Site icon HWCooling.net

When color matters: Corsair AF120 Elite weaker in white

Ľubomír Samák

Results: Vibration, in total (3D vector length)

It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

In our tests of the Corsair AF120 Elite fan, we had its black variant (CO-9050140-WW). However, there is also a white version (CO-9050142-WW) with which the results of the black fan do not scale. This is because of its lower rigidity. Although the shape of the rotor does not differ in any way, the injection mold is the same, but the white material does not have the same chemical elements.

The white material, in short, is composed of different elements than the black one, and the mechanical properties are somewhat different. Even Noctua, when launching the NF-A12x25 chromax.black fans, disclosed that the long wait for the black fans was due to the fact that it was not exactly easy to change the color (from the original brown) in a way that preserved the original specifications. In Corsair, they didn’t worry about it so much, but we did wonder about this issue and measured the difference. It’s nothing dramatic, but it’s good to know that the white fan is pulling the short end of the stick compared to the black one.

However, the stated parameters of maximum airflow and static pressure are the same and correspond to what is claimed by the manufacturer. The difference is in the maximum noise level, which Corsair shouldn’t list the same anymore – the white AF120 Elite fan is noisier. When aligned by fixed noise levels, it then comes out worse in terms of performance as well. The static pressure is clearly lower and then the airflow as well. In a no-obstacle environment, a 0–5 % peak. The biggest differences are at very low speeds around 600 rpm and then from 1300 rpm upwards.

At higher speeds, compared to the black fan, we have already measured some vibrations – in both the Y and the Z exis (for the black one, there are only symbolic vibrations in the X axis). These are still very low vibrations, but finding them nicely shows that the white fan blades are undulating more. And that is why the differences in noise are achieved. With the hexagonal grille, where the fan is fighting the higher resistance of the environment and the sound of microturbulence enters into it as well, the difference is already noticeable at the same speed. Therefore, when aligned to the same noise level, the difference in airflow is up to 10 % against the white fan in the worst case. That’s at higher speeds, around 1200 rpm. At low speeds, below 600, the white fan has a rare advantage, probably due to the creation of fewer microturbulences on sharp edges, where less stability in blade shape can have a paradoxically beneficial effect. At very low speeds, even resonant frequencies do not enter the sound as strongly.

In the 33 dBA mode already, (i.e. the second quietest mode, where the white fan is still a hair weaker in terms of performance) there are small peaks in the sound in the most intense 300–400 Hz aerodynamic band. These only start to appear with the white fan at higher speeds, along with measurable vibrations. However, amplified are usually the lower, rumbling frequencies between 100–200 Hz. So the white fan can and will be more rumbly at higher speeds compared to the black fan. And it’s also worth noting that all of these findings of ours are based on measurements on two fan samples. Two of both white and black. The black and white results match perfectly. So it is by no means a question of the pieces being different as such, regardless of colour. The influence of the rigidity of the material on the properties is unambiguous. With the AF120 Elite fans, however, Corsair has still managed everything very well. We will definitely be covering this topic and see how other manufacturers will have handled recolouring.

Finally, we will come to the word “rigidity”. This is a term we often use with fans and, unless we’re overstating it, it always means compressive strength. The Corsair AF120 Elite has a high compressive strength (although it’s a bit lower on the white one), but then there’s also the tensile strength. And that’s another property that expresses how much tension the material can withstand when bent before it starts to crack. The tensile strength of the AF120 Elite, on the other hand, is quite low. An example of a material that has a very high compressive strength and at the same time a very low tensile strength is also glass. You can break this fan in a similar way as you can see in the pictures below. One of the samples has already come to us in multiple pieces (or small pieces), which is also a bad calling card for the poor protection of the insufficient packaging, which does not absorb enough shock during transport.

English translation and edit by Jozef Dudáš

Severed rotor, broken blades and stator struts, possible consequence of the absence of any protection in the package

   



It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Results: Airlow w/o obstacles






Why is there a missing value sometimes? There may be more reasons. Usually it is because the fan could not be adjusted to the target noise level. Some have a higher minimum speed (or the speed is low, but the motor is too noisy) or it is a slower fan that will not reach the higher decibels. But the results in the graphs are also missing if the rotor is brushing against the nylon filter mesh. In that case, we evaluate this combination as incompatible. And zero in the graphs is naturally also in situations where we measure 0.00. This is a common occurrence at extremely low speeds with obstructions or within vibration measurements.



