asus/hp 478 board with blown mcz

[kc8adu]

asus built hp board.
blown 1800@6.3 mcz
the ones not bulging were open.
only reason it still ran was the polys were still good.

[zandrax]

Re: asus/hp 478 board with blown mcz

Is a SFF HP like this one? Damn, caps are "glued" to the heatsink with little or absent cooling, no wonder there are leaks. But shorted caps are due to overheating or to a vrm design issue?

Zandrax

[kc8adu]

Re: asus/hp 478 board with blown mcz

no.
its a midtower case.
no bad fans or power supply although the bestec is getting recapped to eliminate the common +5 stby overshoot.
seems these boards get hot in the vrm area.

[yanz]

Re: asus/hp 478 board with blown mcz

So i think it's the limit of mcz then.. running in the hot area for years (s478 era), with minimum vents and high heat. Even good caps will fail. And like zandrax said, they were "glued" to the heatsink...
That's a bit of weakness in design too..

[gg1978]

Re: asus/hp 478 board with blown mcz

Is that a Prescott or Northie P4?? Looks like the motherboard itself is getting pretty hot around the MOSFETs, going by the slightly darker brown color..

[RJARRRPCGP]

Re: asus/hp 478 board with blown mcz

asus built hp board.
blown 1800@6.3 mcz
the ones not bulging were open.
only reason it still ran was the polys were still good.

*Heads up, I recently saw bad Rubycon MCZ caps, too!*

I did at WinCycle, a place that recycles computer parts and sells second hand computers in Windsor, Vermont.

Since 2008, I actually got to see bad caps for real, when in the WinCycle recycling warehouse And strangely, I likely saw more than one occurrence of bulged Rubycon MCZs!

[gdement]

Re: asus/hp 478 board with blown mcz

*Heads up, I recently saw bad Rubycon MCZ caps, too!*

I did at WinCycle, a place that recycles computer parts and sells second hand computers in Windsor, Vermont.

Since 2008, I actually got to see bad caps for real, when in the WinCycle recycling warehouse And strangely, I likely saw more than one occurrence of bulged Rubycon MCZs!

What kind of machines were they?
I don't doubt that the extremely low-ESR conventional caps might be more prone to these issues due to pushing the limits of that technology. Polymers are probably best for small toaster boxes. But there's hardly any examples of MCZ problems so I think they're as good as anything outside of polymers.

If there's ever a next-gen capacitor contest, I'd vote for a lineup of hot SFF GX270's running MCZ, KZJ, Panasonic FL or FJ (not sure which is higher end), Nichicon HZ, etc.

[linuxguru]

Re: asus/hp 478 board with blown mcz

FJ is slightly better, I think - I haven't seen any blown FJs ever, though there may be quiet dry-out failures around.

MCZ failures in the VRM area, especially in Prescott boxes with low ventilation, have been known. However, just about any Al electrolytic will probably die in similar conditions (except maybe high-endurance Rubycon ZT (125c), NCC LXF and Nichicon PX - but those will not meet the ultra-low ESR requirements for the Vcore rail). The only viable option remains polymer caps on Vcore.

[RJARRRPCGP]

Re: asus/hp 478 board with blown mcz

BTW, the motherboards did look like HPs.

[RJARRRPCGP]

Re: asus/hp 478 board with blown mcz

FJ is slightly better, I think - I haven't seen any blown FJs ever, though there may be quiet dry-out failures around.

MCZ failures in the VRM area, especially in Prescott boxes with low ventilation, have been known.

With some posts here, looks like possibly the same thing occurring with Abit AN7s.

And this seems to be as common with United Chemicon KZGs on socket A motherboards and Athlon 64s.

[gdement]

Re: asus/hp 478 board with blown mcz

With some posts here, looks like possibly the same thing occurring with Abit AN7s.

Somebody did post questions about caps on their AN7, but there wasn't much evidence they were bad - it was just a guess.

Other builds could be different, but my AN7 has ZL and MBZ caps, not MCZ.

[RJARRRPCGP]

Re: asus/hp 478 board with blown mcz

Somebody did post questions about caps on their AN7, but there wasn't much evidence they were bad - it was just a guess.

Other builds could be different, but my AN7 has ZL and MBZ caps, not MCZ.

I also meant ones with MCZs and ZLs.

[staze]

Re: asus/hp 478 board with blown mcz

I've been seeing blown MCZs in iMac G5s for a while now. I'm guessing it's heat related due to the limited ventilation. I just replaced 6.3v 1800uf MCZs that were blown with 10v 1800uf MCZs and the new ones are even starting to bulge after 5 days of stress testing.

What's the alternative? These machines do get quite hot while running at high CPU load.

Thanks!

[Scenic]

Re: asus/hp 478 board with blown mcz

dunno if it's really an alternative (haven't looked up the datasheet regarding ESR..) but what about high-temperature caps... !?

http://cgi.ebay.com/ws/eBayISAPI.dll...m=300279687636

if they are ok for mobo use it would be interesting to see how long they last in those dell SFF ovens.. lol

[kc8adu]

Re: asus/hp 478 board with blown mcz

I've been seeing blown MCZs in iMac G5s for a while now. I'm guessing it's heat related due to the limited ventilation. I just replaced 6.3v 1800uf MCZs that were blown with 10v 1800uf MCZs and the new ones are even starting to bulge after 5 days of stress testing.

What's the alternative? These machines do get quite hot while running at high CPU load.

Thanks!

ouch!!!
3 things come to mind.
reversed polarity(you installed them backwards)
counterfiets/fakes.
bad batch.
i would likely use 820@2.5 solid polys on vcore in that unit.
chemicon psa or fujitsu re
i use both with no issues to date.

[PCBONEZ]

Re: asus/hp 478 board with blown mcz

better than, worse than, better than, worse than,,,, AHHHHH.

Here is my cheat list.

Groupings based on just based on ESR/Ripple.
----
This came from looking at 8x20mm and 10x20mm cans between 820uF and 3300uF.
Within a Series ESR changes with CAN SIZE, not with uF.
Numbers after Series are ESR for 8x20 | 10x20 mm for that group.

HZ, GA ...... .009 | .007
-------
MCZ, HN, GC, KZJ ..... .012 | .010-.011
-------
FJ, NRSJ, [FL - I THINK. Don't have an 'official' data sheet for FL.] ...... .016 | .013
-------
MBZ, WG, KZG, GD, NRSK, HM ....... .018-.021 | .013
-------
ZLG, FM, GE ...... .020-.030 | .014-.019
-------
HV, NRSH, KZM, KZH ...... .029-.033 | .019-.020
-------
ZL, HD, GK, KZE, WX, NRSG ...... .041 | .023
-------
ZT (high temp series) ...... .054 | .030
-------
KY, HE, YXG, YXH, FK, PA ..... .044-.069 | .035-.046
-------
FC, PW ...... .065 | .052
-------
NRSX, NRSZ, LXZ, PM ...... .069-.085 | .052-.060

Any that look out of sequence or like they overlap aren't and don't.
It looks that way without the ripple ratings which I took into account too but I don't wanna write all that out.

This isn't a perfect cross reference. (Because there isn't one).
Why? - Some given uF/volts in one brand/series may be a different can size in another brand/series and so for that case their ESR/Ripple won't follow the general rule.
-
This list is an aid to know which data sheets are best to look in.
You still need to look in the data sheets to cross reference for sure.

---

KC is right.
Hot Boxes need poly's to live long (and prosper?)
-
Need to look in data sheets for those too.
Some series are rated worse than the better Electrolytics.

.