Fanless PSU Torture Test Roundup

A roundup torture test of samples from every fanless power supply line in the marketplace instigated by the Seasonic X-400 Fanless, which cruised through 15 hours of full-load operation at nearly 60°C temperature in the “case”. Which fanless PSUs can actually deliver rated power with virtually no forced airflow?

A computer with no moving parts remains the Holy Grail of silent computing.
There are only two types of moving parts in a computer: Disk drives and fans.
The transition from mechanically spinning drives to solid state drives has clearly
begun, but there is still no practical substitute for the axial fan in its role
as a cheap, efficient mover of air for cooling. Devices that rely on fans for
cooling airflow include the power supply, although the latest high efficiency
models run so quietly that they might as well be fanless at lower power loads.
Still, there’s always been a trickle demand, somewhat frustrated by lack of
choice, for fanless PSUs.

Silent computing’s Holy Grail is still a computer with no moving parts.

Seasonic has always been extremely conservative in its approach to long-term
and high-stress reliability, which easily explains why there has never been
a fanless model in their lineup, until now. Seasonic’s announcement to introduce
fanless power supplies actually came as a bit of a surprise to SPCR. The best
of Seasonic’s own fan-cooled models are now so quiet that there seemed little
point going fanless — with the added cooling and reliability challenges.
The higher power X-series models feature an effective hybrid cooling design
in which the fan simply does not spin below ~200W load.

Perhaps Seasonic’s marketing team recognized the dearth of serious fanless
PSUs, and made a calculated risk that a PSU which only needs to dissipate 10%
of its input power as heat could, in fact, run cool enough and be reliable enough
to wear the Seasonic badge, a mark of serious quality in power supplies in recent
years. Our recent X-400
Fanless review review confirmed that the new PSU is a marvel of advanced
technology, intelligent engineering and quality manufacturing which combine
to deliver what Seasonic promises: Silent, utterly stable DC output, all the
way to full rated power.

In direct discussions, company reps were brimming with confidence that their
new fanless baby could not only perform just like a fan-cooled unit of the same
rating, but even deliver full power under extremely low airflow conditions,
as one might find in a fanless (or single fan) silent PC. This is the claim
that led to the 15-hour long, full-power test without at exhaust fan on the
SPCR PSU load tester in our review — an extreme torture test that the SS-400FL
passed with flying colors. It is a test that we’d never performed before with
any other power supply, fan-cooled or not. As far as we know, no other review
web site has published any articles of fanless PSUs under such test conditions.

The question that arose naturally was, "How would other fanless
power supplies fare in such a test?"

The question was the genesis of a two-week project that culminates with this
article. It is a roundup review of fanless power supplies, in a test setting
that can only be described as torturous. Admittedly, the test is not representative
of practical usage; few users would push their PSUs to steady-state maximum
load for 15 hours in red-hot conditions. The reality is that most people who
buy and use a fanless power supply are not looking to push the power or thermal
envelope of their PC — usually they’re looking to build a silent computer,
most often with less-than ultra-power-hungry components. The issue this roundup
addresses is just how close to the rated power of a fanless PSU you can run
before the cooling requirements of the PSU itself makes the system noisier than
just running a good fan-equipped PSU.

FANLESS PSU BACKGROUND

Fanless power supplies for DIY computers began appearing in significant numbers
some half a dozen years ago. Silent computing was only just beginning to scratch
at mainstream tech consciousness, but there was a burst of activity among power
supply providers to capture early adopters’ disposable cash. They were invariably
much higher priced than conventional fan-cooled models. Not coincidentally,
a scan of the fanless PSU’s history in computing is a stroll through SPCR’s
review archives. Our focus on quiet/silent PC products means we’ve examined
almost every fanless PSU that made it to market in the past decade. This hands-on
experience puts us in a unique position.

SPCR’s first close look at a fanless PSU came in June 2003, when the TK
Power 300 was examined in an article about a fanless VIA C3 PC built
by reader John Coyle. That was a real DIY effort, as the TKPower 300 was not
an ATX form factor device. Our own sample proved only marginally reliable, and
we never posted an official review.

The Silentmaxx
ProSilence PCS-350W, also reviewed that month, was an ATX form
factor PSU with a hefty external heatsink fins sticking out the back. It was
a massive fanless unit that set the pattern for the breed. The basic concept
was to take a standard ATX power supply and rebuild it with massive heatsinks,
often turning the casing into an extension of the internal heatsink. The general
assumption is that there will be some peripheral airflow in the case/PC; most
DIY systems have at least one case fan, and a fan on the CPU heatsink. Sometimes
a higher power model was used and derated for hotter operating conditions; the
components from a 500W model, for example, might be used in a fanless PSU rated
at 350W. This was to try to ameliorate the higher failure rate that surely prevailed
at a time when power supply efficiency rarely went much past 70%. This meant
that to deliver 200W to the components in a PC, a power supply had to draw nearly
300W, and nearly 100W of waste heat had to be safely dissipated from the power
supply. Without a fan, that’s a challenging task. Not surprisingly, there were
reports in the SPCR forums of many failures with the ProSilence PCS-350W.

Coolmax Taurus
CF-300, reviewed in 2004, also used the heatsink-case concept, but in
a lighter build altogether, perhaps with more modern circuitry. The efficiency
ran a bit higher, climbing up into the high 70s, and overall performance was
not bad. Lots of vents allowed air from peripheral devices to flow through it.
Unfortunately, our sample began showing glitches toward the end our testing,
and we could not recommend it in good faith. Coolmax is still offering fanless
PSUs today, however.

The SilverStone
ST30NF ATX12V PSU entered the fray in 2004. Manufactured by Etasis,
this 300W PSU was the first to employ much higher efficiency (80%) circuitry,
the obvious route to making practical fanless power supplies. It was of a much
higher quality than the Silentmaxx, with the entire extruded aluminum casing
acting as a heatsink and heatpipes used to conduct the internal heat to the
casing. It also featured modern auto-range AC input, APFC for very high power
factor, and even a thermal indicator LED on the back panel. A fair amount of
vent holes allowed some airflow, but the interior of the unit was quite tightly
packed. Silverstone claimed that the circuitry and components were from a 450W
fan-cooled design, which is credible. Our ST30NF sample became a permanent fixture
in the lab, called to perform silent duties for many applications for years.
This is the longest-lived fanless PSU, still
offered in the market today, along with a higher power, more efficient 400W
model, the ST40NF. We reviewed another variant, the ST45NF,
in late 2008. That model oddly had a maximum power rating of 450W with 220VAC
input and 400W with 110VAC input. The ST40NF may simply be a derated, re-tagged
version of the ST45NF model we reviewed.

Antec, one of the biggest case/PSU brands, introduced the Phantom
350 in the fall of 2004. This was another heatsink casing design at
least cosmetically similar to the Silverstone ST30NF, with similarly high efficiency
(in the low 80s). While it, too, offered much higher build quality than earlier
fanless efforts, it must have failed at a higher than normal rate, for Antec
introduced a variant a year later, the Phantom
500, the same basic design, but with a fan that turned on beyond a certain
internal temperature. Neither of these models is offered in the market today.

Fortron-Source Power
(FSP) came to the game in late 2005 with the 300W
Zen, which deviated from the previous heatsink-casing models. The top
was an extruded aluminum fin piece typical of the fanless breed, but the rest
of the casing was mostly open mesh. The mesh cover allowed greater airflow through
the Zen, which meant that an exhaust fan mounted on the back panel of a case
would tend to pull outside air through the PSU, helping to cool it in the process.
It also featured high efficiency, reaching near mid-80s, the highest seen to
that point, and modern features such as APFC and auto-range AC voltage input.
. The next year, a virtually identical version with upgraded components for
400W rated power appeared rebadged as the Silentmaxx
Fanless 400W MX460-PFL01 power supply. Silentmaxx had abandoned
its earlier fins-out-the-back design for the more integrated and modern Zen.
This 400W fanless model is also sold under the FSP badge, as well as Amacrox.

There were also fanless PSUs built into the now-discontinued, massive fanless
TNN-500 and TNN-300 "heatsink cases" by Zalman, as well
as the fanless Hush mini-ITX and Hush ATX HTPC cases/systems made
by Hush Technologies. These power devices were integral to the cases, however,
not in standard ATX form factor, nor availble for separate purchase.

One fanless power supply we never reviewed was the SilenX Luxurae series.
These PSUs appeared some time around 2004-5, but they were largely ignored by
the SPCR community due to the
dubious marketing tactics of the brand’s founder. It’s unclear whether SilenX
is actively selling or promoting the series, as the Luxurae line does not appear
on the current product pages, yet they are listed at steep prices at two SilenX
online sales sites, as well as one or two other online retailers.

CURRENT FANLESS PSU OFFERINGS

A search for fanless computer PSUs with online retail engines turns up familiar
names but not many: Amacrox, Silentmaxx, Fortron-Source Power (FSP), Silverstone,
Coolmax, and SilenX. The first two are differently-badged variants of the fanless
PSU made by FSP as the Zen. We’ve reviewed both the eariler 300W FSP Zen and
a more recent 400W model under the Silentmaxx brand. We also had several samples
of the fanless PSUs currently available (Silverston and FSP), but since they
had been in use in the lab for some time, with unknown attendant wear and tear,
we decided to obtain brand new units. Rather than contact the manufacturers
and wait for available freebie samples, we purchased several from a couple of
different sources. The only one not purchased was the Seasonic SS-400FL, which
is a retail package sample that came directly from Seasonic. (A note for those
who are suspicious of gift horses: There is absolutely no indication that Seasonic
has ever "tweaked" free review samples for unusually good performance.
In a quiet power supply, the only really tweakable thing might be the fan or
fan controller; in this case, there’s no fan at all.)

The fanless contenders:

• Seasonic X-series SS-400FL
• Silentmaxx fanless 560W — a rebadged Fortron-Source Power (FSP) Zen
  400W
• SilenX Luxurae 460W
• Silverstone Nightjar ST40NF
• Coolmax CF-480B

As far as we know, there are no other fanless ATX12V power supplies available
on the retain market.

These five boxes are representative of all the fanless PSU lines available to the DIY computer builder today. Clockwise from top: Silentmaxx fanless 560W, Seasonic X-series SS-400FL, SilenX Luxurae 460W, Silverstone Nightjar ST40NF, and Coolmax CF-480B. There’s some question whether SilenX is still actively offering the Luxurae series.

One thing that became clear during our online shopping for these samples is
that computer retailers generally don’t carry fanless PSUs. There are only a
few sources in the US or Canada, and they are only a bit more numerous in the
EU. Why? A quick, somewhat random survey was done of a handful of computer component
retailers in the US and Canada, mostly by phone. Where a marketing or stock
manager was available, here are the things he/she had to say:

"The returns killed us when we carried the ‘______’ fanless power supply.
Anything close to a 10% return rate we drop like a hot potato."

"The demand is just not there, we need a certain number of turns to
justify carrying the inventory. Now, we bring in small numbers when orders
come in."

"Reliability, user confidence and demand hasn’t been good, so we stopped
carrying them."

"No one really cares, so many fan-cooled power supplies are already
so quiet. We carry a couple and maintain small quantities, the Silverstone
is our best seller but it’s still just a trickle."

Any new fanless PSU must overcome the history of the fanless breed and the
skepticism of retailers.

FANLESS PSU APPLICATIONS

The silent PC DIY enthusiast is an obvious target market, but it isn’t that
big. So what are the other potential markets for fanless PSUs?

• Specialist system integrators who build high end silent systems, especially
  home theater PCs where buyers tend to be well-heeled and less price-conscious.
• Specialist laboratory PCs that must not add to either noise or airflow in
  the test rooms.
• Digital audio workstations for music production.
• Hospital PCs where high reliability, low noise and no airflow are all highly
  valued. In a premium PSU, the fan is the least reliable part. Aside from the
  usual bearing-related issues, PSU makers report that fan-related returns include
  regular instances where the fan blades rub against its own wires, causing
  an annoying noise.

THE TORTURE TEST

We’ve been running a complete set of tests for PSUs for many years with minor
adjustments here and there. You can read all the details the tests in SPCR
Power Supply Test Rig, v.4.1. It can be summarized thus:

• A PSU load tester customized to provide up to ~1,000W loading, with three
  separate load circuits for 12V lines.
• Both current and voltage are monitored on each output line and manually
  summed to obtain the power load.
• We employ the same loading formula that Intel and 80 Plus
  uses for PSU testing, and multiple load test points are checked from a low
  of 20W to full rated power, with 15~20 minutes of operation at each test point.
• Voltage regulation, power factor, noise & ripple, efficiency, low AC
  input performance are tested at every test point. Efficiency at 220~240VAC
  is also checked for a few test points.
• Noise and temperature are closely monitored, the latter with the enture
  test rig in a 10~11 dBA ambient anechoic chamber.
• The heat from the load is in the box where the PSU is mounted, with a single
  120mm fan used for cooling exhaust.

For this roundup, the test has been modified:

• Low power tests (under 250W) conducted quickly to establish efficiency,
  power factor, electronic noise, voltage regulation and other parameters. Only
  120VAC input used.
• 250W load test run for 15 hours continuously while monitoring all
  parameters periodically, with no exhaust fan in the hot box. The basic rationale
  for this test is that any fanless PSU rated for 400W or higher maximum load
  should be able to handle 250W indefinitely. 250W is a fairly typical maximum
  peak demand for the kind of system most DIY builders would use with a 400W
  PSU.
• Full load test run for 15 hours continuously while monitoring all
  parameters periodically, with no exhaust fan in the hot box.
• If Over Temperature Protection (OTP) shutdown occurs, wait till the power
  supply can be powered up again, then run the test with the Nexus 120
  exhaust fan, starting at 800 RPM. If OTP shutdown recurs, retry with exhaust
  fan at 1060 RPM (maximum speed of Nexus 120). If OTP shutdown still occurs,
  then try a Scythe Gentle Typhoon 120mm fan at full speed (rated for
  1450 RPM and more than double the airflow of Nexus 120). If OTP shutdown,
  stop test.
• Don’t run the test in the anechoic chamber, which is not ventilated, but
  in the open adjacent room to ensure some air circulation.
• Place the PSU for best airflow/cooling, ie, either right side up or upside
  down. The Seasonic has an advantage with its open vented side facing up here,
  so let’s give all the others a bit of an edge by optimizing cooling for them
  in our test setup, even if the PSU does not have the screw mount holes to
  allow upside down mounting. Remember, this is not a realistic test anyway,
  it’s plain torture.

The OTP function in the tested PSUs could be critical to successful completion
of the torture test, one measure of which is to survive it, at least. If the
components in the power supply get too high, failure can occur. Hopefully, without
drama, explosions, sparks or flames. OTP is supposed to halt operation before
excessively high temperature is reached. In modern PSUs, normal operation is
usually restored after OTP shutdown when the internal temperature drops to a
safe level, and the power switch is reset. One simple reason for the seemingly
arbitrary 15 hour testing time: Safety. It’s about as long as we could stand
to be hovering near the test rig to ensure that some catastrophic failure would
not occur unwitnessed.

The SPCR PSU tester is a Frankenstein mix of a commercial load tester and a home-brew hotbox. The box simulates a typical ATX tower case, with the PSU positioned just above the rear panel exhaust vent. The bottom, back and one side of the PSU is exposed to the heat generated in the box, which comes from banks of loading resistors in the box and in the commercial load tester. The PSU has the thermal benefit of being open to the outside air on the top and one side, unlike in a case. With no exhaust fan at the exhaust vent, the air temperature directly below the PSU can exceed 60°C at 400W load. This is hotter than any computer enthusiasts would tolerate in their own PCs. Four 80mm fans running slowly at 5V keep the DBS-2100 commercial load tester cooled; they also blow the heat into the hotbox. Resistor banks in the hotbox account for more than 50% of the heat when the test load is at or above ~400W.

Vancouver’s climate is relatively mild, and over the years, most PSU tests
have been conducted in 20~22°C ambient temperature. July and August were
hot in 2010, with just trace amounts of rain and peak temperature on most days
reaching 25~32°C. That meant the ambient temperature in the test room was
typically 23-24°C, and often several degrees warmer. It also made OTP shutdown
much more likely than at lower ambient temperatures. No tests were run when
the ambient temperature in the room (at several points around the load tester)
was higher than 24°C… which is one of the reasons this project took a
couple of weeks. The ambient noise in the room measured 14~15 dBA — quiet
enough to hear any excessive electronic squealing, buzzing or whine (from the
PSU under test) worth measuring or recording in the anechoic chamber.

TEST RESULTS

A. SEASONIC
X-400 FANLESS (SS-400FL)

We might as well start with the model that sparked this project. You can read
the entire detailed review
so just some key photos and info are presented here along with the torture test
results.

It’s well ventilated on almost every side… …even the normally unvented "bottom". The X-400 (and soon-to-come X-460) is the only modular cable fanless PSU, not only in our roundup, but in the world market. The label is in several languages and boasts a myriad of safety approvals.

If you’ve read our review of the X-400 Fanless, you already know the results,
but here’s the summary in the format we’ll be using for all of the other torture-tested
PSUs.

ANALYSIS

As you will probably have read already in our review of the X-400 Fanless last
week, the sample performed flawlessly throughout the torture test, with no change
in the high quality of the DC outputs. About the only significant change was
in efficiency; during the hottest times of the day, AC power input rose slightly,
by about 10W. This is expected and natural.

When forced to Over Temperature Protection shutdown, the sample did so without
any fuss, and recovery occurred with power off/on when the additional heat was
removed.

The Seasonic X-400 sets the standard by which other fanless PSUs must be judged.

B. COOLMAX
CF-480B

Coolmax offers several fanless models, 350W, 400W and 480W; the last is the
one we’re examining here. We
reviewed a 300W fanless Coolmax previously. It’s surprising that the current
model looks so similar, considering the first sample was received was over six
years ago. There have been tremendous improvements in computer PSUs in those
years.

There were inconsistencies in labeling and specifications between the web site
and the PSU label. Still the presence of the RoHS sticker, a three-year warranty,
and more current/power on the 12V lines than on the lower voltage lines (according
to the
web site info) are encouraging.

Modest size/appearance retail box for the Coolmax. The photos on the box are slightly misrepresentative; the actual PSU has no vents on the sides. There’s no printed manual, but a sheet with links to download a PDF copy. The PSU itself uses extruded aluminum for its outer casing, with large vent grills on the front and back. It weights about 4 lbs. There are no vent holes on the sides, top or bottom. The bottom is a serious heatsink-style extrusion… which takes the place of a 120mm fan. The inside panel is quite an open grill. The front/back open grills enable cooling airflow through the PSU, especially if there is an exhaust fan on the back panel of the computer case; it will tend to suck air from the outside through the PSU. The output cables are mostly quite long, and only the main 20/24-pin ATX cable is sleeved. There is one PCIe 6-pin connector. Even though the label suggests a switch for 115V/230V, there is no such switch. The PDF manual also states the Voltage Selector should be set to 115V or 230V. One assumes the PSU was upgraded for auto-range AC input but the manual and tags not changed. 230VAC operation was not checked. NOTE: The output specs on the label differ from those given on the product website; we used the web site info for our loading formula.

ANALYSIS

The CF-480B turned out to have efficiency in the low 70s, similar to the first
fanless PSU from Coolmax we reviewed all those years ago. The change to auto-range
AC voltage input was not accompanied by an improvement in efficiency, but Active
PFC was probably part of the change, as the PF was quite high.

Overall electrical performance, apart from the low efficiency, was solid throughout
the short term testing. Electronic noise / whine was modest enough to be ignored.
Not surprisingly, the unit ran quite hot, with the casing reaching over 80°C
in spots at 400W load. At 400W output, the Coolmax has to dissipate a whipping
160W of heat, compared to just 44W for the Seasonic X-400.

The sample passed the 250W torture test without issue. OTP shutdown occurred
very soon after the 400W test was started. It restarted after cool-down without
any problem, but overheated and shut down again in a few minutes. The Nexus
120 fan at full speed was added to blow air out of the hot box. At that point,
it ran successfully for a couple of hours, but OTP shutdown occurred when the
ambient room temperature exceeded 26°C in early afternoon. The Scythe Gentle
Typhoon 120mm fan was installed as the exhaust fan and run at full speed (~1500
RPM). The increased airflow kept the hotbox temperature under 43°C, and the
Coolmax successfully ran at full power for 15 hours.

C. SILENTMAXX
MX-560PEL01

The Silentmaxx 560, along with the Silverstone offerings, are the most widely
available fanless PSUs today. We tested a version of this power supply, the
Silentmaxx
MX460-PFL01, some four years ago. In fact, it’s still being used in a SPCR
lab heatsink test platform system. Both of these PSUs are made by Fortron-Source
Power (FSP), who introduced the 300W Zen many years ago.

The 560 looks very similar to the earlier 460, the main visible difference
being a heftier external base made of extruded aluminum for additional cooling.
It’s worthwhile to compare the specifications of the two models:.

The label and output specs on the Silentmaxx MX-560PEL01.

The differences in the capacity of the individual lines are minor. In the newer
model, the 5V and 3.3V lines deliver 15W less, while the 12V lines deliver 30W
more. But the essential piece of information is the total power rating: It’s
400W for both models. The peak power rating is 460W for the MX-460 and
it’s 560W for the MX-560, which suggests that both model naming and power rating
are marketing exercises; they are really 400W models. We deplore Silentmaxx’s
deliberately misleading naming scheme.

The MX-560 is tagged as ATX12V V2.2 PSU Guide compliant.

The Silentmaxx box looks designed for retail store display. Like the earlier 460, it is vented all around except for the top. All the attached output cables are sleeved. A 20/24-pin main ATX and a 6pin PCIe are among the output connectors. The bottom is a heavy heatsink-style extrusion with large vent openings that show big internal heatsinks. This is a heavy PSU, weighing nearly 6 lbs. The open vents enable cooling airflow through the PSU, especially if there is an exhaust fan on the back panel of the computer case; it will tend to suck air from the outside through the PSU.

ANALYSIS

The Silentmaxx displayed solid performance in the short-term testing, with
excellent voltage regulation, high PF, and good 80 Plus Bronze level efficiency.
There were no anomalies of any kind, and electronic noise was low enough to
be considered silent. Our impression was much the same as with the Silentmaxx
460 sample we tested four years ago and continue to use in the lab: A fairly
efficient and silent workhorse.

The 250W 15-hour test passed without any issues. About an hour into the 400W
test, the unit failed. Unfortunately, this event was not witnessed, but a post-failure
forensic analysis (heh) showed that the AC power meter through which the PSU
was powered had blown its 20A protective fuse. This suggested that the protective
circuits (thermal or output overload) failed to trip, and the Silentmaxx suffered
a terminal breakdown. It could not be powered up again. Some of the data had
not been collected when the unit blew, which is why it is not reported.

We were surprised at this failure. FSP, the manufacturer of the PSU, has a
long history in its field and it is reasonably well-respected. Our 4-year old
lab sample of the earlier version continues yeoman-like — albeit in open
air and usually with <200W loads, rarely for more than 8 hours continuously.
We had expected a possible failure with one of the older designs (SilenX and
Coolmax).

D. SILVERSTONE
ST40NF

The ST40NF looks very similar to the ST45NF, a sample of which was reviewed
less than two years ago. These fanless PSUs are made by Etasis
Electronics. The much earlier 300W ST30NF was reviewed back in 2004.
Like the Silentmaxx, the ST40NF appears to be a re-named version of the earlier
ST45NF — but in this case, the product has been de-rated instead of up-rated.

The specs of the Silverstone ST40NF.

The ST40NF is a modern 80 Plus Bronze approved PSU. It has two PCIe power connectors,
one with 6-pins, and another with 6/8 pins. At over 6 lbs, it is the heaviest
of the PSUs tested here. Unlike the earlier Silverstone fanless PSUs, the ST40NF
does not features an extra indicator LED that changes color to show when thermal
overload protection is engaged.

The big Silverstone box screams ZERO dBA but is otherwise tastefully silver in color. The ST40NF is well packaged and presented, with a nice manual, Ac cable, zap straps and velcro ties for cable management. The top panel of the PSU is the only one without vent holes. All five other sides are vented. The bottom is a heavy aluminum extrusion with large vents that show big internal heatsinks. This is a heavy PSU, weighing over 6 lbs. Ignore color and the similarity to the Silentmaxx is striking. The open vents enable cooling airflow through the PSU, especially if there is an exhaust fan on the back panel of the computer case; it will tend to suck air from the outside through the PSU.

ANALYSIS

The Silverstone had good performance in the short-term testing, with excellent
voltage regulation, high PF, and good 80 Plus Bronze level efficiency. Unlike
the ST45NF, the ST40NF has low standby and no-load power consumption. There
were no anomalies, and electronic noise was very low. Our impression was much
the same as with the ST45NF sample we tested two years ago and continue to use
in the lab: An efficient, silent workhorse.

Our sample passed the 15-hour 250W torture test without any problems. Only
about 10-15 minutes into the 400W test, OTP shutdown occurred, at only a bit
over 40°C. After a couple minutes of accelerated cool-down with a small room
fan, the unit started up and ran at full power. There was no sign of damage.
However, thermal shutdown recurred in the same way. The usual Nexus 120 fan
was installed and run at 12V as an exhaust fan for the hotbox to see if reducing
the temperature would keep the unit running at full power. OTP shutdown happened
again, this time about 20 minutes into the test when the internal temperature
of the hotbox exceeded 41~42°C.

Finally, the Scythe Gentle Typhoon 120mm fan was installed as the exhaust fan
and run at full speed (~1500 RPM). The increased airflow kept the hotbox temperature
under 40°C, and the ST40NF successfully ran at full power for 15 hours.

Although the conservatively set OTP circuit worked well and protected the PSU
from damage, the torture test findings suggest that the ST40NF must be well
isolated thermally from other components and well ventilated to work in a system
which demands 400W. It’s probably not possible to do this without a fan that
has substantial airflow near the PSU.

C. SILENX LUXURAE 460
– $240 MSP

The SilenX fanless power supplies have not been reviewed by SPCR before. It
is a brand we’ve never held in high regard, for a variety of reasons. The caption
comments below show how this product reinforced our opinion.

There’s some question whether the Luxurae line is still in production. SilenX
USA’s on-line store lists them, but the SilenX
web site has no mention of them. Documentation is sparse and somewhat conflicting.
For example, the web site claims ATX 2.10 compliance, while the PSU’s label
adds ATX12V v1.3 compliance. There’s also a reference to a 10A / 250V fuse rating,
presumably on the AC input.

The enormous current/power available on the 3.3V and 5V lines, combined with
the modest juice on the 12V line suggests that the ATX12V v1.3 appellation is
correct. Modern systems draw 90% of the current on the 12V line, and modern
PSUs reflect this balance.

Despite the allusion to luxury in its name, there’s nothing luxurious about the SilenX Luxurae packaging, which is a plain white carton with a color sticker on the cover. Points for eco-packaging, perhaps, but little else. The 60mm fan mounted outside the shiny black painted PSU screams for one’s attention. A fan on a PSU marketed as fanless with a "0 decibels" notice on the cover. Hmmm… There s a big heatsink on the back with 8 fins. Large aluminum extrusion heatsinks are also visible inside, through the open top grill. AC cable screws and a couple of photocopied letter sheets, one with a big warning notice. This is an amazing warning: Don’t use the bottom row of standard screw mounting holes lest you damage the internal circuit board! So does that mean only the top screw mounting holes should be used for this >5 lb beast? Take a look at the back panel of the PSU, and we find… … the screw holes that should not be used. There are five others, but only one near the bottom in a non-standard location. The normal case will have only two screw mounts along the top row. So mounted as instructed in a typical case, the SilenX Luxurae’s 6 lb weight will hang off two screws along the top of the PSU’s back panel. Amazingly mickey mouse. While we’re looking at the back panel, note that there is a manual 230/115VAC switch rather than the almost ubiquitous auto-range 100~240VAC input in modern PSUs. Passive Power Factor Correction is used. There’s also a little knob which switches on with a click: This is a manual fine output voltage adjustment that raises or lowers the output voltages, presumably to compensate for voltage sag under load. It’s a concept that disappeared from modern PSUs years ago; they now have voltage regulation that’s tight enough so no one has to "tweak" it any more. Here’s how that fan is powered. It runs at full speed via an adapter connected to one of the SilenX PSU’s 4-pin output connectors. If you disconnect the cable at the join closest to the fan, there will be a live exposed 2-pin connector that could short the 12VDC line — and take out other components, too. The fan is secured with three (instead of four) vibration reducing plugs. It falls off the plugs quite easily, by accident. There are no instructions about why or when the fan should be used. The presence of this output connector suggests an ancient design. It’s been so long out of use that we had to hunt around to remember that it is a 6-pin auxiliary power connector for +5V and +3.3V lines used a decade ago on dual-CPU AMD motherboards. The web page where we found this info stated, "You’re more likely to sight Bigfoot than a motherboard which uses this connector." Speaking of connectors, there is no PCIe 6-pin or 8-pin output connector for video cards that demand it. Finally, one of the skimpiest PSU labels.

All of the above are good reasons for low expectations even before the first
power-up. The external fan was removed and disconnected for the test; it is
supposed to be a fanless PSU.

ANALYSIS

The SilenX Luxurae 480 emitted prominent electronic whining noise upon turn-on.
It diminished and increased with different load combinations, but it never disappeared
completely. For some users with sensitive hearing, the sharp, audible tonal
quality of this electronic noise would make this PSU unacceptable at any power
level. Efficiency would have been considered good 6~7 years ago, but today,
it doesn’t even come close to reaching 80 Plus standard, which is almost a minimum
requirement. None of the measured electrical parameters can be considered good:
Voltages fell nearly out of spec at just 250W load, the power factor was low
as expected, and the ripple and noise set new standards for poor performance.
It was measured on the 12V line about 20 minutes into the 250W long term test,
and found to be peaking around a volt. That’s not a typo. A whole volt, when
a maximum of 120mV is recommended. Such a high reading of the ripple/noise on
the oscilloscope was a new experience; it was double and triple checked to make
sure.

The long test at 250W load was halted after just one hour, and the full power
test was not conducted at all. There was no point doing any testing beyond 250W,
because the ripple and noise was so high at this load as to be risking damage
most components in a PC, and the unit was emitting such a loud. piercing whine
that it was audible from a carpet adjoining room more than 25′ away. It sounded
somewhat like a "dirty" whistle, with multiple tones. It would be
considered unusable by anyone who can hear. The efficiency dropped steadily
throughout this test period as temperature in the hotbox rose. At the start
of the test, it was consuming about 318W from the AC line; by the time the test
was stopped an hour later, the AC power consumption had risen to nearly 350W.

The SilenX Luxurae 460 had the dubious distinction of being the only PSU in this roundup to require frequency spectrum analysis. The electronic noise reached 27 [email protected] in the live lab room; it would have measured a few dBA lower in the anechoic chamber, but still way higher than most of our top ranked fan-cooled PSUs at this power output. (If there is demand, an MP3 format recording of this noise can be posted.)

Our sample SilenX Luxurae 460 failed to qualify as a good PSU in any way. It
still powers up on the PSU tester, but we will not risk any other components
to see if it can actually run a PC.

It’s possible that this 460 is a bad sample… and we happen to have also obtained
an identical-looking 400W SilenX Luxurae. Without going into details, we can
report that the 400W sample performed pretty much the same, that is to say,
poorly, with less electronic noise that was still obviously audible from a couple
meters away. It, too, was unacceptable for the same reasons as the 460W Luxurae.

FINAL CONCLUSIONS

That the Seasonic X-400 Fanless would excel in this roundup was a forgone
conclusion. Our full review
last week showed that the X-400 Fanless passed our torture hotbox test at
400W for 15 hours without any problems whatsoever. It seems able to run indefinitely
at full rated power with minimal airflow (at ~24°C room temperature). Efficiency
and electrical performance is top notch. The X-400 is an SPCR Editor’s Choice
product and our #1 recommendation among fanless PSUs. Add totally modular cables
and a 5-year warranty, and the $140 MSRP looks like good value in this roundup.

The Silverstone ST40NF is a solid second choice, with 80 Plus Bronze
Efficiency (>87% at 50% load), good electrical performance, proven market
longevity, and an effective OTP system. The latter seemed a little too aggressive,
kicking in at a fairly modest 41~42°C in the hotbox. While the unit protects
itself well from thermal damage, you will be hard pressed to get anywhere near
the 400W maximum output without resorting to a fair amount of forced airflow
in your system, which negates the reason for a fanless PSU in the first place.
But up to perhaps 250~300W, the ST40NF should be able to run well with minimal
airflow. The typical $160~170 market price puts it at a distinct disadvantage
against the Seasonic. (Postscript, 23 Aug 2010: Silverstone informed
us that effective immediately, the market price of the ST40NF will be reduced
to $130~140.)

Even though our sample of the misleadingly numbered Silentmaxx MX-560PEL01
was the only one to fail and become inoperative in the torture test, we still
put it (and its variously badged brethren) in third place… but it is hard
to recommend without caveats. The efficiency and electrical performance is similar
to that of the Silverstone, as its overall design and build. Obviously, the
OTP circuitry is remiss, allowing the unit to overheat and fail without intervention.
Still, if the total load is kept under 250W, we think it would work fine with
at least a bit of airflow. Any higher, and you’d be well-advised to add significant
forced airflow. The price seems to be extremely variable; we’ve seen the FSP
branded 400W version with 2-yr warranty for as little as $110 in the US, but
the identical Silentmaxx "560W" version as high as € 175 in Germany
(warranty unknown). (Postscript, 23 Aug 2010: Why this sample failed
when the virtually identical FSP-sourced Silverstone ST40NF did not is a bit
of a mystery. It is possible that the OTP circuit in the Silentmaxx is tweaked
to allow higher temperature operation without shutdown while that in the ST40NF
is set for best protection. Perhaps the aggressive OTP setting is required for
this PSU to run safely.)

The Coolmax CF-480B fared far better than expected, given its
low efficiency. Electrical performance was quite solid even through the torture
test, but it ran so hot that caution is advised. Its OTP feature worked well
to protect itself, but the low efficiency requires a lot of forced airflow at
400W or higher for the unit to work without thermal shutdown. The high power
consumption, hot operation, and typical market price of $100~120 in the US makes
it a tough sell; at least Coolmax offers a 3-year warranty. Still, most users
would be better off with a more modern, higher efficiency, quiet, fan-cooled
PSU.

The SilenX Luxurae 460 proved to be poorly documented, and a poor performer
in almost every way. It had only one edge over any of the other contenders:
Higher efficiency than the Coolmax. In every other way, the SilenX did not make
the grade. The horrific electronic whining noise at 250W load was enough after
just an hour for us to pull the plug on it and fail the unit. The MSRP of $240
is a joke; we obtained it from Directron at a sale price of $46 with a 1-year
warranty. We’re hoping they will take it back. Enuf said.

Only time can really tell, but it seems safe to say that with the X-400, Seasonic
has single-handedly transformed the fanless PSU into a really viable product,
not only for the users but for the resellers.

* * *

SPCR Articles of Related Interest:
Power Supply Fundamentals

Recommended Power Supplies
SPCR PSU Test Rig V.4.1
Seasonic X-400 Fanless
Silverstone Nightjar ST45NF:
450W Fanless
Silentmaxx
Fanless 400W MX460-PFL01 PSU (FSP Zen)
Coolmax Taurus CF-300
Tiny, Silent and Efficient: The picoPSU
Winmate DD-24AX DC-DC Module for Silent,
Efficient Power

* * *

Discuss
this article in the SPCR Forums.