It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Results: Static pressure w/o obstacles

Explanatory note: We measure the static pressure without obstacles in two different situations – with laminarized and with turbulent intake. The results in this section reflect the first case. That is, the air intake to the rotor is laminarized by the guidance tunnel. In this setup, the fan has a larger volume of air available and the achieved static pressure values are higher. This is the optimum environment in which to measure fan parameters correctly.








It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Results: Frequency response of sound w/o obstacles

Measurements are performed in the TrueRTA application, which records sound in a range of 240 frequencies in the recorded range of 20–20,000 Hz. For the possibility of comparison across articles, we export the dominant frequency from the low (20–200 Hz), medium (201–2,000 Hz) and high (2,001–20,000 Hz) range to standard bar graphs.

However, for an even more detailed analysis of the sound expression, it is important to perceive the overall shape of the graph and the intensity of all frequencies/tones. If you don’t understand something in the graphs or tables below, you’ll find the answers to all your questions in this article. It explains how to read the measured data below correctly.

Noise levels in the tables below -85 dBu (but watch out for the negative sign, -70 dBu is louder than -80 dBu) can be ignored. They are in fact extremely weak and always far below the limit of human perception, and are often defined by the “noise” of the measurement string. Therefore, do not take even the dominant frequencies within the treble band that exceed 12 kHz too much into account.



FanDominant sound freq. and noise level, no obstacle@33 dBANF-F12 PWMNF-A15 PWM
Low rangeMid rangeHigh range
Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]
Corsair AF120 Elite (white)123,4-80,5369,7-82,419330,5-90,9
Corsair AF120 Elite (black)130,7-76,9369,7-81,619330,5-90,7
Cooler Master MasterFan SF120M50,4-80,91566,8-87,54305,4-89,2
Akasa Alucia SC1250,4-78,81566,8-91,519330,5-90,7
BeQuiet! Silent Wings Pro 4 (BL098)50,4-81,2339,0-80,82712,2-87,1
Thermalright X-Silent 12050,4-79,0369,7-88,64974,287,1
Fractal Design Aspect 12 RGB PWM10,4-72,6380,5-80,519897,0-90,9
BeQuiet! Silent Wings 3 (BL066)97,9-75,21107,9-79,119897,0-90,8
Gelid Zodiac119,9-71,1380,5-80,919897,0-91,0
Fractal Design Dynamic X2 GP-12 PWM47,6-72,81015,4-79,919897,0-90,7
BeQuiet! Pure Wings 2 (BL039)47,6-78,3958,9-79,618780,2-91,0
Gigabyte Aorus 120 ARGB130,7-82,8369,7-80,719330,5-90,9
Arctic BioniX P120 A-RGB127,0-73,2219,8-70,419897,0-90,9
Akasa OTTO SF12N/AN/AN/AN/AN/AN/A
Cooler Master SickleFlow 120 ARGBCooler Master SickleFlow 120 ARGB59,9-82,71076,3-83,119330,5-90,7
Alphacool SL-15 PWMAlphacool SL-15 PWM190,3-79,1380,5-78,818245,6-90,9
Arctic BioniX F120Arctic BioniX F120123,4-76,5246,8-80,019330,5-90,9
SilverStone SST-AP123SilverStone SST-AP123100,8-81,2415,0-78,619897,0-90,8
Noctua NF-P12 redux-1700 PWM103,7-76,9329,4-73,219330,5-90,9
SilentiumPC Fluctus 120 PWMSilentiumPC Fluctus 120 PWM130,7-76,5369,7-80,919897,0-91,0
MSI MEG Silent Gale P1250,4-81,7369,7-81,019897,0-90,8
Asus ROG Strix XF12050,4-80,2329,4-76,219330,5-90,8
Akasa Vegas X7123,4-77,0369,7-83,319330,5-90,7
Reeven Coldwing 1238,9-79,71317,5-84,019330,5-90,7
Reeven Kiran138,5-80,6369,7-83,319330,5-90,8
SilentiumPC Sigma Pro 120 PWMN/AN/AN/AN/AN/AN/A
SilentiumPC Sigma Pro Corona RGB 12092,4-83,0369,7-78,018780,2-90,8
SilverStone SST-AP12147,6-77,5261,4-86,419330,5-91,0
SilverStone SST-FQ12131,3-87,51208,2-79,919330,5-90,9
Xigmatek XLF-F125641,8-69,7213,6-77,319330,5-91,0
/* Here you can add custom CSS for the current table */ /* Lean more about CSS: https://en.wikipedia.org/wiki/Cascading_Style_Sheets */ /* To prevent the use of styles to other tables use "#supsystic-table-1752" as a base selector for example: #supsystic-table-1752 { ... } #supsystic-table-1752 tbody { ... } #supsystic-table-1752 tbody tr { ... } */



