Viewing page 5 of 6 pages.
Previous 1 2 3 4 5 6 Next
TEST RESULTS
The Fans
The two fans are identical. Each is a somewhat unusual nine-blade 92mm
design. Translucent plastic is used; this is a material we've found in many fans that have strong tonal qualities. We suspect it adds tonal resonances to the overall noise signature. The basic geometry is not great, as the trailing edges of the blades are nearly parallel to the struts. For minimal tonal contribution, they should be as perpendicular as possible. (See Fan Blade Geometry on page 3 of the Anatomy of the Silent Fan for more details.) The other notable aspect of its design is that the frame has been cut away, perhaps to reduce mass and/or to minimize its turbulence effects.
|
Fan Specifications
|
| Brand |
Cooler Master |
Power |
0.31A (3.72W) |
| Model |
A9225-18CB-3BN-L1 or PLA09225S12M-3 |
Airflow |
43.8 (total) |
| Bearing |
sleeve (likely) |
RPM |
1800 |
| Hub Size |
1.54" |
Noise |
19 dBA |
| Frame Size |
92 x 92 x 25 mm |
Header |
3-pin |
| Weight |
80 grams |
Start Voltage |
3.2V |
|
The data in the blue cells is provided by the manufacturer; we measured the data cited in the green cells
|
|
FAN MEASUREMENTS
|
|
Voltage
|
SPL - dBA@1m
|
RPM
|
|
1 fan
|
2 fans
|
|
12V
|
30
|
30
|
1750
|
|
9V
|
22
|
23
|
1340
|
|
7V
|
18
|
19
|
1030
|
|
5V
|
14
|
14
|
720
|
Since there are two identical fans, they were tested individually to collect the data in the tables above. They were not identical, but close; the slightly faster of the two fans was cited for the single fan measurements. (The speed difference was about 30 RPM at 12V and less at the lower voltages.)
During all the audio testing, the fans were left on the heatsink, and the heatsink left mounted on the motherboard. The motherboard was placed on a table in the anechoic chamber. A standard letter paper pad was used under the motherboard to keep it from wobbling during audio testing.
In theory, two fans in push/pull configuration have the benefit of higher pressure than one, which makes it useful for a high impedance load between the two fans. However, the total airflow is limited to the free-air airflow capability of a single fan; they do not push double the air. So with a low impedance load, there is no advantage with using the dual-fan, push-pull configuration. The Hyper N520 isn't 100% push-pull, however, because of the slight offset of the two fans, which represents perhaps 20% of the fan blades' diameter. The other question is whether the impedance represented by the ~2" distance through the 1.72mm gaps between 50 thin fins is a high enough impedance to make the push-pull setup worthwhile.
One fan was disconnected for the single fan measurements. There was no appreciable difference in the noise whether the "push" fan or the "pull" fan was disconnected. You could say, judging from the SPL measurements above, that there's not much of a price to be paid for running two fans compared to just one. But subjectively, it was a different story.
Sound Impressions
Despite the S in the model name, which usually denotes sleeve bearing, this fan did not sound like a typical low-speed sleeve bearing fan. The buzzy quality was very similar to the sound of a bad ball-bearing fan. The two fans together sounded worse than one, despite the absence of change in the measured SPL.
@ 12V: While it could not be described as loud, the overall noise was highly tonal, with more than one or two main tones. There was both a buzzing as well as a clicking aspect to the sound, which made it basically unpleasant.
Furthermore, the two fans did not run at exactly the same speed, even though they were fed from the same source. This is not unusual, as one fan could have been a touch more efficient. The slight difference in speed (and primary frequencies) caused intermodulation to arise. This is sometimes referred to as a "beat" frequency, a third frequency of sound that arise from the interaction between two primary ones, and here it sounded like a kind of slow warble or a quick flutter, depending on the exact difference in speed (which varied as the voltage to the fans was changed - for example, the effect was not the same at 12V as at 9V).
If there is any variability in the voltage fed to the fans, the frequency of the intermodulation goes up and down, making it even more distracting and annoying. There was some variation in fan speed at 12V, and this drift might have been a function of the lab power supply or AC power variations. The effect was much more audible in the live (normal) test room where the cooling performance test was conducted than in the anechoic chamber, where there are no reflections to exacerbate the effect.
With just one fan, the overall noise improved, becoming a bit less tonal and the intermodulation effect disappeared, but the measured SPL did not change. The performance cost was 2°C, which is small. The change in overall noise might not be enough to accept, however. If you didn't like the sound of the two fans at 12V, one fan at 12V might still sound too noisy to you.
@ 9V: Most of the turbulence noise faded, but the buzzing and clicking remained, at a much reduced level. The intermodulation sound was mostly gone. Inside a well-damped case, the overall sound level might be considered quiet. There seems little to be gained by running just one fan at this voltage and below; subjectively, the improvement is not great.
@ 7V: Again, the overall noise dropped further, the buzzy quality remained.
@ 5V: Very subdued in level, but up close, the buzzing and clicking character of the fan remained in evidence.
Cooling Results
|
Cooler Master Hyper N520: Cooling Performance
(data in brackets is with a single fan)
|
|
Fan Voltage
|
SPL dBA@1m
|
Temp °C
|
°C Rise
|
°C/W
|
|
12V
|
30 (30)
|
37 (39)
|
17 (19)
|
0.22 (0.24)
|
|
9V
|
23 (22)
|
39 (42)
|
19 (22)
|
0.24 (0.28)
|
|
7V
|
19 (18)
|
42 (47)
|
22 (27)
|
0.28 (0.35)
|
|
5V
|
14 (14)
|
49 (55)
|
29 (35)
|
0.37 (0.45)
|
Load Temp: CPUBurn for ~10 mins.
°C Rise: Temperature rise above ambient (20°C) at load.
°C/W: based on the amount of heat dissipated by the CPU (78W); lower is better. |
The Hyper N520 was a decent performer at the normal 12V, with a temperature rise of 17°C. It was too noisy for our taste, however, especially considering the unpleasant quality of the sound. With one fan, the subjective quality improved a bit, but probably not enough.
The performance at 9V dropped by only 2°C, while the noise dropped to 23 dBA. This is a much better balance.
At 7V, the performance dropped off by 5°C, which might be too much for an enthusiast system. A °C/W number above 0.25 is probably not acceptable for higher performance rigs that use 65W and higher TDP processors. The cooling performance at 5V is unacceptable.
With the N520, 9V is the highest drive voltage recommended for a quiet PC, and this is still well over 20 dBA. The 19°C rise over ambient is good enough for a processor as hot as ours in a cool room. In the confines of a case, in hotter weather, you might easily add 20°C to the CPU temperature, and while that's still under the maximum recommended of 65~70°C, it's probably as high as you'd want to go. Runing this heatsink with just one of the two fans was not a good option, cooling-wise, except at >9V. Below that, it really can't keep the CPU cool enough.
COMPARISONS
When judged on noise vs. cooling, it fares reasonably well against other stock heatsink/fan combinations SPCR has tested. Here's a quick comparison against a few heatsinks with stock fans at around 23 dBA@1m.
|
Comparison: With Stock Fans at 12V
|
|
Heatsink
|
°C rise
|
SPL
|
|
Zalman CNPS9300 AT
|
15
|
24 dBA
|
|
Cooler Master Hyper N520
|
19
|
23 dBA
|
|
Xigmatek HDT-SD964
|
20
|
21 dBA
|
|
Thermaltake MaxOrb
|
21
|
24 dBA
|
The slightly smaller Xigmatek HDT-SD964 is the closest competitor, and if the SPL levels were equalized, the Hyper N520 would probably lose by a nose... or degree or dB. It also loses in price, the Xigmatek selling for only around $30, complete with fan.
How close are any of the above to the best reference quality silent coolers? Not very. Almost any one of the top 120mm fan coolers stomps over them all. Here are some of the best:
|
°C rise of Top 120mm fan CPU Coolers
|
|
Heatsink
|
Nexus 120 fan voltage / SPL @1m
|
|
12V
|
9V
|
7V
|
5V
|
|
16 dBA
|
13 dBA
|
12 dBA
|
11 dBA
|
|
Scythe Ninja 2
|
17
|
18
|
20
|
23
|
|
Xigmatek HDT-S1283
|
13
|
15
|
18
|
22
|
|
Thermalright U120E
|
12
|
14
|
17
|
24
|
|
Noctua NH-U12P
|
14
|
16
|
17
|
21
|
|
Thermalright HR-01+
|
13
|
15
|
16
|
20
|
|
Prolimatech Megahalems
|
10
|
14
|
17
|
20
|
|
All results generated with our reference Nexus 120mm
fan.
|
Why compare the N520 against the best? Because the price difference between them and the $45~50 Cooler Master Hyper N520 is very small. The Ninja 2, for example, sells for $40~50 with its own very quiet fan. The Xigmatek HDT-S1283 runs just $35~40, again, with a fan.
| Help support this site, buy the Cooler Master Hyper N520 dual 92mm fan cooler from one of our affiliate retailers! |
|
