Viewing page 6 of 7 pages.
Previous 1 2 3 4 5 6 7 Next
8. FAN, FAN CONTROLLER and NOISE
We attempted to monitor the fan voltage for the 2-wire Muhua fan,
but found that the voltage remained railed to the 12V line throughout the test,
in spite of audible cues that the fan was spinning faster. This implies PWM
control, or some kind of ground level control, so instead we relied solely on
the more relevant SPL readings and our hearing to monitor the fan performance.
The slowly spinning Muhua fan did not register on our measurement
system, the SPL remaining unchanged at the ambient of 11 dBA@1m from idle all
the way up to 150W. It was barely audible from even 30cm in our anechoic test
chamber. The fan finally started speeding up around 200W, a fractional increase
rounded up to 12dBA@1m, with no real perceivable difference in the sound.
The first perceivable difference was heard at 250W (31ºC intake), where
the fan finally sped up to a slightly more audible, calm "wooshing"
sound at 14 dBA@1m, more from air turbulence than the fan's bearing.
The fan's noise increased most noticeably at 300W (32ºC intake),
to a more audible 22 dBA@1m. To take a reading in the anechoic chamber, the cooling fans in the loading apparatus must be turned off to ensure they do not contribute
to the noise level. This decreases the amount of heat pushed into the "hot
box" test fixture, decreasing the thermal stress on the power supply under
testing. Very soon after the loading apparatus' cooling fans were
turned off, the NX-5000's fan would slow down, implying that a 32ºC intake
is perhaps a threshold for the NX-5000's fan controller. Above 32ºC/300W, the fan
noise increased steadily, up to a maximum of 25 dBA@1m at full load. Even this
noise level is just a quiet, smooth rustle of moving air, which is very impressive
for a 530W PSU at full load.
The screen captures below from the SpectraPLUS audio spectrum
analyzer should be self-explanatory. Note that the noise floor of the anechoic
chamber is just under 11 dBA. The red line is the ambient level of the chamber
without any noise sources.
Idle to about 200W. Peaks are seen at 400Hz and 200Hz, and are related
to commutator buzz and what little turbulence is audible at this level.
At 250W, the fan becomes slightly more audible at 14dBA@1m, with smooth,
low tones dominant.
At 300W, a very noticeable jump in fan noise is seen, with a broad spectrum
that translates to something more like a fuzzy white noise.
As the noise increases past 400W, we mostly hear increasing volume on
the lower pitched hum from blade turbulence and the bearing.
|
Outside the Hot Box
In a modern high performance case that has independent cooling airflow
for the power supply, thermal conditions are much milder than in a conventional
case or in our test box. The PSU does not have to work nearly as hard to exhaust
hot air from the downstream components. As in our most recent PSU reviews,
a noise/load test was performed out of the hot box to simulate noise performance
in a modern, high performance case.
|
Nexus NX-5000 SPL: In Hot Box vs. Out
|
|
Power load
|
200W
|
250W
|
300W
|
400W
|
530W
|
|
in hot box
|
12
|
14
|
22
|
24
|
25
|
|
out
|
12
|
12
|
12.5
|
14
|
19
|
|
Measurements are SPL in dBA@1m
|
On the test bench with ambient room temperature at 23°C, the Muhua
fan managed to stay under a barely audible 14 dBA@1m all the way up to 400W,
and peaked at 19 dBA@1m, which is noticeably quieter compared to its already
superbly quiet performance in the test box.
COMPARISONS
The comparison table below shows the SPL versus Power Load data on all the PSUs tested in the anechoic chamber thus far.
|
Comparison: Various PSUs Noise Vs. Power Output
in Anechoic Chamber
|
|
Model
|
90W
|
150W
|
200W
|
250W
|
300W
|
400W
|
500W
|
6-700W
|
850W
|
|
Seasonic X-650
|
<10
|
11
|
12
|
14
|
16
|
31
|
31
|
32
|
n/a
|
|
Nexus Value 430
|
11
|
11
|
16
|
18
|
18
|
19
|
n/a
|
n/a
|
n/a
|
|
Nexus NX-5000
|
11
|
11
|
12
|
14
|
22
|
24
|
25
|
n/a
|
n/a
|
|
Antec CP-850
|
12
|
12
|
12
|
14
|
14
|
26
|
40
|
44
|
45
|
|
Enermax Eco80+ 500W
|
<11
|
12
|
16
|
19
|
26
|
32
|
33
|
n/a
|
n/a
|
|
Seasonic M12D 850W
|
14
|
14
|
14
|
14
|
14
|
24
|
37
|
42
|
42
|
|
Enermax Modu82+ 625*
|
13
|
13
|
14
|
15
|
16
|
26
|
36
|
37
|
n/a
|
|
Coolermaster M700W
|
14
|
14
|
18
|
21
|
25
|
27
|
34
|
34
|
n/a
|
|
Chill Innovation CP-700M
|
15
|
15
|
15
|
15
|
17
|
30
|
34
|
34
|
n/a
|
|
Antec Signature 650
|
15
|
15
|
15
|
18
|
18
|
28
|
36
|
47
|
n/a
|
|
SilverStone DA700
|
18
|
18
|
18
|
18
|
23
|
32
|
35
|
41
|
n/a
|
|
Nexus RX-8500
|
14
|
14
|
17
|
22
|
28
|
32
|
32
|
33
|
33
|
|
NesteQ ECS7001
|
22
|
22
|
22
|
21
|
23
|
25
|
36
|
37
|
n/a
|
|
PCPC Silencer 610
|
20
|
24
|
24
|
24
|
24
|
30
|
40
|
50
|
n/a
|
|
Corsair CX400W
|
22
|
26
|
32
|
35
|
35
|
35
|
35
|
n/a
|
n/a
|
The green boxes represent >30 dBA@1m SPL.
The Nexus NX-5000 is one of the quietest power supplies reviewed in the anechoic
chamber, and though it is louder than the Nexus Value 430 at 300~400W, it's
practically inaudible below that power load, and still has reserve to spare
for dual-GPU systems, 8 hard disks, or whatever other components you feel it
must power.
Caution: Please keep in mind that the data in the above table is specific to the conditions of our test setup. Change the cooling configuration, the ambient temperature and any number of other factors, and you could change the point at which the fans start speeding up, as well as the rate of the rise in speed. The baseline SPL is accurate, however, probably to within 1 dBA.
| Help support this site, buy from one of our affiliate retailers! |
|
