The SilverStone ST30NFis probably the most sophisticated fanless PSU we've examined thus far. Is it the best?

NOTE this caution further down the thread:
http://forums.silentpcreview.com/viewto ... ht=#125061

Looks awesome!

Me wantsssss....

*preciousssssss....*

EDIT: Um.. Wouldn't convection work better if the cold air enters from a low point and exits, from a high point? Because when placed in the chenbro gaming box with the PSU duct the openings for air intake and exit are roughly at the same height. IMO it would work better if you ducted the 120mm fan opening in the back to the bottom of the PSU, so that the air goes into the box via the intkae, then up into the PSU, and then out through the exit. (there won't be any probelms with hot air going back into the intake because the hot air will rise upwards after exiting the PSU.

Nice review....looks like this is the one. Many ways are possible to vent this PSU, even without mounting it external to the case. The main thing (for me) would be to prevent heat from the PSU from getting back into the case. The setup in the article with a foam barrier and a front bay opening is one way.

I suppose a positive pressure case with airflow blowing out through the PSU is another, but I've got a few ideas that I've been working on that differ somewhat. This PSU looks like a winner.

Thanks MikeC......

Here's the other version of the Silverstone. Might cool better with its' outside heatsink. Insides the same? My questions about both versions ...is there airflow through the outside heatsink or just the heatpipe interface to the heatsink?

Hmm yeah this psu looks really good, tight efficiency! I was thinking about those Lian Li V series casings and Bluefront's venturi mod, about how the exhaust air would cool the back aluminium of the psu. Perhaps with positive airflow it will work even better?

This is the best passive PSU available in Germany. Great that also US customers are able to buy it.
Overview of passive PSU available in Germany:
http://forums.silentpcreview.com/viewto ... ilverstone

There does not seem to be much airflow. As you can see in the pics, this PSU is chock-full, packed really tightly. I think all the vent openings would have to be as big as the ones on the inside panel for convection to really have a chance of drawing any significant amount of air through the PSU.

The SilverStone has been in that system now for almost a week, and the temp sensor has been on it all this time. I don't have a log, but any time I glance at the display, the temp has ranged from a low of 48C to a high of 56C. This is over a range of ambient temps 22C~27C, and stress loads -- probably the most stressful thing I've done on it is some powerpoint presentations with huge numbers of images and Photoshop multitasking.

No signs of instability.

If I were making a custom rig with the ST30NF, I'd pick a decent aluminum case with extra mounting holes for the PSU on the top inside of the case, and use the extra bracket to mount the PSU. This would help increase the heat conduction path between PSU and case. Then I'd either cut open the whole area over the PSU or find some highly conductive material to fit between the top of the case and the PSU. Conduction is mostly how this PSU cools itself; my approach would be to try and extend it to as big an area as possible. Perhaps a channel open from a lower CD bay would work to draw air through the PSU, but if the conduction to the case is done well, I think you would not it.

It's an excellent review, as always. I just want to pick one nit.

I don't see the point of the intake duct. It makes perfect sense with forced air, but convection needs a vertical air path to work. I think you've effectively choked off any convection through the PSU by forcing the air to move horizontally.

That jibes with the fact that the PSU temp. rise is higher during Folding (29 degrees over ambient, 142W A/C) than at 150W on the test bench (26 degrees above ambient, 150W / 78% efficiency = 192W A/C). I realize that the test bench and the Chenbro aren't directly comparable environments, but I'm still concerned.

Would it be possible to run a few of the in-PC tests again with the duct removed?

Thanks.

On an unrelated topic, if this PSU sheds heat through conduction into the case, is it time to start smearing thermal interface material between the PSU and case?

kbob

Mike, if we can seriously get those heatpipe samples, we could do some amazing things in combination with this PSU!!!

Regarding the duct -- I see the point. I had meant to do a 2nd test w/o the duct; not sure why it was not done... been very hectic this last couple weeks.

Anyway, the system has been running 24/7 w/ folding@home, and the ambient temp in my office this morn is 23C, same as during the in-system test, so...

The closed-cell foam 1/2" thick sheet was removed. That's the only change, the top CD bay remains open. Now it is 25 minutes later (probably not enough time for temps to fully stabilize).

Before: PSU 54.6C, CPU 55C, SYS 44C
+25 mins: PSU 51.3C, CPU 55C, SYS 45C

So you were right, the duct was keeping airflow from helping with cooling. The temp was dropping at the rate of about 0.1C/min for the last 10 minutes, but seems to have stopped. Will report if there is any further drop.

re -- the TIM, I think it could help -- a little anyway. I am going to try wadding some aluminum foil and wedging it between the PSU and the top of the case. Because the HS and casing is so tightly coupled together, the whole PSU enclosure really does seem to get evenly hot.

Mayeb doing something like I said a few psts back would be worth testing?

Like this:

Aren't there a couple of top end aluminium cases that have provision for placing the PSU at the bottom of the case? Situating the PSU at the bottom could mean lower temps - since, as you stated in your review Mike, hot air rises from other components and is normally vented through a top located PSU (as well as the rear case fan). What do you think?

