Noctua NF-B9 Review

hardnrg - 2008-06-09 11:01:43 in Cooling
Category: Cooling
Reviewed by: hardnrg   
Reviewed on: June 23, 2008
Price: $18.99

Introduction:

Noctua is an Austrian company that combines industrial experience and knowledge of cooling products, with scientific research in fan technology. This is a recipe for high performance cooling, but with the least amount of noise, which is something Noctua continues to prove possible with each new product.

This time, Noctua enters a 92mm fan into the cooling arena. The claim is that the fan's design gives rise to a high level of airflow and air pressure for its speed. So, even though its RPMs are low, it can provide the same level of performance as faster turning fans, and the NF-B9's lower speed means that the fan produces less noise.

At least, this is the theory that Noctua would have us believe, so let's put the theory to the test!

 

Closer Look:

The box design does not differ from that of Noctua's other fans, and to a certain extent, all their products. You get a cheeky peek at the fan and some of the extras through the clear plastic window, and a summary of the unique aspects of the fan's technology. Each of these aspects is explained briefly on the back of the box, along with the more commonly expected fan specifications.

 

 

If you look closely at the top-left corner of the back of the box, you can see a small white triangle that says "OPEN". I haven't noticed this before, but the other Noctua fan boxes I have also have this, and the rear of the box actually opens up to explain Noctua's unique technology even further, with illustrated diagrams and graphs. It's actually pretty cool to see all this information on the box as it answers the question that would be going through a potential buyer's mind, "what's so special about this fan that makes it better than others?"

 

Closer Look:

After having a good read about the fan, it's time to look at what you get for your money.

The inner packaging keeps the fan separate from the various accessories.

 

 

Taking a look at the fan itself, you can see that the frame of the fan is quite standard, but also the fan blades blatantly stray far from the norm.

 

 

It's not very often that fans come with "solid" corners, but I find it extremely annoying when they are, as it can limit the methods of mounting the fan with clips or screws. Nice to see that Noctua doesn't partake in such tomfoolery.

 

So, these fan blades are the most unique and important part of this fan design. Noctua develops its fans alongside the Austrian Institute of Heat Transmission and Fan Technology (Österreichisches Institut für Wärmeübertragung und Ventilatorentechnik, ÖIWV). Catchy name, huh?! I guess it's less of a mouthful in Austrian. Anyway, the ends of the fan blades are bevelled to give a precisely designed gap away from the fan frame, and by reducing the impedance of air right on the surface of the frame, the airflow and pressure are optimised. The trailing edge of the fan blades are notched with the intention of spreading the noise over a range of frequencies by making multiple small vortices.

Fancy a better look at them after all that techy-talk? Ok, here you go.

 

 

Noctua do a fantastic job of sleeving the fan cable, and it's not your run-of-the-mill nylon braided sleeving either, as it's like a woven fabric - flexible and doesn't snag on things like nylon sleeving.

 

Along with the fan, you get several accessories. In the photo below, you can see from left to right, four vibration compensators for vibration-isolating the fan from your case, four standard self tapping fan screws, an Ultra Low Noise Adaptor for the maximum reduction of noise (and speed), a Low Noise Adaptor for moderate reduction of noise (and speed), and a 4-pin "Molex" to 3-pin fan socket adaptor.

 

"Fine fine fine, but how does it compare to other 92mm fans?", I hear you ask. Okay, keep your hair on! Let's see how the NF-B9 does against a range of other fans, from near silent to extreme performance.

 

Specifications

 

Size 92x92x25 mm
Bearing SSO-Bearing
Blade Geometry NF-B9 Blade Design with BBT & VCN
Rotational Speed (+/- 10%) 1600 RPM
Rotational Speed with L.N.A. (+/- 10%) 1300 RPM
Rotational Speed with U.L.N.A. (+/- 10%) 1000 RPM
Airflow 64.3 m³/h
Airflow with L.N.A. 52.6 m³/h
Airflow with U.L.N.A. 41.1 m³/h
Acoustical Noise 17.6 dB(A)
Acoustical Noise with L.N.A. 13.1 dB(A)
Acoustical Noise with U.L.N.A. 7.9* dB(A)
Static Pressure 1.61 mm H2O
Static Pressure with L.N.A. 1.09 mm H2O
Static Pressure with U.L.N.A. 0.64 mm H2O
Input Power 1.32 W
Input Current 0.11 A
Voltage Range 12 V
MTBF > 150,000 h
Scope of Delivery
  • Ultra-Low-Noise Adaptor (U.L.N.A.)
  • Low-Noise Adaptor (L.N.A.)
  • 4 Vibration Compensators
  • 3:4-Pin Adaptor
  • 4 Fan Screws
Warranty 6 years

*extrapolated values

 

Features

 

Testing:

The most unique features of this fan in terms of performance, are its claims of having a low noise signature, and having decent airflow and static pressure for its relatively low rotational speed.

