Discussion in 'Modding' started by trodas, May 8, 2007.
I did laugh a bit.
Stygian Shane -
Definitively. What capacitor type this was? And what circuit this is? Don't looking to me that it could overstress the cap... so, it got me asking
Unknown mainboard and user, found on BadCaps forum, YEC caps:
Nightmares will be had
This came out of an airconditioner which wasn't working for some reason...
Click to view full size!
Uhuguhguh, that makes me feel a little dead inside.
I've seen remarkably few catastrophic cap failures these last couple of years... wonder what the go is. Technology is getting crappier and more consumerised, things should be failing more! Solid caps on PC parts have all but killed 99% of the cap death porn I guess. Haven't seen an A/C unit with a failure like that in forever though either!
This is from a HP Proliant ML350 server of all things. Holy shit you think a professional grade server for 24/7 use would have some decent caps, right?
I have some ML110 and ML150's that have similar rubbish (with failures) too. Yet other same models of different generations have better caps.
I've retired mine them from service even though they are still running OK. The older (<G6) HP ML motherboards (working or repaired) are worth more than the entire machine, so repairing them for resale can be quite a good earner.
Well you would think they would have fixed this issue ages ago.
I'd say that they have.
That photo above is of a ML350 G5, which was released in 2006 or 2007, right at the end of the "Capacitor Plague" era.
Most of the "critical" caps are solid electros, but there are also a smattering of standard ones.
Yeah I was going to say, I would expect that to have been an issue of the past.
HP Proliant ML350 server - what caps are they? From the top, they looks like Rubycon caps, witch are good caps... unless they are fakes...
And what the hell is this?! That much stolen caps?!
PS. Smokin Whale, could you please tell me, what capacitor is this failed one?
IMHO they are likely fake Rubycon caps. Is there a way to clarify this? Thanks!
ECS MCP61PM-GM mainboard witch use Panasonic FL (fake!) caps, user jsc1973, Overclock.net forum.
Fixed the Czenglish
Yes, those are genuine (NOT fake) Rubycon MCZs, as are the caps in Smokin' Whale's ML350. Ruby MCZs may be decent, but it obviously gets too hot in there for them (or they are being subjected to too much ripple). In all likelihood, only solid/polymer caps would hold up OK in there. I have an ML150 G3, it also have a lot of caps which weren't installed from the factory on the VRM-low side. In face, there was only one 820uF 2.5V poly there on each of the CPU sockets. I filled the blank spots up with 560uF 2.5V OS-CON SEPC polies.
Same kind of thing probably goes for that ECS MCP61PM-GM. Those caps are NOT fakes, nor were they poorly made, but were just overstressed. Remember, motherboard manufacturers never ever use fake parts. They could only be fakes if someone else has already recapped it with fakes from fleabay.
Hopefully this is based on some solid evidence, not just your wishfull thinking? I say this, because you are proven to be wrong on many cases (as almost any human)...
A picture will be worth thousads of words... but... what a ripoff! This is crazy...
This is, where you are wrong. And I prove my case, instead of just saying that you are wrong, that mobo manufacturers never use fake caps.
This is Apple PoweMac G5:
These caps are labeled as Nichicons HN, witch are quality ones (except for one bad batch). The catch? HN(M) stays for Nichicon, yes, but these are black-white, not black-gold. Only gold ones in Nichicon line are HZ ones... Hence they are fakes and Apple bought fake caps and then selling hardware with them to their customers... that will fail.
Real life difference between Nichicon HN and HZ:
(true Nichicon HM and Nichicon HZ on my own mobo)
Wanna more proof? When I posted that on Overclock.net forum, a user Xylene quoted my post, saying:
So, it is a common knowledge, that even Apple used fake caps. Update your knowledge, pretty please with suggar on top
(and yes, sorry for my poor grammar - I did too many typos, not need to be sarcastic about it... or it was meant more as attack?)
ASRock P4S61, Chemi-con KZG caps, user JKuhn, Overclock.net forum.
100% Wrong. Geniune HNs, even from the proper distributors are always black and gold. Only HM are black and white. Also, Nichicon themselves acknowledged that there was a defect which affected all HMs and HNs from before 2005 - when the mac G5s were made, and this caused them to fail. It was not just one bad batch. Sorry, but those caps are NOT fakes. If you try telling everyone on badcaps.net that HNs should all be black and white (rather than black and gold), then I think everyone will give you the same reaction and tell you otherwise.
That quote from Xylene doesn't even mention them being fakes. It doesn't prove anything.
Sorry if I sound a bit agro there. I can't really think if a way of putting it more gently.