Qwertyiopisme --

the trouble with turning the back panel fan/opening into an intake is: What happens to CPU/system heat exhaust? You'll see temps for both rise quite a bit, guaranteed, unless there was some other way for that heat to escape the case.

Anyway, some 45 mins later...

Before: PSU 54.6C, CPU 55C, SYS 44C
+25 mins: PSU 51.3C, CPU 55C, SYS 45C
+45 mins: PSU 50.8C, CPU 56-57C, SYS 46C -- PSU temp has not changed for nearly 10 mins.

The PSU temp has dropped nearly 4C, the CPU temp is probably up 1.5C, and the SYS temp is up 2C. Implication is a bit more airflow would probably help -- maybe a 120mm fan instead of the 92mm in there. I don;t feel any more air coming out the back of the PSU, but one has to assume there must be at least a tiny bit more. More importantly, the "sidestream" of the exhaust fan airflow must be having some cooling effect across the bottom of the PSU.

Thanks, Mike.

As for the aluminum foil, might I suggest tightly rolled tubes of aluminum? Solid core, or as near as you can approximate, so they can be squeezed tightly for good contact. The smooth outer surfaces would have a large contact area with the PSU and case surfaces for good conduction, and using tubes instead of sheets assures short, all-metal paths from bottom to top.

Downside is you'd use a whole box of foil. But you wouldn't need to fill the entire space above the PSU, I don't think.

kbob

I wonder how this PSU would fare alongside a Reserator. With the two together one could go pretty close to fanless I guess.

Just go with a notebook drive, and make sure there is a very open airflow path from low front to high back. No fan needed, IMO.

Ah yes.. I'm getting used to not dealing with lots of heat in the case

*wuves my watercooling*

maybe (if there is no fan in the back hole), the duct could cover the top ½ or top 1/3 of the hole, so that air can both escape and enter there (although there would be some air leakge from hte "hot" exhaust into the "cool" intake..

All of Antec's larger (and presumably therefore a lot of Chieftecs and nonames based on that design) cases have about a full inch of completely unused space above the PSU. Wonder if it'd be worth it to try cutting out a small opening above the PSU in cases like that... in the back of the case that is, not the top, unless the top of the case is above your head and won't ever be seen.

What about placing the psu on top of the case? would that give it better cooling?

nice review



probably they mean operation at 25°C room temperature
(since this was the worst-case spec for their ambient operating conditions)

impressive with the warranty - if it burns, get another!

Hello,

I recently purchased this power supply and received it about 2 weeks ago. It's been running on top of my case since then. (The case is on top of my desk).

My computer's specs:

Opteron 142
2 gigs ecc ram
radeon 9600 pro
spinpoint 7200
dvd drive
tv tuner

Overall, I've noticed that it seems to have significantly better voltage regulation than the Seasonic Super Silencer 400 watts that it replaced. Running outside the case, the temperature status light is always green. My motherboard (Asus Sk8n) reports the 12 volt line as 11.3 volts, which I am pretty sure is an innacurate reading because my machine has been perfectly stable.

One negative: After a few days, I've noticed that there is a soft, very high pitched, constant whine coming from my computer. Originally, I assumed it was originating from the hard drive, because it sounded like a much softer version of the whine that my old western digital hard drive makes. After unplugging all my fans and the hard drive, I noticed that the whine was still there! Ultimately, I ascertained that the silverstone ps is indeed making this high pitched whine.

So, now I want to put this fanless power supply inside a case to reduce or eliminate the whine. I'm surprised that the review recommends an aluminum case, when aluminum cases were always so criticized before. I was planning on putting the power supply into an evercase 4252 with sound dampening material (and a single 120 mm exhaust fan), but now I am a bit concerned that this would perhaps cause the power supply to overheat. Is it wise to have sound dampening material on the top part of the computer case above the power supply? From my understanding, if there is no contact with the power supply on the top part of the case, then different case materials will not give any appreciable differences in power supply temperatures. So, the recommendation for an aluminium case is to improve thermal transfer for the part where there is physical contact between the power supply and the case. Is this correct?

I was hoping to get a fanless powersupply. However, my computer is now so quiet that spending $159 just wouldn't work for a difference I wouldn't notice now.

The high-pitched whine is a HUGE bummer. Mike, can you confirm this? I think most people don't have the hearing range to detect high frequencies, but (unfortunately?) I do. Samsung Spinpoints are unusable for this reason, although very few people report it as a problem, most likely due to it being undetectable by the average user rather than being a symptom of product variance. It's disappointing to purchase something that's supposed to be "silent", when in fact it emits a mind-numbingly loud squeal. I was about to buy one of these PSUs but it's a lot of money to spend on something which is inherently flawed.

alleycat, as Mike mentioned in his review, and as he does for pretty much every PSU review he does, coil whine (the squeel being discussed) is not particular to the PSU, but rather, the combination of components within the system, in combination with the PSU. In all the tests that Mike ran for this review of this PSU, he did not experience any high pitched coil whine whatsoever. Clearly, Mike's equipment is not the same as Hello's rig.

Even if the systems had identical components, the slightest manufacturing variance will change the whine as well; you could have what's supposed to be two identical builds, but only one whines. Such is the nature of coil whine; there's no way of knowing 100% whether or not you'll get whine without trying it out in its precise final configuration. As a matter of fact, it is also known that overclocking often changes the effects of coil whine; some machines whine more at certain clockrates etc. It's obvious enough, being that a great portion of the people experiencing coil whine experience it only, for example, during full load periods or during certain types of tests or calculations.

-Ed

Also, look out for devices such as cold cathode/flourescent lights. That is the only time I ever ran into coil whine, when using lighting. Now I use LED lights.

Just some thoughts that may be totally without merit.

Is it possible that the reported high whine in the fanless Silverstone PSU might be prevented or mitigated by line conditioning and/or isolation? There is so much hash riding on the power lines these days (much of it due to microprocessors) that "golden eared" audiophiles routinely spend hundreds (if not thousands) of US$ to rid themselves of the plague in their sound systems. Most recording studios employ all sorts of filters, chokes, and isolators. Is it also possible that such a high pitched whine exists in all power supplies but is usually masked by the noise from their own fans? Could the whine also be due to the higher than average AC to DC conversion efficiency?

Well, I already have a power filter on the way, so we'll see if that cures the whine.

Some more thoughts:

I noticed a high pitched whine from my computer before I swapped power supplies -- I just assumed that it was the hard drive because that was the nature of the sound. I swapped the Seasonic and the Silverstone, and I still heard a high pitched whine -- albeit, the sound signature was a little different. I was planning on purchasing a hard drive enclosure, but first I wanted to make sure that the hard drive was indeed the source of the whine. After a lengthy trial of shutting on and off fans (and the hard drive, dvd drive), I determined that the Silverstone was the sole source of high pitched whine in my computer. Of this I have no doubt. I am thinking that the Seasonic Super Silencer also emitted a high pitched whine, but I have not actually confirmed this.

There is some good news in regards to high frequency sounds. High frequency sounds lose volume more quickly in air than low frequency sounds, and sound dampening material is more effective on high frequency sounds. So, I'm hoping that putting the Silvestone in a case under my desk will drastically improve things. As it stands now, the power supply is in the worst position for silence -- at head level about 3 or 4 feet away, outside the case. Still, I am able to hear the whine from at least 15 feet away.

The whine is not loud. I highly doubt the average person would notice it right away. I am not even sure if the average person would even pick up on it if he was asked if he heard the sound. It's possible that I can hear higher frequency sounds better than most.

A note about the upper limits of hearing -- a person's hearing does not drop off suddenly, but gradually. So, if a person can hear a 40 db, 14 khz sound as being 40 db, he/she might hear a 40 db, 15 khz as 35 db, he/she might hear a 40 db, 16 khz as 30 db, and so on. My numbers are probably not accurate, but my point is that a person does not go from perceiving 40 db, 14 khz sound as 40 db to perceiving a 40 db, 14.1 khz sound as 0 db. And if a loud, high pitched sound seems quiet, it can still cause hearing damage. So, a 120 db 20 khz sound that is barely audible can still damage your ears. I am sure that the Silverstone or other power supplies do not pose any hearing damage risks; I only mention this because a main reason I pursue silent computing is to give my ears a "rest", which means lowering both audible and almost audible sounds.

Please be sure to post your before and after results with the power filter! I've been wondering about getting one myself.

I'm at the stage where my computer is silent enough that I can hear the whine of the CRT, the ceiling fan, and the UPS system although only late at night. Interesting how the pursuit of quieting my computer has "trained" my ears although as an older person I have, presumably, less acute hearing than when I was younger.

Speaking of younger days, I can recall reading the Popular Electronics article on the first "personal computer", the MITS Altair 8800. Yes, a 2MHz Intel processor was hot stuff back in 1975. I wonder if the power supply also had a whine?

http://wwwcsif.cs.ucdavis.edu/~csclub/m ... _8800.html

yes, precisely, for this PSU and probably most fanless PSUs, because they all rely at least somewhat on conduction of heat into the PSU case. This is in marked contrast to fan-cooled PSUs, whose heatsinks are always isolated from the casing. If care is taken to decouple mount all other sources of noise/vibration, then the dreaded (by me) aluminum hum should not be a factor.

re -- whine. I had the unit in 3 different load situations:

The test loader; The in-case system described in the review; A p4 testbench system (p4-2.8, used usually for HS testing)

There was no high frequency noise from the PSU at any time.

Other than PSUs and HDDs, there is another occasional source of high freq noise in PCs: The power components on the motherboard. I have had at least 2-3 different motherboards whose PS coils/caps emitted high freq whine. One was so bad I returned the sample (SFF barebones) as a reject.

Thanks, Ed for your explanation of the causes of coil whine. I suppose the word "buzz" didn't leave the same impression as "whine". Thanks Mike for your comments and thorough testing. One of the points I was trying to make was that I feel 'handicapped' by my ability to discern high-frequency sound, ie it is probably abnormal (my girlfriend thinks so!). It is difficult to know how others perceive the world.