SilverStone ST30NF Fanless ATX12V PSU

Power
Viewing page 5 of 5 pages. Previous 1 2 3 4 5

IN-SYSTEM TESTING

This is not a standard procedure for all PSUs, but fanless ones do require extra attention. The test system is a mid-high end one similar to that used in the Coolmax Taurus Fanless PSU review but with a different case.

  • Chenbro Xpider case
  • Zalman 400B PSU (ATX12V v1.3) modified with Panaflo 80L fan
  • AOpen AK89 Max (nVidia3 Athlon 64 board)
  • AMD Athlon 64 3200+ (Cool 'n' Quiet enabled)
  • Zalman ZM7000A-AlCu heatsink w/fan at 6V (via SilentBIOS of motherboard)
  • ATI 9800 Pro VGA with Arctic Cooling VGA Silencer (fan set to low)
  • Crucial PC3200 memory, 256MB
  • Samsung MP0402H 40G 2.5" 5400rpm notebook HDD (for Windows XP OS), placed on foam in 3.5" drive bay
  • Seagate Barracuda V SATA 80G hard drive (for data), placed on foam just behind front intake fan
  • Asus QuietTrack DVD/CDRW Drive
  • Panaflo 92mm low speed exhaust fan set to 7V
  • Panaflo 80mm low speed intake fan set to 7V
  • Case placed on thick piece of closed cell foam to insulate vibrations from hardwood floor

This case is set up with a PSU intake vent and duct. The top optical drive bay is left open, and a piece of close cell foam divides the PSU and intake path (from the optical drive bay to the PSU). This ensures that the PSU stays well cooled to ensure its fan never ramps up. The duct/vent was left in place for the ST30NF.



Some temperature and power measurements were taken on the above machine. Then it was shut down and the modified Zalman PSU swapped for the SilverStone ST30NF PSU. No other changes were made.


SilverStone ST30NF simply swapped in place of the Zalman PSU.

A thermistor was placed on the back exhaust grill of the ST30NF (in the same spot as during the lab testing). The following temp measurements were taken after 30 minutes in each state. The ambient temperature for the in-system testing in this room (my office) was 23°C, 3°C lower than in the lab.

PSU
State
Temperature (°C)
AC Power
Noise dBA@1m
CPU
Board
PSU
SilverStoneST30NF
Idle
38
37
47
104W
23
Folding
56
42
52
142W
CPUBurn
64
44
55
155W
Zalman 400B modified
Idle
38
37
42
112W
24
Folding
56
42
45
151W
CPUBurn
64
44
47
166W

Ambient room temperature: 23°C.

THERMAL ANALYSIS

It's evident that the case and CPU temperatures were unaffected by the PSU swap because the PSU intake channel / vent isolates the PSU from the rest of the system. The open duct to the front helped keep the ST30NF stay a bit cooler; in the absence of a fan, the difference was probably not more than a couple of degrees.

More significant is the insulating effect of the duct in keeping the PSU heat from the rest of the system. Chances are, the CPU and case temps would have gone up by a couple of degrees without the PSU duct. As the table shows, the CPU and case temps remained identical with either PSU. The casing temp after 30 minutes of CPUBurn was 55°C. The 155W AC draw puts the DC output load at around 120W; with the 3°C differential applied, this is a couple of degrees hotter than it would have been in the PSU thermal simulation box (on the test bench). Both LEDs on the PSU always remained green throughout the in-system testing.

The top panel of the steel case became quite warm during stress testing, but remained relatively cool when the system was idling. Folding@home is probably fairly close to the max stress a home system would see with real applications. The temperature stayed relatively modest even after many hours of Folding.

The voltage lines were perfectly stable throughout the in-system testing, which took place over a 4-day period. See the screen capture below (showing the highs and lows) from one of the several runs of stress testing.


Screen capture of high/low report in Motherboard Monitor 5 (MBM5).
NOTE: This motherboard does not draw on the -5V line.

CONCLUSIONS

The SilverStone ST30NF is the most capable fanless PSU I have tested or used thus far. The features alone are impressive enough, but more importantly, real performance was exemplary throughout bench and in-system testing.

The isolation of the PSU with the closed-cell foam sheet helped to keep its heat from affecting other components in the in-system test. This is a technique fanless PSU users may want to experiment with. Given the high efficiency of this PSU, even without the isolation, the increase in the temps of other components would have been minor.

The aplomb with which the ST30NF shrugged off the extreme high temperature it was accidentally subjected to is impressive. Few systems can actually draw 250W DC; if you have one that can draw this much power on a long term basis, good airflow cooling around the PSU is recommended. It would be particularly beneficial to cut an opening in the case to expose the top PSU panel.

I believe a key to understanding the performance of this PSU lies in the unsubstantiated comment in the product web page, "a peak rating of up to 400W+." It is likely that the core components could allow this PSU to be fairly rated at over 400W if it was fan-cooled. Despite the modest 250W output rating, this unit can easily handle much more than the power demand of the system used for the in-system test. With some additional airflow it can probably do more than 250W fairly easily.

The ST30NF may be the perfect PSU for use in aluminum HTPC cases like SilverStone's own SST-LC01, where the high conduction of the aluminum would help dissipate the PSU heat more efficiently than a steel case. The SilverStone SST-LC01 deserves a firm SPCR recommendation.

PRO
  • No fan; silent operation
  • No buzzing or whining during in-system trial
  • Very high efficiency
  • Very high power factor
  • Very good passive cooling system
  • Easily meets output claims
  • Excellent voltage regulation
  • Stable in-system operation
  • Great manual
CON
  • What's not to like?
  • Price?
  • No PCI Express compatibility

Much thanks to SilverStone for this ST30NF PSU sample.

POSTCRIPT: Efficiency Correction
October 22, 2005

Recently, we discovered that our power supply testing equipment and methodology were providing erroneously high efficiency results. In general, the biggest errors occurred at higher output load points above 300W. At lower output levels, the efficiency error was often no more than one or two percentage points. No other tested parameters were significantly affected.

Through a fairly arduous process of discovery, analysis and old fashioned problem solving, we modified our testing equipment and methodology to improve the accuracy of the efficiency results and described it all in the article SPCR's PSU Test Platform V.3. As part of this revision, we re-tested most of the power supplies on our Recommended PSU List. In most cases, the same sample was used in the second test.

The corrected and original efficiency results for all the re-tested PSUs are shown in in the article, Corrected Efficiency Results for Recommended Power Supplies. The relative efficiency of the tested power supplies has not changed. If the tested PSUs are ranked by efficiency, the rankings remain the same whether we use the original results or the new results.

This data is also being added to relevant reviews as postscripts like this one.

CORRECTED EFFICIENCY: SilverStone ST30NF
Target Output
65W
90W
150W
200W
250W
Actual Output
65.0W
93.6W
149.6W
200.5W
251.5W
Efficiency
Corrected
75.6%
78.7%
80.8%
80.5%
79.6%
Original
76%
77%
78%
81%
80%

In this case, our original efficiency calculations were very close: Slightly low through to about 150W output, then a touch high at 200W and 250W.


POSTCRIPT #2: Updated Cable Sets
January 11, 2007

As power supplies go, the ST30NF has good longevity. It's been on the market for more than two years, and a lot has changed since it was first released. As such, Silverstone has updated it to keep up with the times. Most significantly, the new revision includes the connections that are now standard in the industry. At the time of writing, the ST30NF now comes with the following connections:

  • 20+4-pin main ATX connector — changes for the old 20-pin plug
  • 4-pin AUX connector — unchanged
  • 6 x 4-pin IDE connectors — unchanged
  • 2 x Floppy connectors — one more than before
  • 4 x SATA connectors — two more than before
  • 1 x 6-pin PCIe connector — this one is completely new

 

 

* * *

Discuss this article in the SPCR Forums.



Previous 1 2 3 4 5

Power - Article Index
Help support this site, buy from one of our affiliate retailers!
Search: