In addition to basic fan tests in our Massive Fan Shootout, we've thoroughly analyzed noise characteristics of the contenders and determined noise pressure levels in 1/3 octave bands. Our experiment is compliant with ISO 3744. We use a measuring face in the form of a parallelepiped (five measurement points), 0.25 m measuring distance (for better detection of individual peculiarities of noise generated by test objects). The background noise level in the process of the experiment is constant - 15 dBA.
Let's proceed to our test results and comments! 80x80x25 mm Fans.
Arctic Cooling Arctic Fan 8
The spectrum is not quite illustrative, but we can still see some peculiarities of noise generated by Arctic Fan 8. There is a noticeable surge in low frequencies (200 Hz and 250 Hz bands) — that's the so-called characteristic frequency, which is determined by the formula f=(RPM*n)/60, where RPM is the rotational speed, n is the number of blades, physically being a fundamental component of the aerodynamic noise. The second "hump" in the spectrum in 400 Hz and 500 Hz bands can also be interpreted as the second harmonic in the characteristic frequency. Another surge in the 1250 Hz band is not easy to explain. Here is our hypothesis — it's manifestation of engine popping (electromechanical component of noise, which is generated by non-smoothed-out switching between inductor coils.
Arctic Cooling Arctic Fan 8L
This situation is much better than with Arctic Fan 8. The main "acoustic energy" is concentrated in the low frequency zone — characteristic frequency (160 Hz and 200 Hz). Then goes a neat spectrum. As a result, noise generated by Arctic Fan 8L is subjectively perceived as a superposition of well-muffled low-frequency sound and soft rustling of air.
Cooler Master Tri-Blade TBF-B81-E1
Quite an illustrative spectrum. The main peculiarity is a distinct peak of the characteristic frequency at 125 Hz (higher than the "neighboring" levels by 16-17 dB!) Subjectively, this tone in the noise sounds like a cardinally muffled propeller jet (quiet "din"). There is also the second distinct harmonic in the major tone at 250 Hz and the third harmonic spread out at 400 Hz, 500 Hz, and 630 Hz (there is evidently some turbulent noise generated by impeller/stator. Another peak (1250 Hz) indicates engine popping, which you can hear in Tri-Blade TBF-B81-E1.
Cooler Master SAF-B83-E1-GP
We again hear that aerodynamic noise, "generated" by the characteristic frequency, prevails — a distinct peak at 250 Hz, and the second and third harmonics at 500 Hz and 1 kHz. There is also some rise in the middle of the spectrum (1-2 kHz), which probably has to do with noise effects at stator's carriers spiced up with "engine popping".
Cooler Master SAF-S84-E1-GP
Aerodynamic noise of the low-speed SAF-S84-E1-GP impeller is expectedly even — harmonics of the characteristic frequency are not prominent (there are some insignificant surges at 160 Hz and 400 Hz). At the same time, we can already notice some electric traces of engine popping against the muffled aerodynamic background (the first harmonic is at 1 kHz and 1.25 kHz, the second harmonic — at 2.5 kHz).
GlacialTech SilentBlade GT80252BDL-1
This spectrum is very interesting and even unique. Its main peculiarity is spread-out aerodynamic noise with three harmonics — the first harmonic covers 200 Hz, 250 Hz, and 315 Hz, the second harmonic is audible at 400 Hz, 500 Hz, and 630 Hz, the third — 1-2.5 kHz. The reason is the tight sweep and lean impeller with a high solidity factor (1.67 at the hub and 1.12 at the tip) that levels tonality of the characteristic frequency. As a result, equivalent noise of SilentBlade GT80252BDL-1 is uniform, without irritating harmonics. Mid-frequency air rustling subjectively prevails here.
GlacialTech SilentBlade II GT8025-HDLA1
Acoustic influence of the fundamental frequency is distinct here — the main surge is at 160 Hz, the second harmonic appears at 315 Hz. But the middle and top of the spectrum look much neater than in SilentBlade GT80252BDL-1. The resulting noise of SilentBlade II GT8025-HDLA1 is subjectively more ergonomic and uniform.
This is another illustrative spectrum. We can see the distinct tone of the characteristic frequency at 400 Hz (higher than the "neighbors" by more than 10 dB), a very high middle part and high "tops" — that is clear acoustic signs of a high-speed impeller with straight blades. Subjectively, the impeller generates unpleasant "howling" accompanied by loud air hissing.
It's a similar situation in many respects — there is a characteristic tone again (now at 315 Hz) and higher mids and tops of the spectrum. Howling of the impeller is subjectively less loud, but air hissing is still irritating.
The situation expectedly goes back to norm (more or less) at moderate speed — characteristic tonality remains in the TFD-8025L12S spectrum, but it's less pronounced and splits into two harmonics (at 250 Hz and 500 Hz). But the lower middle part of the spectrum reveals "popping of the engine" (1.25 kHz), which was hidden in TFD-8025M12C and TFD-8025H12B behind the high equivalent noise level.
Zalman ZM-F1 Normal Mode
It's an interesting spectrum. Its main distinctive feature is a sub-harmonic of the characteristic frequency — along with the fundamental peak (at 315 Hz), there is a surge at 160 Hz (that is approximately ? the characteristic frequency). The nature of this sub-harmonic is not quite clear. To all appearances, it may be the result of some individual traits of the aerodynamic configuration of ZM-F1 impeller (or structural noises of the fan). This combination of a low mid section and neat tops is very interesting — there is a significant difference from Titan TFD-8025H12B operating at similar speed! So if we don't take into account the sub-harmonic problem, the noise reduction tendency is evident here.
Zalman ZM-F1 Silent Mode
There are no surprises here — we can see a weak fundamental component (200 Hz), a nearly unnoticeable harmonic at 400 Hz, quite a uniform mid section, and a neat high section. As a result, the noise level generated by ZM-F1 in silent mode is quite ergonomic and not irritating.
Vitali Crinitsin (email@example.com)
June 8, 2007
Write a comment below. No registration needed!