FanDominant sound freq. and noise level, no obstacle@39 dBANF-F12 PWMNF-A15 PWM
Low rangeMid rangeHigh range
Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]
Corsair AF120 Elite (white)169,5-76,1369,7-74,419330,5-90,7
Corsair AF120 Elite (black)164,7-79,2339,0-72,819897,0-90,7
Cooler Master MasterFan SF120M146,7-78,8380,5-77,92635,0-83,9
Akasa Alucia SC12151,0-80,6226,3-73,419897,0-90,7
BeQuiet! Silent Wings Pro 4 (BL098)20,3-76,0339,0-73,92791,7-83,3
Thermalright X-Silent 120119,9-74,0369,7-75,010540,1-88,4
Fractal Design Aspect 12 RGB PWM10,4-64,7380,5-72,418780,2-90,8
BeQuiet! Silent Wings 3 (BL066)130,7-70,1369,7-75,218780,2-90,9
Gelid Zodiac146,7-66,0369,7-73,419330,5-90,9
Fractal Design Dynamic X2 GP-12 PWM113,1-78,5329,4-74,618780,2-90,9
BeQuiet! Pure Wings 2 (BL039)130,7-76,1369,7-75,319330,5-90,7
Gigabyte Aorus 120 ARGB127,0-63,9380,5-74,219330,5-90,6
Arctic BioniX P120 A-RGB50,4-77,9439,7-72,418780,2-91,1
Akasa OTTO SF12155,4-67,1380,5-75,24431,5-87,2
Cooler Master SickleFlow 120 ARGB109,9-72,81076,3-72,418245,6-90,9
Alphacool SL-15 PWM184,9-73,4380,5-73,119897,0-90,8
Arctic BioniX F120151,0-69,3369,7-75,318780,2-91,0
SilverStone SST-AP123SilverStone SST-AP123164,7-75,1369,7-74,719330,5-90,9
Noctua NF-P12 redux-1700 PWM146,7-68,0329,4-66,718245,6-91,1
SilentiumPC Fluctus 120 PWM190,3-69,2380,5-68,819897,0-90,9
MSI MEG Silent Gale P1223,1-73,5380,5-73,318780,2-90,9
Asus ROG Strix XF120130,7-70,9369,7-75,219330,5-91,0
Akasa Vegas X7127,0-77,8369,7-76,119330,5-90,9
Reeven Coldwing 12160,0-74,0369,7-76,819330,5-90,9
Reeven Kiran184,9-75,2369,7-75,217726,2-89,8
SilentiumPC Sigma Pro 120 PWM97,9-81,7369,7-78,12635,0-86,2
SilentiumPC Sigma Pro Corona RGB 12020,3-67,8380,5-77,32487,1-85,6
SilverStone SST-AP121103,7-76,2339,0-73,42031,9-86,5
SilverStone SST-FQ121138,5-75,21208,2-71,118780,2-90,9
Xigmatek XLF-F1256190,3-77,1761,1-75,519897,0-91,1
/* Here you can add custom CSS for the current table */ /* Lean more about CSS: https://en.wikipedia.org/wiki/Cascading_Style_Sheets */ /* To prevent the use of styles to other tables use "#supsystic-table-1755" as a base selector for example: #supsystic-table-1755 { ... } #supsystic-table-1755 tbody { ... } #supsystic-table-1755 tbody tr { ... } */



