I think a lot of people do hear the high-frequency sound, but they just don't see it as a problem or they happily tune it out. When we point it out, they say "Huh? THAT is what's bothering you?" and look at us funny.

Quite a bit like the goal of making computers quieter. Same response.

Some people hear the really high frequency stuff and others dont, i remember playing with a freqency generator and speaker in electronics class. A kid named Charles would alway yell at us when i turned it really high. Just to test it out i would randomly turn it on with a frequendy beyond my hearing range, like 22Khz. every single time charles would yell at us even when he couldnt see me, because his earts were sensistive to those frequencies and they hurt his ears. Luckily high frequcncies are very directional and can easily blocked/absorbed by puting something between you and the offender.

I wonder if there could be a way to mount this vertically. I'd have to assume that a straight-through path for the airflow would be way more efficient and cool better.

It would not be pretty to have the power cord sticking straight up, but perhaps if there was a 90degree cord... This would also enable you to see the LED status; but it seems like that is not wholly accurate anyway.

That's an interesting idea. Its effectiveness would depend on whether or not the heatpipe must be on the horizontal as it would be in the normal mounting arrangement for a PSU. Some heatpipes are able to work at full efficiency on any axis while others cannot.

MikeC do you happen to know?

It would probably work fine. In that config, at least part of the heatpipe would be running up/down, with the heat source on the down side, which is good. But I don't know whether the amount of hacking you'd need to do to an existing case would be worth the trouble. Simply arranging for as much tight contact surface area between PSU casing and case as possible would be a bettery strategy IMO. Plus, the test system is running 24/7 (now well over a week) and no problems whatsoever.

Howdy everyone, sorry I'm a bit late coming in to this one!

I use my PC for audio recording, and despite some reasonable efort on my part it's not really quiet enough yet. I'm planning, as soon as I can get enough money, on something of a refit, and the fanless PSU idea seems pretty groovy. I saw this puppy advertised somewhere a while back, so I'm really pleased you've reviewed it.

I'm still debating whether to go for an Acousticase with all the damping material or to try some kind of rack-mount arrangement. I think the former might not be the ideal home for this PSU, but perhaps a horizontal layout might actually help? I figure a central extract fan at the back would keep the case clear and the PSU could remain largely independant over on one side.

I've read a few discussions on rack-mount cases here, and it's an area which most pro-audio types would be interested in so it'd be really great to see some reviews. I think I found this link on these forums, seems to be the best value rack case available in the UK http://www.xcase.co.uk/acatalog/Compucase_S411_.html - but could it be made quiet?!? I should think a fanless PSU would help.

I've gone a bit off-topic here, haven't I?!? Anyway, thanks for the review and the site in general. I found the Etasis version a little cheaper than this in the UK at Scan http://www.scan.co.uk/Products/ProductInfo.asp?WebProductID=116236 and as that's got a bigger heatsink it might be a better deal?

Joe.

Wellwellwell,

thread hijacking all the way... Sorry about this but I had to reply to GB Joe´s post. As it happens I bought that case for my workstation just a couple of months ago. The design is really open airflow wise and well, if there is something inside that makes noise then you can hear it. I´m really pleased with it though, for a rackmount case it is actually fun to work with and can be made even quieter with the right selection of components.

Here is a cut-and-paste of a short evaluation and advice I gave at Arstechnica´s forum:

--
Hi,

I actually bought the Compucase (HEC Group) version of that case (S411) from Finland a couple of months ago. It is the same case as the Antec 4U22ATX400 but without a powersupply, paid something like 140€ for it.

All in all one of the best cases that I have ever bought or worked with. Setting up the thing can take some time as setting up rackmounts usually does, the 5´25" drive bay holders need to be taken out (4 screws) to install drives etc. Not a real pain but something that can be a bit bothersome if you need to swap those components often. The case is also rather deep (60cm) so placing it can also be troublesome. It has a quite an open design so airflow is not a problem.

It is currently running:
- Abit IC7-G v1.1
- 2.8C P4
- Zalman´s 7000A-CU with a Zalman Fanmate 1, dialed to minimum.
- NB fan turned down with a Zalman Fanmate 1, dialed to minimum.
- Tagan 380W quiet PSU
- 2x Seagate SATA 7200.7´s (80GB and 200GB) in Supermicro CSE-M35T1 5 bay hotswap enclosure (Replaced the original fan, a real hoover, with a 92mm Zalman).
- LiteOn CD-RW.
- 2x 80mm fans (Globefan? the ones that came with the case) at the back at 5V.

Temps are nice (HDD´s idle at 37C, CPU and NB heatsinks are running quite neutral to touch) and the rig is quietish, can´t make it absolutely quiet with all those fans providing the unfortunately necessary airflow. And because the Supermicro backplane doesn´t isolate the drives elastically the seek noises are somewhat audible.

I´m planning of buying another one for my next computer, it really is that nice to work with.
--

Linkage : Rackmount and Quiet?

All in all, my opinions about the case haven´t changed during this time.

What comes to the PSU with the external heatsink, I wonder if we will have the option of affixing a system consisting of heatpipes and additional heatsinks between the fins of the external heatsink to take the heat away even further. Kinda like taking the heat out-of-teh-box completely. It would be fun to have for example a video card which would have heatpipes that run out of the case via a modified PCI-slot cover to an external passive heatsink. Would like to try that one someday.

ALERT!

After a little over a month of 24/7 folding @ home & fine service in my backup PC, the SilverStone fanless PSU sample tested for the reviewstopped working a couple days ago. The power/function LED lights up RED when the PSU is turned on, either in the PC or on the test bench with any load. I don not know what the cause for the failure is.

The PSU was drawing a maximum of ~165W AC in the backup PC, typically 125W or less. The temp at the rear heatsink ranged 48~52C.

I've contacted SilverStone and am arranging for a return so they can investigate cause of failure. Will report back when I have more info.

...hopes dashed...

Please keep in mind...

A review of an individual product sample tells you details about that sample -- and it is hopefully representative of other samples of that product. However, there are always variances in any product line. The failure of a sample does not condemn the model in question; nor does a positive review about a sample mean every sample of that product will fare as well. Reviews are meant to illuminate strengths and weaknesses of a product's design and quality, but cannot possibly address consistency issues nor every aspect of its design. This is especially true for something as complex as a PSU for a DIY PC where thermal and load conditions can vary hugely.

Then let us all hope this is a fluke.

Then let us all hope this PSU has a good warantee.

I plan on running mine outside the case so i dont think the temps will be too high, I'm very curios as to the cuase of the failure, and the nature of the failure. (crazy EE student)

I just realised this PSU is on sale in New Zealand. Its particularly difficult to find fanless PSU's over here.

Will be keeping an eye on this thread, hopefully not all models suffer the same problem as Mike's review sample.

Some updates:

I have the Silverstone connected to a power filter and the whine is still there. The whine is soft and high pitched, and I think it is surely noticeable. So, I may send it in for warranty, but I probably won't do that anytime soon.

I've since put this PSU in the evercase 4252. And, it appears the PSU is running cooler than outside the case. I had the PSU upside down before, so maybe that was interfering with the heatpipe. Though, I think the more likely reason is that the 120mm Nexus exhaust fan (at 9v) is pulling in cool air through the ridged heatsink on the bottom the psu. Either way my temp led has always been green (which means normal operation).

Any puffs of smoke, burning or similarly dramatic damage or did the PSU simply shut itself down?

Maybe failure is just fluke, but long term reliability, especially in machines that are on 24/7, is still pretty much unknown in fanless PSUs. Well, I was a coward and already decided last week to go with Noisetaker rather than Silverstone.

No, just a quiet failure.

Aside from manufacturing inconsistencies, and realizing that the failure of Mike's sample could be due to just one single component rather than a design or system failure, we should bear in mind that the ATX packaging contraints make the implementation of a fanless solution a real challenge. There are a large number of parts jammed into a small space. Locating it internally in a case would just, it seems to me, add to the cooling difficulties.

I wouldn't be surprised that if an intrepid enthusiast took one of these PSU's apart to give it "room to breathe" in a custom installation then the PSU would fare much better. Sometimes, it's simply not possible to get everything one wants in one little package within current technology and financial constraints. Perhaps that is why Antec's Phantom took so long to release and is priced at $199? Makes one wonder what Zalman's forthcoming external fanless PSU will end up costing.

For the time being, is a high efficiency PSU with a slow turning fan still the safest and most cost effective way of approaching our goals?

Actually, the lowest quoted price I found for the Phantom is $148 (but not in stock). It's $199 at NewEgg. Lowest quoted price for the Silverstone was $125. Lowest quoted price for a Seasonic Super Tornado 300 with 120 mm fan? $58. (I realize that this is just USA pricing but expect that the disparities would apply in other parts of the globe, assuming availability).

Sorry, straying somewhat off topic, but out of fairness to an SPCR sponsor should mention that Silicon Acoustics is having a sale on the A3 version of the Super Tornado at $55. http://www.siliconacoustics.com/seasoni ... 0watt.html

Ok, since prices are always changing, and I'm not a tout for vendors, I've learnt my lesson and will stay away from posting prices on serious minded threads such as this one in the future (there is a Section in the Forums for that very purpose - if one is so inclined).

almost a month later; any news on this yet? i'm really curious about why this psu gave up the ghost.

UPDATE on the Silverstone...

Nothing to report on the one that failed. My guess is that analyzing this particular failure is low on the list of priorities for Silverstone and we should not hold our breath.

A replacement arrived at the beginning of Oct, and it has been working fine in test systems and other lab duties. I have not had it long term in a 24/7 system -- but perhaps this is one of the caveats about fanless PSUs: don't run it 24/7? It's probably wise...

Been running mine for 6 weeks now, 24/7. No problems at all. No whining or buzzing. Mine is located half-inside, half-outside a tower case (Lian-Li V2000), although it still runs hot (as in burn your hand hot).

dunno if the phantom might be a better choice? It doesn't seem to run THAT hot for me.

REPORT from SILVERSTONE on failed test sample

They apologized for the long delay, but finally sent an email explaining the problem found by the PSU engineering team:

"The diagnosis was that the "U1" IC that controls the AC to +5VSB rail went bad. There are no obvious causes for this part's failure because this is a simple IC that is highly unlikely to be overloaded or affected by high temperature. They assured us that this is a rare case of defective part and we probably won't see a failure like this again."

This seems fairly reassuring. The other good news is that the second sample that's been working hard in the lab continues like a workhorse, mostly 24/7.

That's reassuring to know.

Mine's been running fine. I've had it for 4 months now, running 24/7.

Wonder if this PSU would fit into my case.
(check my hardware profile at bottom)

Today..I can feel that the PSU is pushing normal hot air out..
Will the new PSU burn up ? (or just stop)

still going ok?

(i'm wondering why everybody bought phantoms and not this psu. i'd hate that big light on the back of the phantom and with all the failure reports i wouldn't want to disconnect it if it would stuff the warranty)

wim --

Yup. It's perfectly stable running an A64-3800+ system that I'm actually using for audio recording these days. (Things get juggled around according to need.)