still bad, or has Nippon Chemi-Con changed the electrolyte?

Cause, you know, it has to hold at least a year when I will provide warranty reselling it, on the other hand, if they are good (or at least better than used to be), I can spare my Rubycons and the time, too

I don't know the exact manufacturing year, but they are darker than used to be, marked 8(7) 4R, 8(7) 4Z and 6(T) 2I. I guess this is new dating system as the older one I found definitely won't apply on this. Or they are just OEM's with different marking.

I have that same board but in a custom case with much better cooling than the factory setup and have had no issuses with it (even with the KZG's) its all about the cooling.

If your just looking to unload it and not worry about it then change the caps but if you know its going somewhere where it will be kept cool and clean I would just offer the year warranty and replace the caps if it starts acting up.

[Behemot]

Re: KZG's still bad?

Changed them, burned MOSFET, chaged it, burned again, changed, fan didn't work. Plugged fan on the disk molex and hopefully the guy will come for it soon :-D

[RJARRRPCGP]

Re: KZG's still bad?

I have that same board but in a custom case with much better cooling than the factory setup and have had no issuses with it (even with the KZG's) its all about the cooling.

I'm thinking the same thing, too.
KZG don't seem to be quite as bad as Fuhjyyu was on Antec PSUs.

Because I cannot even recall seeing a bad KZG at WinCycle LOL.

And I can't find any signs of bad caps on my Asus A7N8X-X, which has a lot of them and a Dell Vostro 200 that was given to me in 2011.

And it looks like the ones on my MSI 845E Max could be fine.

Because the random cold boot problems on my MSI 845E Max seems to be related to my "StupidPower" 2.0 500W, which in November, 2011, found a bulging Fuhjyyu cap next to the outer fan.

But the last time I booted the MSI 845E Max, was before I saw any bad caps in the PSU.

[Th3_uN1Qu3]

Re: KZG's still bad?

I'm thinking the same thing, too.
KZG don't seem to be quite as bad as Fuhjyyu was on Antec PSUs.

I see them failing even on boards running Sempron CPUs. I also have an intel manufactured 775 board that i haven't gotten around to recapping yet, because it had 8 3300uF 16v KZGs scattered all over the place, and ALL but one were bloated. Even the two that hadn't bulged yet tested out of spec for ESR. The board was still working when i got it tho - the VRM out is all poly. Seeing bad KZG even in low-stress positions like RAM and chipset is enough to put them on my "replace on sight" list.

On the other hand the 3300uF 6.3v seems to be the most affected part, others do better. One of the boards i stripped for parts had in the VRM one 6.3v 3300u and the rest 6.3v 1500u (a whole bunch of 'em). All looked and tested good, including the 3300u. That's the first 3300u i've found good so far.

[goodpsusearch]

Re: KZG's still bad?

I had a KZG 3300uF 6.3V that bulged in the sofa unused, so yes they are very very bad...

[RJARRRPCGP]

Re: KZG's still bad?

I had a KZG 3300uF 6.3V that bulged in the sofa unused, so yes they are very very bad...

That's not like United Chemicon! I have a better chance seeing a bulging Samxon.

[Behemot]

Re: KZG's still bad?

You may not be aware of that, but KZG has bad electrolyte. It is known bad series (together with one or two which share this electrolyte) for many years now. On the other hand, it is the cheapest from comparable caps. This together with known short life (shorter than specified) makes it great for MoBo makers cause it can die shortly after 2 year warranty.

Every manufacturer (maybe except for Rubycon) had or has its bad series: HN for Nichicon, KZG (KZJ/TMV) for NCC, the same (short life time and low price) but for power supplies (usually in displays) is GF for Man Yue.

[goodpsusearch]

Re: KZG's still bad?

https://www.badcaps.net/forum/showthread.php?t=19441

Every manufacturer (maybe except for Rubycon) had or has its bad series

and Panasonic and Sanyo too

[japlytic]

Re: KZG's still bad?

Should I get rid of used KZGs in my inventory?

[Rulycat]

Re: KZG's still bad?

I recommend so. Junky caps. Having said that, I've an old board running in my dad's PC using 4 3300uf KZGs and an Opteron 939 which is fine. OSTs on that board too.

[ratdude747]

Re: KZG's still bad?

Should I get rid of used KZGs in my inventory?

Yes.

[momaka]

Re: KZG's still bad?

Should I get rid of used KZGs in my inventory?

If they test fine on an ESR meter, you can still keep them around for testing. And even if you can't test them, they may still work fine for general purpose projects.

At least that's what I do with my crappy caps. Sure they may not be very reliable, but if I need to recap something just to test it (i.e. no intention of leaving the caps there long term) or something that I don't care much about, then I use them.
The only thing worse than unreliable caps is failed caps.

[Behemot]

Re: KZG's still bad?

Too much work with soldering just for testing. I change it, if it works, OK, if not, remove them and disassemble the HW for components.

[Agent24]

Re: KZG's still bad?

If they test fine on an ESR meter, you can still keep them around for testing. And even if you can't test them, they may still work fine for general purpose projects.

I wouldn't bother with using them for testing, too much hassle as Behemot said, but I generally keep cheap branded ones that test out OK in terms of ESR and capacitance for use in general purpose situations, random projects and such.

[ant3202]

Re: KZG's still bad?

somehow... if possible... i will still avoid them