I decided that testing the fan with no resistance would not be all that meaningful, and difficult to quantify in measurements. So I devised a testing setup that would let me compare the fans ability to provide sufficient airflow in a demanding scenario.

Heatsinks without heatpipes rely solely on forced air to remove the heat from the fins. The Thermalright SLK-948u is one such heatsink.

 

 

This heatsink is pure copper and the fins are arranged very closely together.

 

The largest fan size that the heatsink can take is 92mm; the fan clips (not pictured) allow either 25mm or 38mm thick fans to be secured in place.

 

If the fan can provide sufficient static pressure through the tightly packed fins, it will result in good cooling of the heatsink. Also, noise levels can be observed with the fan in a real world situation.

 

Testing Setup:

Comparison CPU Fans:

 

Brand Manufacturer Model Bearing Speed Static Pressure Air Flow Noise
        RPM mm-H2O CFM dBA
Thermaltake VetteCorp A9225L12S Sleeve 2000 1.59 36.78 30
Noctua Noctua NF-B9-1600 SSO (Oil Pressure) 1600 1.61 37.8 17.6
Panaflo NMB-Mat FBA09A12M Hydro Wave 2450 2.79 48 30
Jamicon Kaimei JF0925H1MSAR Sleeve (HTLS) 2500 2.92 46.31 32
Zalman Qwikflow RDH9025B Ball 2800 3.34 54.63 32.96
Thermaltake Everflow F129025BH 2 Ball 2850 3.98 56 35
Vantec Vantec TD8038H Double Ball 4800 16.76 119 56.4

 

Overclocking:

Overclocked settings:

As well as stock voltage and speed on the CPU, I tested the highest stable CPU overclock for a higher heatload for the fans to deal with.

 

Benchmarks:

 

Testing:

OCCT v2.0 Perestroika supports multi-core processors and will produce the most CPU heat of any stress test. Prime95 and its variants can produce the same heat load, but only after a longer period of time. OCCT has a standardised test that lasts for 1 hour, and produce temperature charts at the end of a successful run. The mean average of the CPU temperature over the second half of the test was recorded as a meaningful sustained heat load.









Settings:


bleh


Testing:

There is a lot of discrepancy between the measured noise levels that each manufacturer specifies for their fans. So, in this section, instead of measuring a raw noise level with a microphone, I chose to observe the relative noise levels of each fan by ear in a room with near silent ambience. In this way, I could gauge the perceived intensity of the noise (volume), as well as how annoying the sound was for each fan.

Some fans spread the noise over a range of frequencies so that the resulting sound is smooth, whereas other fans have a characteristic whine or whirring sound.

In this test I have devised a noise level scaled from 1 to 10:

  1. Silent, inaudible
  2. Near silent, only just audible
  3. Near silent, very quiet
  4. Audible, soft noise
  5. Audible, moderate noise
  6. Fairly loud, moderate noise
  7. Fairly loud, starting to be annoying
  8. Fairly loud, annoying
  9. Loud, annoying
  10. Very loud, ridiculously annoying

 

 

All the fans were run at 12V, connected to the PSU rather than motherboard headers. Surprisingly, the Noctua NF-B9 was actually slightly louder than the Thermaltake TT-9025A. Both of these fans were only just audible however, and much quieter than the other fans in the test.

 

Conclusion:

Well, I was hoping for a bit more performance out of the NF-B9, but at just 1600 RPM, it's asking quite a lot for it to perform better than fans turning at 2000 RPM and higher, and yet it matched or exceeded the performance of the Thermaltake TT-9025A (2000 RPM) fan. So, in this respect, the design of the NF-B9 fan blades can be seen as having a performance advantage over standard fan blades.

Where the NF-B9 fell a bit short of my expectations, was that it was actually slightly louder than the Thermaltake TT-9025A, a fan that comes free with many Thermaltake cases. So, how does the Noctua NF-B9 warrant the price premium for this fan over a "free" standard fan?

The NF-B9 is not as strong a performer as louder fans, but it's also not as quiet as a standard fan, and barely edges out in performance. I think the fan fails to deliver the required air pressure or airflow to be used in overclocked applications, but would be very much suited to cooling a stock clocked CPU, or as a case fan.

I really would like to see Noctua start producing some higher speed fans using these types of blades, to see how the performance increases and how the noise levels compare to medium and high performance fans. All the Noctua fans I have tested so far have been extremely quiet even at 12V, rendering the adaptors almost useless as the difference in noise is overshadowed by the loss in performance. Having a higher speed to begin with should give a more wholesome range of performance ranging from silent to high cooling ability.

 

Pros:

 

Cons: