80x25mm Fan Round-Up #1

Table of Contents

Here it is at last! You’ve read about it for years, we’ve tantalized you with a methodology article without publishing results, but we’ve finally come through: It’s the first episode of SPCR’s Calling All Good Fans series of fan round-ups, with detailed noise information about twelve different fans. This round-up has a little bit of everything. The obvious ones: Nexus and Panaflo models that regulars have known and loved for years. Some less known but occasionally recommended fans from Papst and Mechatronics. We’ve even tossed in a Delta screamer so everyone knows what to avoid. And, last but not least, a surprise gives the Nexus a run for its money…

November 13, 2006 by Devon
Cooke

May 5, 2008
Our fan airflow measurement system has recently undergone a major
revision to improve accuracy and repeatability. We’ve updated airflow data
for some but not all fans; only fans that fared well acoustically were retested
with the new system. There will be a new methodology article coming soon.

Silent PC Review’s first fan roundup is the culmination of over a month of
hard work testing, re-testing, and documenting a dozen different fans as we
learned the subtle ins-and-outs of fan testing
. It’s also the first concrete
result of a project we started over two years ago, called Calling
All Good Fans
.

This inaugural review covers a wide and somewhat random range of 80mm fans
(other sizes will be covered in subsequent articles), starting with our current
favorite (Nexus), passing over some past favorites (the two Panaflo models),
tossing in a few that have been recommended by others (Papst, SilenX), and ending
with some that are just plain bad (Delta). While the focus of our selection
was on low speed, quiet fans, we’ve included a wide range of fans to make judgments
easier — the bad fans serve as points of reference just as much as the
good ones.

This roundup is the first in a long series, and is not meant to be comprehensive
or even representative of what’s out there. The only thing that these fans have
in common is that, at some point in the last two years, they were acquired by
SPCR and added to our enormous pile of fans.

This roundup is primarily a summary of our test results with a few interesting
tidbits about each fan thrown in. We have kept theory to a minimum, so you do
not need to know how a fan works to get the most out of this article. You need
to know two things:

  1. Fans are designed to push air — the faster the fan, the more air it
    pushes
  2. Fans produce noise — the faster the fan, the more noise it produces

For our purposes, the best fan is the one that pushes the most air for the
least amount of noise. For users who are interested, a more technical discussion
of fan technologies can be found in our recent article, The
Silent Fan
. Users who want to know exactly how the fans were tested
should refer to our methodology
article
. The rest of you: Sit back and enjoy! We hope you find our work
useful.


This roundup barely dents the veritable mountain of fans we have waiting
to be tested.

HOW TO USE THIS REVIEW

Each fan in this roundup has its own data table and write-up that summarizes
what we learned about it. Use these to find specific information about the fan
you’re looking for. In addition, every fan was recorded four times, according
to our standard Audio Recording
techniques.
These recordings can be used to make A/B comparisons between
fans to help illustrate the differences between them. The four recordings are
as follows:

  1. Alternating ambient noise and the fan running at 5V, 7V, 9V, and 12V,
    recorded at a distance of one meter.
  2. Alternating ambient noise and the fan running at 5V, 7V, 9V, and 12V,
    recorded at a distance of one foot (30 cm).
  3. Five seconds of ambient noise, followed by the fan running in the constant
    airflow test, recorded at a distance of one meter.
  4. Five seconds of ambient noise, followed by the fan running in the constant
    airflow test, recorded at a distance of one foot (30 cm).

As always, we recommend that you listen and compare the recordings in a specific
way. The green box below describes how we make our recordings and what you’re
supposed to do with them.

At the end of the roundup is a conclusion that summarizes the best and the
worst that we found. This is where to look if you just want to cut to the chase
and find out which fan we liked best.

HOW TO LISTEN & COMPARE

These recordings were
made with a high resolution, studio quality, digital recording system,
then converted to LAME 128kbps encoded MP3s. We’ve listened long and
hard to ensure there is no audible degradation from the original WAV
files to these MP3s. They represent a quick snapshot of what we heard
during the review. Two recordings of each noise level were made, one
from a distance of one meter, and another from one
foot
away.

The one meter recording
is intended to give you an idea of how the subject of this review
sound in actual use — one meter is a reasonable typical distance
between a computer or computer component and your ear. The recording
contains stretches of ambient noise that you can use to judge the
relative loudness of the subject. For best results, set your volume
control so that the ambient noise is just barely audible. Be aware
that very quiet subjects may not be audible — if we couldn’t
hear it from one meter, chances are we couldn’t record it either!

The one foot recording
is designed to bring out the fine details of the noise. Use this recording
with caution! Although more detailed, it may not represent how the
subject sounds in actual use. It is best to listen to this recording
after you have listened to the one meter recording.

More details about how
we make these recordings can be found in our short article: Audio
Recording Methods Revised
.

THE FANS

The following fans were included in the roundup:

The Conclusion can be found
on page 8.

NEXUS REAL SILENT CASE FAN SP802512L-03

Ambient noise at the time of testing was 18 dBA.

Brand Nexus Power Rating 0.15A
Manufacturer Bisonic Airflow Rating 20.2 CFM
Model Number SP802512L-03 RPM Rating 1,500 RPM
Retail Availability Yes Noise Rating 17.6 dBA
Bearing Type Sleeve Header Type 3-pin & Molex
Hub Size 1.35″ Starting Voltage 5.0V
Frame Size 80 x 80 x 25 mm Number of Samples 5
Voltage
Noise
RPM
CFM
Power
12V
20 dBA@1m
1540 RPM
22 CFM
0.57W
9V
~19 dBA@1m
1130 RPM
14 CFM
0.42W
7V
<18 dBA@1m
840 RPM
10 CFM
0.34W
5V
<18 dBA@1m
530 RPM
5 CFM
0.25W
@10 CFM (7.2V)
<18 dBA@1m
870 RPM
10 CFM
0.34W
May 5, 2008
The updated airflow results here are the result of improvements
in our testing procedures. They are more accurate than the original results
above, but they are not directly comparable. Please compare these only with
fan reviews published after May 5, 2008 — or ones that have updated
results published in a box like this one.
12V
20 dBA@1m
1540 RPM
18 CFM
0.57W
9V
~19 dBA@1m
1130 RPM
14 CFM
0.42W
7V
<18 dBA@1m
840 RPM
10 CFM
0.34W
5V
<18 dBA@1m
530 RPM
6 CFM
0.25W
@10 CFM (7.0V)
<18 dBA@1m
870 RPM
10 CFM
0.34W


* Much thanks to EndPCNoise for supplying these samples.
*
The Nexus brand name should be familiar to anyone who has spent any length
of time poking though Silent PC Review. Their Real Silent fans are well liked

because they demonstrate two very important qualities:

  1. The default speed is slow enough that they are reasonably quiet even at
    12V.
  2. The noise character is superior to almost every fan we have heard. It doesn’t
    get much smoother than this!

The Nexus is the unofficial benchmark that other fans will have to measure
up to. We are intimately familiar with its noise signature, as we use it in
our own personal systems. Best of all, it is so smooth that it is effectively
inaudible below seven volts. Even at full speed, the only noise is a low, broadband
hum that blends easily into the background.

The smooth noise character can probably be attributed in part to the use of
sleeve bearings, which tend to generate less clatter than the more common ball
bearings. A word of warning, however: Sleeve bearings are less robust than ball
bearings, and are not suitable for use in high heat situations (such as on a
heatsink). They are also prone to damage if operated in a horizontal orientation.

That’s not to say it’s without its bad points. Its low speed makes it inappropriate
for small heatsinks and hot cases. Even at full speed, it just doesn’t blow
that much air. However, from a noise perspective, the best solution to these
problems is to get a better heatsink or build a cooler system.

It also generates a surprising amount of vibration, which may end up producing
noise on its own when hard mounted to a flimsy case. Soft-mounting with silicone
grommets or some other soft material is recommended to get the most out of this
fan. It’s also a sleeve bearing design, which should only be used vertically, and it may be more at risk for rapid loss of lubricant than with other bearing types. (For more information on sleeve bearings, see Anatomy of the Silent Fan.)

We have both closed corner flange and open corner flange samples of this fan. The closed fan version might be earlier samples. While the closed corner flanges may be good for
the stability of the frame, they are a bit of a pain to work with. Many clip
or grommet-based mounting systems to not work well with this arrangement. In
fact, the problem has annoyed so many people that the
instructions for cutting the corners away
has earned a permanent sticky
topic in our fan forum.

Sample variance was fairly minor through the five samples that we tested. One
sample had a slight ticking noise, while another had a bit more chuffing and
hum than the others, but these differences were very minor, and did not change
the noise character enough to provoke more than a comment in passing.

Noise Recordings

PAPST 8412 NGL

Ambient noise at the time of testing was 18 dBA.

Brand ebm-Papst Power Rating 0.6W
Manufacturer ebm-Papst Airflow Rating 33 m³/h (19.4 CFM)
Model Number 8412 NGL RPM Rating 1,500 RPM
Retail Availability Yes Noise Rating 12 dBA / 3.5 Bel
Bearing Type Sintec Sleeve Header Type Bare wire
Hub Size 1.37″ Starting Voltage 5.0V
Frame Size 80 x 80 x 25 mm Number of Samples 2
Voltage
Noise
RPM
CFM
Power
12V
21 dBA@1m
1520 RPM
23 CFM
0.63W
9V
~19 dBA@1m
1020 RPM
14 CFM
0.50W
7V
<18 dBA@1m
670 RPM
7 CFM
0.41W
5V
<18 dBA@1m
320 RPM
0.31W
@10 CFM (7.7V)
<18 dBA@1m
790 RPM
10 CFM
0.43W


* Much thanks to EndPCNoise for supplying these samples. *
Ebm-Papst has an enviable reputation for quality that sets them apart from
their competitors. They are unusual in the tech industry in that they do not
have any direct ties to China or Taiwan. The company is based in Germany, and
manufacturing and distribution are both centered in Europe. Their fans do not
seem to be all that common on the retail market, especially in North America.

Because ebm-Papst is an OEM, most of their products are bought by electronics
companies who use them in their products — retailers who carry these fans
have to go out of their way to find them. Hard to get or not, Papst fans are
often recommended as quiet fans, especially by readers from the EU. They have a reputation for being robust and long-lived, and are quite costly.

Our two samples are quite old — they are branded without the “ebm”
prefix, indicating that they were manufactured before the company adopted its
current name in 2003. In fact, the exact model number on our samples no longer
appears on Papst’s web site. The specifications listed above are from the closest
match we could find — the 8412 NGLE.

Unfortunately, this particular sample did not live up to the Papst low-noise reputation, despite the Sintec modified sleeve bearing.
Loud isn’t quite the word to describe it — its low rotation speed ensured
that it measured almost as low as the Nexus — but it’s a perfect example
of a fan that measures quiet but sounds noisy. The reason: The noise character
is impossible to ignore. The fan had all manner of buzzing, clicking and chuffing
that didn’t really go away at any speed. The most intrusive was the repetitive
clicking that varied with the speed of the fan. The fan was audible even at
5V — a point where the fan didn’t blow enough air to measure.

The fan achieved 10 CFM at around 7.7V, where it was clearly audible from one
meter. Audible or not, it wasn’t quite loud enough to be picked up by the sound
meter, and it’s possible that other sources of system noise (such as the hard
drive) would muffle the noise character enough to live with it. Bottom line:
If you believe that Papst’s reputation for quality is worth paying for, it’s
not impossible to use this 8412 NGL in a quiet system, but there are definitely
quieter choices out there. We’ll have to get our hands on some other Papst samples in the future.

Noise Recordings

JUMP TO:

CHENG HOME ELECTRONICS SUPER RED CHA8012AS-A

Ambient noise at the time of testing was 18 dBA.

Brand Cheng
Home Electronics
Power Rating 0.09A
Manufacturer Cheng Home Electronics Airflow Rating 25.07 CFM
Model Number Super Red CHA8012AS-A RPM Rating 2,100 RPM
Retail Availability No Noise Rating 23 dB
Bearing Type Sleeve Header Type 3-pin
Hub Size 1.37″ Starting Voltage 3.7V
Frame Size 80 x 80 x 25 mm Number of Samples 4
Voltage
Noise
RPM
CFM
Power
12V
29 dBA@1m
2210 RPM
32 CFM
1.06W
9V
22 dBA@1m
1700 RPM
25 CFM
0.81W
7V
20 dBA@1m
1310 RPM
19 CFM
0.64W
5V
~18 dBA@1m
860 RPM
12 CFM
0.49W
@10 CFM (4.6V)
<18 dBA@1m
780 RPM
10 CFM
0.38W

These fans are not available on the retail market, but we have included it
because many long-time silencers should be familiar with its noise characteristics:
Seasonic used these fans in some of their earlier power supplies before they
switched to their current supplier.

Cheng Home Electronics is about what you’d expect for a fan supplier to the
electronics industry. They’re based in Taiwan, manufacture in China, and sell
“worldwide”. The most notable aspect of their design is the use of
three support struts instead of the usual four. Presumably this has the effect
of reducing impedance slightly, but it’s hard to imagine that the effect is
large. In any case, we didn’t notice that the Super Red produced substantially
more airflow than any other fans.

The Super Red is faster than either of the two previous fans, so it’s no surprise
that it was noisier, although it’s still classed as a low speed fan by the manufacturer.
More critically, it remained audible even when undervolted to the 10 CFM level,
and that level could not be achieved without dropping the voltage below 5V (the
lowest easily achievable voltage in most systems).

Once again, the culprit was noise character, which is typical of ball bearing designs: The fan produced a severe buzz
whenever it was running, although the whine from the motor eventually replaced
it as the dominant noise at higher (>10V) speeds. The fan also produced a
squealing high frequency overtone that was very irritating. This tone was most
audible at lower speeds.

Noise Recordings

NMB-MAT 3110KL-04W-B10

Ambient noise at the time of testing was 18 dBA.


An oddity: This particular sample ran slower than the others, despite bearing
the same model number.

Brand NMB-MAT Power Rating 0.10A
Manufacturer NMB-MAT Airflow Rating 24.7 CFM
Model Number 3110KL-04W-B10 RPM Rating 2,150 RPM
Retail Availability Yes Noise Rating 22 dB
Bearing Type Ball Header Type 3-pin PSU (2 wire)
Hub Size 1.37″ Starting Voltage 5.4V
Frame Size 80 x 80 x 25 mm Number of Samples 3
Samples 1 & 2 (rated for 0.10A)
Voltage
Noise
RPM
CFM
Power
12V
25 dBA@1m
1930 RPM
29 CFM
0.86W
9V
21 dBA@1m
1420 RPM
21 CFM
0.55W
7V
19 dBA@1m
1050 RPM
14 CFM
0.42W
5V
@10 CFM (5.7V)
<18 dBA@1m
800 RPM
10 CFM
0.32W
Sample 3 (rated for 0.06A)
Voltage
Noise
RPM
CFM
Power
12V
21 dBA@1m
1355 RPM
21 CFM
0.48W
9V
~19 dBA@1m
970 RPM
14 CFM
0.28W
7V
~19 dBA@1m
680 RPM
9 CFM
0.21W
5V
<18 dBA@1m
390 RPM
3 CFM
0.15W
@10 CFM (7.2V)
~19 dBA@1m
770 RPM
10 CFM
0.22W

NMB-MAT is a large OEM manufacturer that is the end result of a three-way merger
between three well known manufacturers: NMB,
Minebea, and
Panasonic’s
Panaflo division
. Because of this, the specifications for the company’s
fans can be found in a dizzying number of places, as all three companies still
maintain independent web sites. Even worse, the specifications published in
different places do not always agree with each other, making it difficult to
know exactly which fan you’re looking at.

Our samples bear NMB’s mark, not the more recent NMB-MAT brand, so it seems
likely that they were manufactured before the merger took place. For now, we
will assume that the specifications on NMB’s web site are the correct ones.

Making matters even more confusing, we ended up with a lower speed sample that,
for all intents and purposes appeared to be identical to the others except for
its electrical rating its default speed, and the presence of an RPM monitor
wire. According to NMB’s
model numbering system
, the only difference between the fans should have
been the RPM monitor wire — the differences in electrical rating and speed
are not accounted for in the model number. Because the model numbers are identical
in every important respect, we have included the lower speed fan with the rest
of the samples and listed the test results for this “unofficial” model
side by side with the results for the two other samples.

Oddly labeling nonwithstanding, all three of our samples sounded very similar
when run at similar speeds. We were unable to identify the slower model on the
basis of noise quality alone. All suffered from a persistent buzz that never
quite went away, even at the quietest levels. At certain speeds, the noise sounded
like wet high voltage power lines — a singularly irritating quality.

Aside from the noise quality, the fan was reasonably low speed and was therefore
reasonably quiet. The maximum measured noise level of 25 dBA@1m is not loud,
especially given the amount of airflow it generated at that level. In this respect
it is like the Papst that we looked at above: Slow enough to be quiet, but with
a poor noise character that ruins it for use in very low speed situations.

One word of warning: One of our samples had a very high starting voltage: It
would not start consistently below 5.4V, which is above the minimum voltage
given by several popular fan controllers such as Zalman’s Fanmate. NMB rates
the minimum input voltage at 6.0V, so we were technically operating the fan
out of spec, but we were surprised nevertheless — most fans of this speed
rating have no such issues.

Noise Recordings

JUMP TO:

SILENX IXTREMA THERMISTOR FAN IX-08025-14T

Ambient noise at the time of testing was 18 dBA.

Brand SilenX Power Rating 0.07~0.09A
Manufacturer Globe
Fan
Airflow Rating 18~28 CFM
Model Number IX-08025-14T RPM Rating 1,400~2,200 RPM
Retail Availability Yes Noise Rating 11.8~14.4 dBA
Bearing Type Hypro? Header Type 3-pin
Hub Size 1.29″ Starting Voltage 3.5V
Frame Size 80 x 80 x 25 mm Number of Samples 3
Voltage
Noise
RPM
CFM
Power
12V
33 dBA@1m
2850 RPM
42 CFM
2.52W
9V
28 dBA@1m
2310 RPM
33 CFM
1.82W
7V
22 dBA@1m
1870 RPM
26 CFM
1.40W
5V
~19 dBA@1m
1330 RPM
17 CFM
1.01W
@10 CFM (3.6V)
<18 dBA@1m
750 RPM
10 CFM
0.53W


* Much thanks to Anitec for supplying these samples
*
SilenX has a reputation for selling fans with absurdly low noise ratings, and
the 11.8 dBA rating for this one is no exception — it’s too low to be plausible.
Fortunately, SilenX also has a reputation for being fairly quiet.

Evaluating this fan is somewhat difficult because it’s thermally controlled,
meaning it rarely, if ever runs at full speed, but it expects full voltage nonetheless.
For the purposes of this test, we short-circuited the thermistor temporarily,
causing the fan to run without the thermistor slowing things down. Assuming
we can take SilenX’ RPM specifications at face value, the fan’s actual operational
range should be the equivalent of what we heard between 5~9V. Knowing the way
most thermally controlled fans work, there’s a good chance that the fan will
spend most of its time at minimum speed, which bodes well for noise.

SilenX sells to the retail market, and this fan looks the part. It comes in
a color cardboard package that includes screws, rubber grommets for soft-mounting,
and a Molex to 3-pin adapter for flexibility. All in all, not a bad package.

A tiny globe logo on the hub marks the fan as being sourced from Globe Fan,
most likely. However, the mention of hypro bearings — supposedly unique
to Adda — makes us wonder. Have we mis-identified the logo (unlikely, as
we’ve encountered the logo on Globe’s own fans), do Adda and Globe fan have
a business relationship, or is SilenX playing fast and loose with the terms
in their marketing material?

The fan spins quickly and noisily at full speed. No fan we’ve encountered is
quiet at 3,000 RPM, and the Ixtrema held true to this rule. That said, it was
quiet enough at 5V, and it still pushed a decent amount of air. We had to drop
the input voltage down to 3.6V (just 0.1V above the starting voltage) to achieve
10 CFM for our constant airflow test.

Noise quality was nothing too impressive. It didn’t disappear into the background
like the Nexus, but the papery throbbing at 10 CFM was easy to ignore as background
hiss. Above 5V, motor whine quickly became a problem, but it was still acceptable
at its default speed around 5V.

Noise Recordings

MECHATRONICS A8025S12D

Ambient noise at the time of testing was 18 dBA.

Brand Mechatronics Power Rating 0.075A
Manufacturer Mechatronics Airflow Rating 25 CFM
Model Number A8025S12D RPM Rating 1,500 RPM
Retail Availability Minimal Noise Rating
Bearing Type Ball Header Type bare wire
Hub Size 1.33″ Starting Voltage 5.3V
Frame Size 80 x 80 x 25 mm Number of Samples 1
Voltage
Noise
RPM
CFM
Power
12V
21 dBA@1m
1570 RPM
19 CFM
0.76W
9V
~19 dBA@1m
1200 RPM
14 CFM
0.57W
7V
<18 dBA@1m
900 RPM
9 CFM
0.46W
5V
@10 CFM (7.2V)
<18 dBA@1m
930 RPM
10 CFM
0.46W

Mechatronics made its name making bearings for the Aerospace industry, so they
are not well known in the electronics industry. They are headquartered not far
from SPCR in a suburb of Seattle, but the “made in Korea” on the fan
hub shows that they follow the common industry practise of out-sourcing their
manufacturing (and design?) to Asia.

This particular fan has unusually small fins and a very low power rating. It
has been on our recommended list for some time, so we already know that it’s
quiet. Can it retain its recommended status, or will our newly stringent
test techniques reveal some flaw that was overlooked before?

The most obvious place to look for flaws is the airflow. Those small, backwards-swept
blades don’t push much air, and we weren’t surprised to learn that the motor
had to spin about 20% faster to generate the same airflow as the more conventional
fans that we tested. Despite the extra speed, the fan stayed quiet enough to
give the Nexus a run for its money.

We had to strain to hear the fan at 9V, and it disappeared almost entirely
below this level. Straining our ears, we might catch the occasional click that
could be attributed to the fan, but these would not be heard in general use.

At full speed, it produced slightly less airflow and measured slightly louder
than the Nexus. The higher noise measurement can be attributed mostly to differences
in noise character. The Nexus sounded slightly smoother and more tonal, while
the Mechatronics had a muffled, rhythmic throbbing that made for a less constant
sound.

The fan’s starting voltage was quite high — above the 5V threshold that
we consider failsafe. However, in this case, the noise quality is good enough
that a smart builder might be able to use this fact to his advantage. In combination
with a thermally regulated controller, the Mechatronics is a good candidate
for use as an emergency activated fan that only turns on when the case begins
to overheat. Just bear in mind that the stall voltage is considerably less
than the starting voltage, so once it comes on it will probably be on for good.

Noise Recordings

JUMP TO:



DELTA AFB0812SH

Ambient noise at the time of testing was 18 dBA.

Brand Delta
Electronics, Inc.
Power Rating 0.34A / 4.08W
Manufacturer Delta Electronics, Inc. Airflow Rating 46.62 CFM
Model Number AFB0812SH RPM Rating 4,000 RPM
Retail Availability Yes Noise Rating 40 dBA
Bearing Type Ball Header Type 3-pin
Hub Size 1.47″ Starting Voltage 3.0V
Frame Size 80 x 80 x 25 mm Number of Samples 1
Voltage
Noise
RPM
CFM
Power
12V
44 dBA@1m
3790 RPM
59 CFM
3.45W
9V
35 dBA@1m
2910 RPM
45 CFM
2.23W
7V
28 dBA@1m
2230 RPM
34 CFM
1.54W
5V
24 dBA@1m
1440 RPM
22 CFM
1.02W
@10 CFM (3.4V)
20 dBA@1m
760 RPM
10 CFM
0.67W

We said we would include one really noisy fan in the round up, and this is
it. Delta fans have long had a reputation for being screamers, and this model
certainly justifies it. However, given the large number of fans that Delta sells,
it’s not inconceivable that they do have some quieter models. It’s hard to expect
any fan to be quiet when it’s whirring along at 4,000 RPM. This fan has
all the hallmarks of a noisy, high speed fan. Foremost among them: A high electrical
rating (the label reads 0.51A, but Delta’s most recent spec sheet claims less)
and a chunky center hub with a ball bearing that can stand up to the torque of high speed operation.

Strangely, while our measurements showed the fan spinning slightly below its
rated speed, noise and especially airflow were higher than Delta’s specifications.
We measured the fan’s rated airflow at just above 9V. Perhaps Delta’s specifications
take into account the fact that fans are not typically used in open air?

One thing that surprised us was how much power was required to drive the fan
at full speed. Most of the other (slower) fans that we looked at required less
than a full watt at full speed, but the Delta ate up almost three and a half
watts. Some of this can be attributed to the Delta’s higher top speed, but consider
this: The two thirds of a watt required to drive the Delta at 3.4V was higher
than the 0.57W required to drive the Nexus at full speed and twice the RPM.

The noise from the fan was… noisy. At lower speeds, the Delta clearly lost
out in terms of airflow-for-noise. At 5V, it blew as much air as a full speed
Nexus but was 4 dBA@1m louder. In the constant airflow test, the Delta was the
only fan we tested that we were able to measure above the background noise.
The noise at this level was chiefly the sharp clicking that seems prevalent
in so many ball bearing fans.

Noise Recordings

SCYTHE KAMA FLOW SA0825FDB12SL

Ambient noise at the time of testing was 19 dBA.

Brand Scythe
Co., Ltd.
Power Rating 0.09A
Manufacturer Sony? Airflow Rating 19 CFM
Model Number SA0825FDB12SL RPM Rating 1,500±10% RPM
Retail Availability Japan Only Noise Rating 14 dBA
Bearing Type FDB Header Type 3-pin w/ Molex adapter
Hub Size 1.41″ Starting Voltage 4.6V
Frame Size 80 x 80 x 25 mm Number of Samples 2
Voltage
Noise
RPM
CFM
Power
12V
22~23 dBA@1m
1570 RPM
24 CFM
0.48W
9V
~20 dBA@1m
1150 RPM
17 CFM
0.36W
7V
<19 dBA@1m
850 RPM
11 CFM
0.28W
5V
<19 dBA@1m
530 RPM
5 CFM
0.22W
@10 CFM (6.1V)
<19 dBA@1m
720 RPM
10 CFM
0.25W
May 5, 2008
The updated airflow results here are the result of improvements
in our testing procedures. They are more accurate than the original results
above, but they are not directly comparable. Please compare these only with
fan reviews published after May 5, 2008 — or ones that have updated
results published in a box like this one.
12V
22~23 dBA@1m
1570 RPM
17 CFM
0.48W
9V
~20 dBA@1m
1150 RPM
12 CFM
0.36W
7V
<19 dBA@1m
850 RPM
8 CFM
0.28W
5V
<19 dBA@1m
530 RPM
5 CFM
0.22W

We subjected your ears to a horrible noise in the Delta AFB8012SH, and we’re
sorry. We hope that this fan will make up for it.

Anyone familiar with Scythe’s line of heatsinks knows that Scythe likes to
juggle about ten current models at a time. They pull a similar act with fans,
with at least four distinct lines. How do they manage to differentiate the various
fans? A very good question — one that we’re not entirely sure how to answer.
Nevertheless, they do carry multiple fan models and this is one of them.

At the time of writing, this particular fan is only available in Japan, under
a name that transliterates to “Furou” (???) or, less literally, “Flow”.
Three different versions are available: Standard, quiet, and super-quiet. Our
samples were the super-quiet variety, and we were pleased to note that they
did, in fact, seem to be quite quiet. Perhaps these are the retail equivalent
of the stock fan that Scythe has shipped with their recent heatsinks?

Aside from being reasonably low noise, this fan’s biggest distinguishing characteristic
appears to be the use of Fluid Dynamic Bearings, similar to those used in modern hard
drives. Scythe’s S-Flex line also uses these bearings, and it is possible that
our fan is a predecessor to the S-Flex line. FDB is an evolution of the sleeve bearing that provides the lowest noise (according to the manufacturers) and much prolonged life even under high heat conditions.
(See the discussion of FDB on page 2 of Anatomy of the Silent Fan.)

This fan is another one to rival the Nexus for noise. It blew a little more
air at full speed and spun a little faster to produce it, but the noise character
was almost the same: A smooth, low hum that only showed up at around 9V. From
our usual listening distance of one meter, we were unable to hear it at 7V and
below. So, here it is at last: A worthy competitor to the Nexus that we can
wholeheartedly recommend. It even has the FDB advantage of being usable in a variety of positions, not just vertical. Now, if only someone would bring us back a few dozen
from Japan…

Noise Recordings

JUMP TO:

SUNON KS1208PTS3

Ambient noise at the time of testing was 18 dBA.

Brand Sunon Power Rating 1.4W
Manufacturer Sunon Airflow Rating 31 CFM
Model Number KS1208PTS3 RPM Rating 2,300 RPM
Retail Availability Yes Noise Rating 26 dBA
Bearing Type Sleeve Header Type bare wires
Hub Size 1.48″ Starting Voltage 3.0V
Frame Size 80 x 80 x 25 mm Number of Samples 2
Voltage
Noise
RPM
CFM
Power
12V
32 dBA@1m
2400 RPM
36 CFM
1.35W
9V
30 dBA@1m
1920 RPM
28 CFM
1.08W
7V
23 dBA@1m
1550 RPM
23 CFM
0.90W
5V
21 dBA@1m
1130 RPM
16 CFM
0.70W
@10 CFM (3.2V)
~19 dBA@1m
710 RPM
10 CFM
0.45W

Like Delta, Sunon is a big name. Walk into an electronics store that buys a
single brand of fans in bulk, and chances are decent that that brand will be
Sunon. And, like Delta, Sunon does not seem to put much effort into making their
product line quiet. To wit: This fan is the slowest and quietest in its series,
and it’s rated for 2,300 RPM and 26 dBA. Our favorite fans are rated for 1,500
RPM and below, and, because they are marketed as quiet, often carry unrealistically
low noise ratings.

We’re a little unsure of the results for this fan because the two samples that
we had were quite obviously different, despite bearing the same model number.
The difference was in the frame: Both frames were unusually sturdy, but one
had closed flanges and seemed even more sturdily built. A quick listen showed
that the noise didn’t vary significantly, but we have to wonder how many other
variations of this model number there are…

Subjectively, the Sunon didn’t sound too bad. It was a touch buzzier than we’d
usually like, but it didn’t have the clicking or rattling that we heard in some
other fans. Unfortunately, undervolting was not very effective; dropping the
voltage to 5V reduced the noise to a level that would just barely be acceptable
in the quietest systems. That’s not to say it didn’t end up being quiet when
we dropped the speed enough to produce 10 CFM, but the 3.2V required to achieve
this is difficult produce for most users. It’s lucky that the fan even started
consistently at this level — at 3.0V, this fan had the lowest starting
voltage of any fans we tested.

The Sunon also had issues with vibration — it had a lot. The fan harness
shook visibly during the test, especially at higher speeds. Soft-mounting could
definitely benefit this fan, as placing the bare fan on our test bench produced
a hum that was not audible when it was tucked into the foam harness that we
use for most of our testing.

All in all, Sunon’s biggest advantage is its wide availability. In situations
where nothing else is available, this fan at 5V may do in a pinch, but it’s
far from the best of the bunch.

Noise Recordings

ARCTIC COOLING ARCTIC FAN 3

Ambient noise at the time of testing was 18 dBA.


Arctic Cooling has been playing with frameless designs for a long time.

Brand Arctic
Cooling
Power Rating 0.12A
Manufacturer Arctic Cooling Airflow Rating 28 CFM
Model Number Arctic Fan 3 RPM Rating 1,900 RPM
Retail Availability Discontinued Noise Rating 0.8 Sone
Bearing Type Fluid Dynamic Bearings Header Type 3-pin
Hub Size 1.39″ Starting Voltage 3.4V
Frame Size 80 x 80 x 45 mm Number of Samples 2
Voltage
Noise
RPM
CFM
Power
12V
21 dBA@1m
1800 RPM
12 CFM
1.18W
9V
19 dBA@1m
1460 RPM
10 CFM
0.87W
7V
<18 dBA@1m
1170 RPM
6 CFM
0.66W
5V
<18 dBA@1m
810 RPM
3 CFM
0.47W
@10 CFM (9.0V)
<18 dBA@1m
1460 RPM
10 CFM
0.87W

Arctic Cooling has been making quiet, inexpensive products for a long time,
and they’ve become quite good at it. They’ve been experimenting with “frameless”
fans for some time with the goal of reducing turbulence noise. Unfortunately,
as the results above show, the approach also seems to affect airflow significantly.
That said, it’s not clear how accurate our airflow measurements are for this
fan; measurement required fitting the fan into our test harness, which imposed
a foam “frame” around the fan that does not reflect standard usage.

The Arctic Fan 3 is a slightly smaller stand-alone version of the fan found
on Arctic Cooling’s last round of low-end heatsinks, the
Super Silent 4 series
. The fan can also be found in Arctic
Cooling’s Silentium T2 case
. It has since been superseded by the Arctic
Fan 8 series, but the design has not changed that much, and there are still
many Arctic Fan 3’s to be found on the retail market.

There’s no question that the Arctic Fan 3 is a quiet fan. It measured 21 dBA@1m
at full speed — the equal of Scythe’s FDB fan and nearly as good as our
favorite Nexus. The trouble is airflow: There’s not enough of it to make voltages
below 9V worthwhile, and, unfortunately, the fan is still audible at this level.
The poor airflow results seriously hurt the fan, as it needs to spin faster
and more noisily to push the same amount of air. In addition, the fan is not
well suited to situations that require higher airflow, since the maximum airflow
we ever measured was just 12 CFM.

The noise character was very, very smooth, with most of the noise scattered
across the higher frequencies. There was very little low frequency noise at
all. At full speed, the fan had just a touch of whine, but it was higher pitched,
and thus more clearly audible than most fans at this noise level. In addition,
the fan seemed to squeal a bit at certain speeds — an odd noise that is
unlike any other fan we’ve heard. The squealing was most evident just before
the fan started, at around 3.3V. The easiest method of dealing with the squeal
is probably to ensure that there is no direct noise path between the fan and
your ears; higher frequencies are easily deflected or absorbed within the case,
and as long as there is something between you and the fan, the squealing should
not be a problem.

Noise Recordings

JUMP TO:

PANAFLO FBA08A12L

Ambient noise at the time of testing was 18 dBA.

Brand Panaflo Power Rating 0.068A / 0.82W
Manufacturer Panasonic / NMB-MAT Airflow Rating 24 CFM
Model Number FBA08A12L RPM Rating 1,900 RPM
Retail Availability Yes Noise Rating 21 dBA
Bearing Type Hydro Wave Header Type bare wires
Hub Size 1.46″ Starting Voltage 4.9V
Frame Size 80 x 80 x 25 mm Number of Samples 18
Voltage
Noise
RPM
CFM
Power
12V
24 dBA@1m
1900 RPM
30 CFM
0.82W
9V
~19 dBA@1m
1400 RPM
21 CFM
0.65W
7V
<18 dBA@1m
1050 RPM
15 CFM
0.55W
5V
<18 dBA@1m
670 RPM
8 CFM
0.37W
@10 CFM (5.2V)
<18 dBA@1m
710 RPM
10 CFM
0.40W

Ah, the infamous Panaflo L1A. We have a soft spot for these fans — for
a long time they served as SPCR’s de facto reference fan thanks to their reasonably
smooth character (in the good samples) and easy availability on the internet.
The development of the “Silent Computing” market sector marked the
end of the Panaflo’s reign — once manufacturers started paying attention
to noise, it became much easier to find quiet fans, but, for us, the Panaflo
was with us at the very beginning. It’s still not a bad choice, especially if
you’re not willing to pay a premium for retail markup.

Panaflo is dead and gone now — the brand was folded into the conglomerate
that is NMB-MAT some time ago — but the longevity of this fan is such that
there are still quite a few floating around on the internet. Panasonic made
a lot of these fans, and most of the ones that individual users
can buy are overstock (or sometimes bad stock) that was originally intended
for the OEM market. Panaflo is something of a legend, and there’s a lot of interesting
and obscure knowledge about it floating around on the net. In fact, Dorothy
Bradbury
is somewhat of an unofficial expert on these fans. This is a company
that sold Panaflo fans exclusively (they now carry only NMB-MAT), and Dorothy
Bradbury herself has left a few gems of knowledge floating around on the SPCR
forums.

Panaflo fans were manufactured in several different factories, and for a long
time the quality control at the Japanese factories seemed to be better than
those in other countries, leading to a very specialized demand for “Japanaflo”
fans by silencers who snubbed the more widely available Chinese-made models.
Perhaps because the fan was so widely known, it eventually developed a reputation
for having a large amount of sample variance (in addition to the “country-of-origin”
variance observed above). In particular, users who bought the fans in bulk began
to notice that the noise quality seemed to vary by box, leading some to theorize
that the variation was caused by rough shipping practices, and was not inherent
to the fans themselves.

This Panaflo fan uses Hydro Wave bearings: A modified sleeve bearing (similar to
FDB bearings) that is not vulnerable to damage when operated in a horizontal
orientation. This is what made the Panaflo so attractive when it was first discovered.
It has the flexibility (and longevity) of a ball bearing fan, but its noise
character is smoother than most ball bearing fans.

There were two popular Panaflo models that SPCR recommended: Low speed and
medium speed versions of the same basic fan. Both fans sounded quite similar,
and we’ve stockpiled a large number of each over the years. We tested both models,
but we will deal with the low speed version first. We tested a total of 18 samples
that we had on hand.

The low speed model was more or less inaudible at 5V and just barely audible
at 7V. The noise character was not quite as smooth as the Nexus or Scythe models
that we looked at, but it wasn’t bad, especially in the better samples. Lesser
samples tended to buzz a bit more, and some developed a chuffing sound at low
speed. On the whole, the noise character remained a deep throaty growl that
“wobbled” a bit, but mostly lacked the sharp clatter that many other
fans produce.

Although classed as a low speed fan, the Panaflo has a higher top speed than
the other quiet fans that we’ve recommended. At 9V, it produced about the same
noise and airflow as the Nexus does at 12V, although the Nexus sounded a bit
smoother. At 12V, a touch of motor whine was clearly audible, but it was still
quiet enough that many may find it worth the sacrifice for the ~50% increase
in airflow between 9V and 12V.

Noise Recordings

PANAFLO FBA08A12M

Ambient noise at the time of testing was 18 dBA.

Brand Panaflo Power Rating 0.124A / 1.49W
Manufacturer Panasonic / NMB-MAT Airflow Rating 32 CFM
Model Number FBA08A12L RPM Rating 2,450 RPM
Retail Availability Yes Noise Rating 28 dBA
Bearing Type Hydro Wave Header Type bare wires
Hub Size 1.46″ Starting Voltage 4.3V
Frame Size 80 x 80 x 25 mm Number of Samples 11
Voltage
Noise
RPM
CFM
Power
12V
31 dBA@1m
2340 RPM
37 CFM
1.28W
9V
23 dBA@1m
1780 RPM
28 CFM
0.97W
7V
~20 dBA@1m
1360 RPM
20 CFM
0.78W
5V
<18 dBA@1m
850 RPM
12 CFM
0.53W
@10 CFM (4.6V)
<18 dBA@1m
750 RPM
10 CFM
0.43W

This is the medium speed version of the low speed Panaflo reviewed above.
It needs no further introduction, so comments will be limited to differences
between the two fans.

As a general rule, the medium speed models seemed to have more issues with
buzzing and chuffing than the low speed models, and it was harder to find a
really good sample among the choices we had. That said, we’d be hard pressed
to tell the difference between the low and medium speed models if they were
running side by side at the same speed. In fact, the sample variance within
each speed category was probably larger than the variation we heard jumping
from slow to medium.

Most users will find that the choice between low and medium speed will probably
be dictated by purpose and availability. Where a large margin of safety is required,
the medium speed model is capable of providing quite a respectable amount of
airflow if required, but it won’t be quiet while doing so. The low speed model
is better suited to situations where a finer grain of control is desired or
with an automatic fan controller that occasionally pushes the fan to full speed.
It reasonably quiet throughout its operating range, while the medium speed version
is best kept to 9V and below.

Another thing to consider is that users who want to supply the fan with 5V
and leave it are better off choosing the medium speed version. Both versions
are effectively inaudible at 5V, but the medium speed provides more airflow
and starts more reliably at this voltage.

Noise Recordings

JUMP TO:

CONCLUSIONS

It should be no surprise to regular readers that the Nexus is
still the quietest fan that we know of. We’ve referred to it so often in our
regular articles that most readers know by now that we think highly of it. What
is surprising is how slim the Nexus’ advantage is, especially when noise-for-airflow
is taken into account. In fact, Scythe’s obscure “Flow”-series fan
is the equal of the Nexus in terms of noise and airflow, and may even be superior
for reliability if its FDB bearings live up to their reputation. However, its
extremely limited distribution outside of Japan will mean that most users will
simply have no option but to continue using Nexus fans. Scythe, are you listening?
There’s an opportunity here…

There are a number of other fans that come very close to standard
set by the Nexus: The unusual fans from Mechatronics and Arctic Cooling and,
of course, the familiar Panaflo fans. In fact, both the Mechatronics and the
Arctic Cooling may be quieter than the Nexus for a given rotation speed; it’s
only once airflow is considered that they drop behind a bit.

All of the fans mentioned above have smooth noise signatures,
and they are completely inaudible from one meter when undervolted to produce 10 CFM.
Any one of them would be quite at home in a quiet system so long as high airflow
is not needed. The advantage of the Nexus and Scythe fans is that they are both
reasonably quiet at 12V while producing a little more than 20 CFM each. This
makes them very flexible: They can be used at a constant speed (~7V is a good
bet) without being heard, or they can safely be used with a fan controller without
worrying that the noise level will ever get too high.

Panaflo, the old standby, also lived up to its reputation so long
as sample variance could be avoided. These fans are not so quiet at 12V, so
they are a bit more difficult to work with, but they are well suited to situations
where the Nexus or Scythe fans don’t provide adequate cooling or a little more
headroom is desired. They can always be turned down if the airflow isn’t needed.

The rest of the fans in the test are less interesting, mostly
because they don’t sound good enough. Let’s face it; most of the fans measured
fairly close to each other, and most could be made quiet simply be turning them
down far enough. Where they differed was noise character, and it was
the fans that sounded bad no matter what speed they were at that fared poorly
in this roundup

Not surprisingly, the screaming Delta was the worst offender here,
but the Sunon was also quite noisy. The biggest surprise was the well-reputed
Papst, which clicked severely and would not be a good choice in a quiet system.
Perhaps we happened across the wrong models or have damaged samples. If anyone out there has a non-clicking
80mm Papst fan that they would like to contribute for analysis, we’d love to
hear from you!

The bottom line is, we are happy to find that our methodical approach
has confirmed what we’ve known all along: It’s tough to find a smoother, quieter
fan than the Nexus. However, we were pleasantly surprised to find a fan that
could equal it in the Scythe. We look forward to testing more Scythe fans in
a future roundup

Many thanks to all the readers, contributers and manufacturers who donated fans
so this project could happen.

*

SPCR Articles of Related Interest:
SPCR’s Fan Testing Methodology
Anatomy of the Silent Fan
SPCR’s Recommended Fans
Simple Fan Controllers from Zalman
Get 5V, 7V, or 12V for your Fans

* * *

Discuss
this article in the SPCR Forums
.

* * *

JUMP TO:

Silent PC Review is reader-supported. When you buy through links on our site, we may earn an affiliate commission. Learn More

Leave a Comment

Your email address will not be published. Required fields are marked *