The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

[Topcat]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

As so many of you requested, I disected one of these Antec PSU's to see what caps were hiding in it. To my surprise and thankfulness, there were NO Fuhjyyu caps in it!! YAY!! It was filled with OST and Teapos. Neither of those are known to be good brands, as I've seen them fail in extremely high numbers, but they are definitely better than Fuhjyyu!!

Here's a few pics:

[Shroomie]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

That a Seasonic-built model? Never seen a CWT that looked like that, or had OSTs (and there's no CWT on the transformer either).

Good stuff!

[killjoy]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

Oh crap. I have one of those powers running in somewhat critical server. It's a dual xeon with quite a lot of scsi drives so the psu is being taxed. It has been running without problems for a year already but now you made me a bit worried.

Could you please list cap values and physical sizes? I'll be ordering replacements for this one asap. I don't want this server to go belly-up anytime soon.

[Topcat]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

Once again, I can not thank Antec enough for their contributions to this project!! Their generosity has gone well beyond my wildest expectations!

Ok, here's the story:
I emailed David Forster of Antec to thank him and his company for contributing the PSU's that they sent for us to use. In the correspondence, he asked what cases I had bought, that I mentioned the cheesy PSU's they had. I responded indicating they were just cases I had looked at and was going to purchase locally (I rarely ship cases, especially cheap ones, as UPS tends to play basketball with them and ruins them a lot of the time). He replied with something along these lines: 'Go to the Antec website and choose a case that you think will suit your needs, preferrably one without a PSU, since you already have good ones. We'll fulfill that need as well.' I had to read that email 3 times to make sure I comprehended it right!! They had some really nice file server cases on there, but I wasn't going to push my luck. I selected the Antec 'SOLO' ATX mid tower. A few weeks later, there's 4 HUGE boxes on my front porch! My girlfriend calls me and rags me out (in a fun way) because she thought I ordered another set of wheels for one of my cars (the boxes were about the right size for it). At the time she called, I had forgotten about the cases, and was wondering what she was talking about.... So I rush home excited (thinking its car parts), and I find these little jewels instead!! All that excitement wasn't in vain!! Ok, I've rambled long enough......

The cases are Antec SOLO's, all brand spankin new!! Antec has contributed over $1,000.00 USD worth of hardware to this experiment (based on their website prices)!!! That's a hell of a lot of merchandise just for a little homegrown experiment like this!! Again, I want to extend my most sincere thanks to David Forster of Antec for all he's done to make this experiment successful!!

Before I post some pics, I also wanted to thank ALL the contributors who donated parts to this one more time! This includes:

David Forster of Antec - 4x PSU's and 4x cases
Joe Lo (Big Pope) - Donated Samxon and Sanyo WG caps
Wetback - Donated the 4th Abit KT7-R

Without all of you, this experiment would have never left the ground!

The cases themselves are VERY nice!! Thick gauge steel, no sharp edges, EXCELLENT quality!! The cases are actually quite heavy! Anyway, here's some pics of the cases:

[willawake]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

4 cases also..you better do a good job there then

looks like rock solid chenming build for those cases, something i like very much....very good build indeed.

i think the internal psu pics should be done again cos they are crap pics. the other pics were great though

[PeteS in CA]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

When you find yourself in a hole:

1. Stop digging;

2. Start climbing out.

Looks like Antec is working on both.

[Topcat]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

Yea Will, I plan on getting some better PSU nudie pics... I'll get them in better lighting, and try andn keep the camera steady... With the macro zoom on, the slightest movement blurs the pic, and I guess my hands aren't as steady as they once were... I guess I need to invest in a mini tripod.

To whoever suggested keeping a spreadsheet earlier in the thread and suggestions on taking the readings/recording data, great ideas, I'll do just that. The Pannies arrived last week, and I'll recap the last remaining KT7 for this experiment. All that's left is to get CPU's. All the boards are rev 1.2 I believe, I'll have to do my homework to see what the fastest AMD CPU these boards were designed to run. ...and no, there will be no overclocking in this experiment.

[denyu]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

Mmm sorry, kind of currious how is the progress now ? supossed nothing happend yet, the project just started

[nania]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

I'm following this thread with great interest. I own a few Epox 8RDA which swelled the little caps by the memory and southbridge controllers. This is an issue for FSB overclockers but probably won't come into play with a box running stock speeds. Once this test is over (or possibly concurrently) I will offer to put up two overclocked 8RDA+ and two FIC AU11 with Athlon XP Mobile chips. In this way, we can test a great overclockers board and see which caps are vulnerable by the type of overclock. I have a plan if anyone is interested, are you listening TopCat?

[PeteS in CA]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

If the parts do at all decently, the test should take many thousands of hours. Running 24x7, that's 168 hours in a week, 8760 in a year. So the test could well take a couple of years or more.

[gonzo0815]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

Well, if there are suckers, we will see this far elaier due to bad ESR. But i thinbk this will not happen as all used caps are probably superior.

[nania]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

Yes, you guys get what I'm saying. Running this test at stock speeds with good caps will bring about a very long MTBF so my idea is to run two tests of two mainboards (the AU11 and 8RDA+) concurrently with TopCat. One test designed to stress the CPU multiplier overclock using both memory channels and one designed to test the bridges. I suspect my proposed tests will tell us a little about which caps are stressed by the type of overclock you use. My experience gives me a good idea about the expected results but it will be nice to corroborate my impressions with a controlled test.