FanDominant sound freq. and noise level, no obstacle@45 dBANF-F12 PWMNF-A15 PWM
Low rangeMid rangeHigh range
Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]
Corsair AF120 Elite (white)155,4-77,9320,0-67,818780,2-90,8
Corsair AF120 Elite (black)130,7-78,5369,7-70,919330,5-90,9
Cooler Master MasterFan SF120M184,9-69,7369,7-67,94305,4-73,8
Akasa Alucia SC12130,7-70,3380,5-68,62031,9-90,4
BeQuiet! Silent Wings Pro 4 (BL098)47,6-75,5339,0-60,52791,7-83,3
Thermalright X-Silent 12082,3-74,9369,7-91,118780,2-91,1
Fractal Design Aspect 12 RGB PWM130,7-69,5285,1-67,219330,5-90,8
BeQuiet! Silent Wings 3 (BL066)164,7-64,2339,0-69,118780,2-90,8
Gelid Zodiac184,9-61,8380,5-68,718780,2-90,8
Fractal Design Dynamic X2 GP-12 PWM146,7-72,2391,7-71,52280,7-83,1
BeQuiet! Pure Wings 2 (BL046)130,7-71,3369,7-69,319330,5-90,6
Gigabyte Aorus 120 ARGB160,0-61,9369,7-70,42487,1-88,3
Arctic BioniX P120 A-RGB151,0-67,6369,7-72.02873,5-90,6
Akasa OTTO SF12Akasa OTTO SF12190,3-66,7380,5-67,74695,1-86,3
Cooler Master SickleFlow 120 ARGBCooler Master SickleFlow 120 ARGB138,5-66,71076,3-67,13835,7-89,2
Alphacool SL-15 PWMAlphacool SL-15 PWM151-71,5232,9-66,52280,7-89,6
Arctic BioniX F120Arctic BioniX F120195,8-68201,6-62,219330,5-90,7
SilverStone SST-AP123SilverStone SST-AP123130,7-70,9369,7-67,719897-90,9
Noctua NF-P12 redux-1700 PWMNoctua NF-P12 redux-1700 PWM184,9-62,2369,7-6719897-91,1
SilentiumPC Fluctus 120 PWMSilentiumPC Fluctus 120 PWM25,6-69,6239,7-622957,7-90
MSI MEG Silent Gale P12MSI MEG Silent Gale P1228,3-73,5380,5-712487,1-85,9
Asus ROG Strix XF120Asus ROG Strix XF12023,1-62,9369,7-71,72347,5-89,1
Akasa Vegas X7Akasa Vegas X7N/AN/AN/AN/AN/AN/A
Reeven Coldwing 12Reeven Coldwing 12195,8-68,3380,5-71,32031,9-90
Reeven KiranReeven Kiran130,7-73,1219,8-70,217726,2-89,4
SilentiumPC Sigma Pro 120 PWMSilentiumPC Sigma Pro 120 PWM138,5-701522,2-71,92560-82,9
SilentiumPC Sigma Pro Corona RGB 120SilentiumPC Sigma Pro Corona RGB 120190,3-70,5369,7-66,52416,3-82,4
SilverStone SST-AP121SilverStone SST-AP121130,7-66,8439,7-66,12347,5-90,9
SilverStone SST-FQ121SilverStone SST-FQ121127-71,2369,7-66,52031,8-81,4
Xigmatek XLF-F1256Xigmatek XLF-F1256130,7-71,5239,7-60,918780,2-91,1
/* Here you can add custom CSS for the current table */ /* Lean more about CSS: https://en.wikipedia.org/wiki/Cascading_Style_Sheets */ /* To prevent the use of styles to other tables use "#supsystic-table-1756" as a base selector for example: #supsystic-table-1756 { ... } #supsystic-table-1756 tbody { ... } #supsystic-table-1756 tbody tr { ... } */
Why is there a missing value sometimes? There may be more reasons. Usually it is because the fan could not be adjusted to the target noise level. Some have a higher minimum speed (or the speed is low, but the motor is too noisy) or it is a slower fan that will not reach the higher decibels. But the results in the graphs are also missing if the rotor is brushing against the nylon filter mesh. In that case, we evaluate this combination as incompatible. And zero in the graphs is naturally also in situations where we measure 0.00. This is a common occurrence at extremely low speeds with obstructions or within vibration measurements.




It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Results: Airflow through a hexagonal grille






Why is there a missing value sometimes? There may be more reasons. Usually it is because the fan could not be adjusted to the target noise level. Some have a higher minimum speed (or the speed is low, but the motor is too noisy) or it is a slower fan that will not reach the higher decibels. But the results in the graphs are also missing if the rotor is brushing against the nylon filter mesh. In that case, we evaluate this combination as incompatible. And zero in the graphs is naturally also in situations where we measure 0.00. This is a common occurrence at extremely low speeds with obstructions or within vibration measurements.



It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Results: Static pressure through a hexagonal grille






Why is there a missing value sometimes? There may be more reasons. Usually it is because the fan could not be adjusted to the target noise level. Some have a higher minimum speed (or the speed is low, but the motor is too noisy) or it is a slower fan that will not reach the higher decibels. But the results in the graphs are also missing if the rotor is brushing against the nylon filter mesh. In that case, we evaluate this combination as incompatible. And zero in the graphs is naturally also in situations where we measure 0.00. This is a common occurrence at extremely low speeds with obstructions or within vibration measurements.



It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Results: Frequency response of sound with a hexagonal grille

Measurements are performed in the TrueRTA application, which records sound in a range of 240 frequencies in the recorded range of 20–20,000 Hz. For the possibility of comparison across articles, we export the dominant frequency from the low (20–200 Hz), medium (201–2,000 Hz) and high (2,001–20,000 Hz) range to standard bar graphs.

However, for an even more detailed analysis of the sound expression, it is important to perceive the overall shape of the graph and the intensity of all frequencies/tones. If you don’t understand something in the graphs or tables below, you’ll find the answers to all your questions in this article. It explains how to read the measured data below correctly.

Noise levels in the tables below -85 dBu (but watch out for the negative sign, -70 dBu is louder than -80 dBu) can be ignored. They are in fact extremely weak and always far below the limit of human perception, and are often defined by the “noise” of the measurement string. Therefore, do not take even the dominant frequencies within the treble band that exceed 12 kHz too much into account.


FanDominant sound freq. and noise level, hexagonal grille@33 dBANF-F12 PWMNF-A15 PWM
Low rangeMid rangeHigh range
Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]
Corsair AF120 Elite (white)50,4-81,5369,7-80,219330,5-90,8
Corsair AF120 Elite (black)49,0-80,0339,0-78,219330,5-90,8
Cooler Master MasterFan SF120M41,8-72,3380,5-83,14305,4-90,7
Akasa Alucia SC1241,8-80,9369,7-80,619897,0-90,9
BeQuiet! Silent Wings Pro 4 (BL098)47,6-75,2329,4-74,919330,5-90,8
Thermalright X-Silent 12050,4-77,7339-76,018780,2-90,9
Fractal Design Aspect 12 RGB PWM59,9-78,7349,0-69,819330,5-90,7
BeQuiet! Silent Wings 3 (BL066)49,079,3329,478,919897,0-90,8
Gelid Zodiac50,4-81,0339,0-78,719330,5-90,7
Fractal Design Dynamic X2 GP-12 PWM38,9-75,4339,0-79,518780,243-90,9
BeQuiet! Pure Wings 2 (BL039)47,6-79,7380,5-81,219330,5-90,9
Gigabyte Aorus 120 ARGB146,7-82,0369,7-82,019330,5-90,9
Arctic BioniX P120 A-RGB169,5-84,7329,4-70,519330,5-90,8
Akasa OTTO SF12N/AN/AN/AN/AN/AN/A
Cooler Master SickleFlow 120 ARGB47,6-82,01076,3-81,918780,2-90,9
Alphacool SL-15 PWMN/AN/AN/AN/AN/AN/A
Arctic BioniX F120169,5-84,7329,4-70,519330,5-90,8
SilverStone SST-AP123SilverStone SST-AP12358,2-80,8302,0-77,319897,0-90,8
Noctua NF-P12 redux-1700 PWM100,8-81,6329,4-69,819897,0-90,9
SilentiumPC Fluctus 120 PWM100,8-82,7369,7-79,919330,5-90,8
MSI MEG Silent Gale P12100,8-81,2369,7-80,319330,5-91,0
Asus ROG Strix XF12038,9-79,7349,0-72,419897,0-90,9
Akasa Vegas X7100,8-85,3359,2-80,319330,5-90,9
Reeven Coldwing 1292,4-79,8369,7-79,719897,0-90,9
Reeven Kiran97,9-84,4391,7-77,819897,0-90,8
SilentiumPC Sigma Pro 120 PWMN/AN/AN/AN/AN/AN/A
SilentiumPC Sigma Pro Corona RGB 12092,4-83,0369,7-77,918780,2-90,8
SilverStone SST-AP121113,1-80,9246,8-84,019330,5-90,9
SilverStone SST-FQ12138,9-76,91522,2-81,718780,2-90,9
Xigmatek XLF-F125663,5-83,3380,5-78,819897,0-90,9
/* Here you can add custom CSS for the current table */ /* Lean more about CSS: https://en.wikipedia.org/wiki/Cascading_Style_Sheets */ /* To prevent the use of styles to other tables use "#supsystic-table-1757" as a base selector for example: #supsystic-table-1757 { ... } #supsystic-table-1757 tbody { ... } #supsystic-table-1757 tbody tr { ... } */



FanDominant sound freq. and noise level, hexagonal grille@39 dBANF-F12 PWMNF-A15 PWM
Low rangeMid rangeHigh range
Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]
Corsair AF120 Elite (white)130,7-71,3369,7-74,119330,5-90,8
Corsair AF120 Elite (black)138,5-73,2369,7-73,519897,0-90,8
Cooler Master MasterFan SF120M95,1-79,1380,5-72,74305,4-89,6
Akasa Alucia SC1250,4-80,5369,7-72,619897,0-90,9
BeQuiet! Silent Wings Pro 4 (BL098)155,4-78,6339,0-71,32712,2-87,9
Thermalright X-Silent89,8-77,8359,2-68,818780,2-91,0
Fractal Design Aspect 12 RGB PWM50,4-82,7380,5-68,719330,5-90,8
BeQuiet! Silent Wings 3 (BL066)BeQuiet! Silent Wings 3 (BL066)58,2-78,5339,0-73,218780,2-90,9
Gelid Zodiac50,4-80,5380,5-73,219330,5-91,0
Fractal Design Dynamic X2 GP-12 PWM50,4-80,4320,0-65,319330,5-90,8
BeQuiet! Pure Wings 2 (BL039)59,9-79,0369,7-73,318780,2-90,8
Gigabyte Aorus 120 ARGB138,5-81,5369,7-73,218780,2-90,9
Arctic BioniX P120 A-RGB151,0-82,0219,8-62,219330,5-90,9
Akasa OTTO SF12N/AN/AN/AN/AN/AN/A
Cooler Master SickleFlow 120 ARGB130,7-81,8369,7-75,42152,7-90,4
Alphacool SL-15 PWM146,7-76,8369,7-73,218780,2-91,0
Arctic BioniX F12047,6-82,0369,7-72,019330,5-90,9
SilverStone SST-AP123SilverStone SST-AP123100,8-80,7339,0-73,418780,2-90,9
Noctua NF-P12 redux-1700 PWM116,5-76,8339,0-69,419330,5-90,8
SilentiumPC Fluctus 120 PWM146,7-78,0586,9-72,718780,2-91,0
MSI MEG Silent Gale P12164,7-80,6339,0-71,119330,5-91,0
Asus ROG Strix XF120Asus ROG Strix XF12092,4-77,5369,7-70,119330,5-91,2
Akasa Vegas X731,3-84,3369,7-74,318780,2-90,8
Reeven Coldwing 12123,4-71,2380,5-71,318780,2-90,9
Reeven Kiran127,0-77,7380,5-73,419330,5-90,8
SilentiumPC Sigma Pro 120 PWM63,5-83,4380,5-74,02347,5-83,8
SilentiumPC Sigma Pro Corona RGB 120134,5-72,0369,7-75,119897,0-90,8
SilverStone SST-AP12153,4-81,9380,5-70,819330,5-91,0
SilverStone SST-FQ12141,8-78,9369,7-75,019330,5-91,1
Xigmatek XLF-F125650,4-70,3246,8-72,719897,0-90,9
/* Here you can add custom CSS for the current table */ /* Lean more about CSS: https://en.wikipedia.org/wiki/Cascading_Style_Sheets */ /* To prevent the use of styles to other tables use "#supsystic-table-1758" as a base selector for example: #supsystic-table-1758 { ... } #supsystic-table-1758 tbody { ... } #supsystic-table-1758 tbody tr { ... } */

FanDominant sound freq. and noise level, hexagonal grille@45 dBANF-F12 PWMNF-A15 PWM
Low rangeMid rangeHigh range
Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]Frequency [Hz]Noise level [dBu]
Corsair AF120 Elite (white)164,7-70,9369,7-69,719330,5-90,8
Corsair AF120 Elite (black)179,6-69,1369,7-67,519330,5-91,0
Cooler Master MasterFan SF120M119,9-74,3380,5-67,94182,8-87,8
Akasa Alucia SC12151,0-76,6369,7-67,219330,5-90,7
BeQuiet! Silent Wings Pro 4 (BL098)190,3-74,5339,0-62,42791,7-83,7
Thermalright X-Silent 120119,9-73,4369,7-69,54974,2-88,6
Fractal Design Aspect 12 RGB PWM119,9-78,5586,9-69,319897,0-90,7
BeQuiet! Silent Wings 3 (BL066)127,0-68,7391,7-67,219897,0-90,9
Gelid Zodiac127,0-71,9380,5-64,319330,5-91,1
Fractal Design Dynamic X2 GP-12 PWM123,4-75,01566,8-76,72215,8-89,6
BeQuiet! Pure Wings 2 (BL039)75,5-78,2369,7-69,919330,5-90,9
Gigabyte Aorus 120 ARGB164,7-78,7678,1-66,819897,0-90,7
Arctic BioniX P120 A-RGB130,7-70,6219,8-62,218780,2-90,7
Akasa OTTO SF12160,0-72,4369,7-66,44431,5-88,3
Cooler Master SickleFlow 120 ARGB127,0-74,1380,5-66,06267,2-90,8
Alphacool SL-15 PWM134,5-72,2369,7-66,719330,5-90,8
Arctic BioniX F12058,2-78,7369,7-68,318780,2-90,7
SilverStone SST-AP123SilverStone SST-AP123130,7-79,2339,0-69,618780,2-90,9
Noctua NF-P12 redux-1700 PWM142,5-72,8586,9-65,018780,2-90,8
SilentiumPC Fluctus 120 PWM174,5-75,1349,0-62,12957,7-90,3
MSI MEG Silent Gale P12130,7-72,7415,0-67,22487,1-89,3
Asus ROG Strix XF120Asus ROG Strix XF120119,9-73,4604,1-69,619330,5-90,9
Akasa Vegas X733,6-81,5427,1-70,218780,2-90,9
Reeven Coldwing 12160,0-64,7369,7-67,719897,0-91,0
Reeven Kiran155,4-73,7369,7-69,719330,5-90,8
SilentiumPC Sigma Pro 120 PWM130,7-77,8369,7-68,14431,5-90,1
SilentiumPC Sigma Pro Corona RGB 12020,3-64,8369,7-73,03044,4-89,2
SilverStone SST-AP121130,7-78,0570,2-64,918780,2-91,2
SilverStone SST-FQ121169,5-64,4246,8-77,419330,5-91,0
Xigmatek XLF-F125658,2-74,2678,0-63,819330,5-90,9
/* Here you can add custom CSS for the current table */ /* Lean more about CSS: https://en.wikipedia.org/wiki/Cascading_Style_Sheets */ /* To prevent the use of styles to other tables use "#supsystic-table-1759" as a base selector for example: #supsystic-table-1759 { ... } #supsystic-table-1759 tbody { ... } #supsystic-table-1759 tbody tr { ... } */
Why is there a missing value sometimes? There may be more reasons. Usually it is because the fan could not be adjusted to the target noise level. Some have a higher minimum speed (or the speed is low, but the motor is too noisy) or it is a slower fan that will not reach the higher decibels. But the results in the graphs are also missing if the rotor is brushing against the nylon filter mesh. In that case, we evaluate this combination as incompatible. And zero in the graphs is naturally also in situations where we measure 0.00. This is a common occurrence at extremely low speeds with obstructions or within vibration measurements.



It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Reality vs. specifications

Explanatory note: For a quick overview of how manufacturers “spice up” specifications, we have a sort of “truthfulness” coefficient. We calculate this by putting our measured values in proportion to those given in the specifications by the fan manufacturers. A result of “1.00” means that the claimed parameters match the values we have recorded. After such a finding, we can conclude that the manufacturer has done his job honestly and the way he presents the fan agrees. The more the coefficient number is different from 1.00, the less accurate the claimed specifications are. Of course, the better case for the user is if the coefficient is higher than 1.00 (and it is, for example, 1.20), then the real parameters exceed the paper ones. Conversely, if the coefficient starts with zero, then the fan does not reach the parameters on paper. For example, a value of 0.80 means that the real airflow or static pressure is 20 % lower than the manufacturer claims.



It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Results: Vibration, in total (3D vector length)

Explanatory note: We have already described how to measure the vibration intensity in the corresponding chapter within the testing methodology. However, it is worth emphasizing here what we mean by “cumulative vibration”. In order to be able to interpret the motion in the three axes (X, Y, Z) as simply as possible, with a single number, we consider the individual axes as vectors and calculate the so-called 3D vector from them. A more detailed account at the level of the individual axes is given in the next three chapters of this article.





Prečo niekedy pri ventilátore chýba hodnota? Dôvodov môže byť viac. Obvykle je to preto, že nebolo možné daný ventilátor zregulovať na cieľovú hladinu hluku. Niektoré majú vyššie minimálne otáčky (alebo otáčky sú aj nízke, ale má príliš hlučný motorček) alebo sa jedná o pomalší ventilátor, ktorý si na vyššie decibely nedosiahne. Výsledky v grafoch chýbajú ale aj vtedy, ak sa rotor šúcha o sito nylonového filtra. Vtedy túto kombináciu vyhodnocujeme ako nekompatibilnú. A nula v grafoch je prirodzene aj v situáciách, kedy nameriame 0,00. To je častý jav pri mimoriadne nízkych otáčkach s prekážkami alebo v rámci meraní vibrácií.



It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Results: Vibration, X-axis





Why is there a missing value sometimes? There may be more reasons. Usually it is because the fan could not be adjusted to the target noise level. Some have a higher minimum speed (or the speed is low, but the motor is too noisy) or it is a slower fan that will not reach the higher decibels. But the results in the graphs are also missing if the rotor is brushing against the nylon filter mesh. In that case, we evaluate this combination as incompatible. And zero in the graphs is naturally also in situations where we measure 0.00. This is a common occurrence at extremely low speeds with obstructions or within vibration measurements.



It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Results: Vibration, Y-axis





Why is there a missing value sometimes? There may be more reasons. Usually it is because the fan could not be adjusted to the target noise level. Some have a higher minimum speed (or the speed is low, but the motor is too noisy) or it is a slower fan that will not reach the higher decibels. But the results in the graphs are also missing if the rotor is brushing against the nylon filter mesh. In that case, we evaluate this combination as incompatible. And zero in the graphs is naturally also in situations where we measure 0.00. This is a common occurrence at extremely low speeds with obstructions or within vibration measurements.



It’s not a joke. It may sound silly, but know that we are not presenting you with anti-physics. On the contrary, the foundation in physics here is very good. Different colors of materials, as long as they are not just a surface finish but a homogeneous mass, can (and often do) affect their strength. Our sensitive measurements reveal well the differences that exist between the white and black variants.

Results: Vibration, Z-axis





Why is there a missing value sometimes? There may be more reasons. Usually it is because the fan could not be adjusted to the target noise level. Some have a higher minimum speed (or the speed is low, but the motor is too noisy) or it is a slower fan that will not reach the higher decibels. But the results in the graphs are also missing if the rotor is brushing against the nylon filter mesh. In that case, we evaluate this combination as incompatible. And zero in the graphs is naturally also in situations where we measure 0.00. This is a common occurrence at extremely low speeds with obstructions or within vibration measurements.


Continue: Results: Vibration, X-axis