[gonzo0815]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

I am afraid, that TC has limited time, place and money to do that a lot of trials. But i think, for me the most interesting thing is, if those Samxon are equla to the others for abouth one or two years. And if they would not stand up to the others, early signs of failure should be visible by the messurements very early. I don`t expect the definite answer if all used caps are on the same endurance, this would seriousely take to mutch time in any konfiguration. The problem with OC is IMHO not the endurance of the capacitor itself, it is more that the generall desing of vrm will be overlodaded. And in this scenario the cap with the lowest esr will win the battle or the one witch can stand the genrated heat. For seriousely OC i think the caps on the board should be alterted to meet the new power requirement. Thus either some of the best ultra low esr caps with higher capacitance should be used or some high quality solid polymer caps. This depends on the particular desing of the board.

[nania]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

gonzo
I think you may be missing my point. I am offering to do the overclock tests so there is no additional burden on TopCat. Boards like the 8RDA+ have sturdy regulators so overclocking tests will more quickly highlight the vulnerable capacitors. Most of the members here probably overclock their boards in some way and I suspect they would rather not wait years for a conclusion so my proposal is to add a little more "stress" to the test

[gonzo0815]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

Thas sounds cool, if you can put some 8RDA+ under severe OC stress i would be glad to see the results (yea your right i oc every system right after i assembled it ). May be some limited airflow would accelerate the research. I am very interested in high temp high stress test`s for any of the newest ultra low esr type, wich are usually only rated at 2000k. Esecially those Rubycon MCZ and similar series from the other brands like Samxon and UCC. I personally have not the parts, money and place to do a lot of high end testing. My personall testprogramm are ATM a old K7VZA with 700mhz duron on a very hot desktop case. I recaped the board (fully blown GSC) with MCZ, but the whole VRM on that is running pretty hot due to the mediocre ECS vrm desing. I am shure, that those MCZ have not that mutch headroom to survive very hot ambient to long. But i hope they have ;-) .

[trodas]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

yanz -

well, about the smallest caps, i agree with you. but if everything is running just right, i would just lazy to recap them too

I think we can't afford to say "I jsut got lazy" to ruin this test, because on one of the board some of the small caps go that bad, that it affect the stability of the mobo, resulting in canceled test - eg. at least that long, before the small caps won't get indentified as source of instability and replaced.
Other than that - I'm glad you agree with me on necessarity of replacing even all the small caps on the testing boards with some reliable proven brand (Ruby, Panny)
Topcat, what do you say about it? I know - more work, but look... there are reasons to believe that this should gretly increase the experiment professionality.
On the other hand - it can ruin it easily.


Topcat - great to hear that no Fuhjyyu caps are into these Antecs!
Teapos and OST aren't exactly what we love to see, but they should manage it for some time (I bet two lears on such light load is not unreasonable) and frankly - we want kill the caps, so, there must be some ripple, right?
So, big THX to Antec - and the cases are just so pretty and nice! HOORAY for that!
BUT you MUST remove the back fan!
Why? I run Samxons for 8 months now in sis JetWay V266B w/o any sign of problems, folding all the time + surfing, etc. and highly overclocked - poor VIA KT266 from 133 to 150Mhz... AND ALL THAT IN FANLESS CASE!
So, the test cases must be fanless too, if we see the Samxons fail in our lifes, I'm affraid.

PS. with just a little overclocking we can see results faster Just a suggestion, don't shoot me for it


Shroomie - your sig come into life again:

You know there's something wrong when you open up a PSU and are glad to find Teapos.

LOL I don't know how you can quessed it, but it is true. We find Teapos and OST and we are happy. Something is wrong with ut

nania - Big Poe showed me, that Epox 8RDA+ with Samxon GC caps everywere and a custom moded bios is overclockable by FSB to 250Mhz easily (CPU-Z verified ) Yet in this test we talking about older brand.
The observed failure of small caps is, as well as other members experiences

kc8adu, Super Moderator: i had similar bugs in a msi 694t pro till i replaced the tiny 10uf@16v sm lytics

just one another reason why replace the small caps too with same trusted brand for every of these boards.
Futhermore, you have 4 boards, so, what caps you want test agains what? And do you have 2 same, or at least two pairs of same mobile chips too?
And you have four same PSUs - preferably recaped with Pannyes? (since we talking about O/C there! )
And could you show us some recap work you did, so we can see what to expect from you? (no offence, but quality of work on recapping play damn big role there )

[Shroomie]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

LOL I don't know how you can quessed it, but it is true.

My ~350w Channel Well is all Teapo inside. I was thrilled that they weren't Fuhjyyu.

[Evnldr]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

I like my Antec 350.. But if im ever in the market for case or Pwr supply Antec will be the one I go with.. Yey Antec.. now charge 5$ more and get the good caps in there!
people will pay for quality...

[Topcat]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

Just a quick update on this project:

I've recapped the last KT7 (the one using pannies), I've located identical RAM, big fat identical heatsink/fans, video cards, hdd's, and NIC's. I still need CPU's. Other than that, we're ready to roll.

[tazwegion]

Re: The Great Capacitor Showdown - Rubycon vs Panasonic vs Samxon

I personally have not the parts, money and place to do a lot of high end testing. My personall testprogramm are ATM a old K7VZA with 700mhz duron on a very hot desktop case. I recaped the board (fully blown GSC) with MCZ, but the whole VRM on that is running pretty hot due to the mediocre ECS vrm desing.

Me too... though I think I'm the only member here using a mix of Hitano EXR series & Low ESR SamXon components, to date @ 24/7 100% F@H (3 months) with no signs of instability and/or capacitor bloating/venting

Nice work TC, very favourable of 'Antec' to be so generous their R&D is on the right track now!

* slaps Fuhjyyu CEO *