Massive Fan Shootout: 25 Products from Arctic Cooling, Cooler Master, GlacialTech, Scythe, Titan, and Zalman

Test Results: 80x80x25 mm Fans

Contents:

The Idea of the Experiment

It's not as easy to test fans (to say nothing of their comparative analysis) as it may seem on the surface of it. The main problem here is to decide upon the criterion of fans' functionality and consequently to choose a reference point for valid comparisons. Technically versed readers may exclaim: "Come on! What about the metering characteristic?! Airflow speed and static pressure! Everything necessary is at hand! No need to "decide upon" anything! Just take fans and get the results!" Indeed, metering characteristic is good. But there are two significant ideological obstacles here.

Obstacle One: metering characteristic is not static, and it does not determine strictly fan's operating mode for possible applications. A diffuser (or confuser) at the output/input of the fan in a circuit (stage) with complex impedance may change fan's behavior beyond recognition. That's why engineers working in the fan industry often rely on experiments instead of "idealized" data from manufacturers. They test fans in their certain conditions.

Obstacle Two: practicality of metering characteristic for PC fans is highly overestimated. A typical PC fan is not a turbocompressor, which pressure is the most important parameter, and it's not an industrial air pump, where the discharge flow is important. It's a cooling device in the first place! So it's actually not important for a computer user, be it a common user or an enthusiast, how much pressure a fan in their cooler generates, and what airflow (accurate within hundredths of m/s) is expelled from the n-th slit of the air grid in a PSU. Cooling efficiency - that's what a user needs! Static pressure and flow rate are just empty phrases, if you don't know the temperature.

Well, we'll use a different method! We'll run the tests on our testbed, which is usually used to analyze CPU coolers (ASUS P5AD2-E Premium motherboard and Intel Pentium 4 550 processor). Fans will be tested with a heatsink from the famous CPU cooler Thermaltake Big Typhoon, which is notable for high impedance (hydraulic resistance) and makes a good reference point in our shootout.

Thus, the main criterion of fan's efficiency is temperature readings (thermal resistance) of the "reference heatsink + fan" assembly. 120x120x25 mm fans are installed directly, and 80x80x25 mm fans are mounted on a simple adapter made of a 120 mm fan case. We'll use additional parameters to evaluate fan's efficiency of cooling near-socket elements - temperature readings of inductance coils of the CPU voltage regulator (temperature of PL24, PL25 and PL26 coil cores installed near the socket). And finally, in order to generate more heat in the testbed, CPU voltage is raised to 1.525 V (the resulting heating capacity is 150 W).

Testbed configuration:

Motherboard: ASUS P5AD2-E Premium rev. 1.05
Processor: Intel Pentium 4 550 (3.4 GHz Prescott, HT Technology)
OS: Microsoft Windows XP

We use the S&M utility to simulate maximum thermal load of a processor, and Speedfan - to monitor temperatures. Thermal Monitor is disabled in all the tests.

Let's proceed to our test results!

80x80x25 mm, normal mode

Diagram 1. Temperature readings (temperature of the processor core)

Diagram 2. Thermal resistance

Note
Thermal resistance θ_ja is defined as the relation
θ_ja = (T_j — T_a)/P_h, where T_j is the temperature of a CPU core, T_a is the environment temperature (it's 25°C in this case), P_h is the thermal capacity of a processor (in this case it's 150 Watts).

Diagram 3. Temperature readings (temperature of near-socket components)

80x80x25 mm fans, 1500 rpm

Diagram 4. Temperature readings (temperature of the processor core)

Diagram 5. Thermal resistance

Note
Thermal resistance θ_ja is defined as the relation
θ_ja = (T_j — T_a)/P_h, where T_j is the temperature of a CPU core, T_a is the environment temperature (it's 25°C in this case), P_h is the thermal capacity of a processor (in this case it's 150 Watts).

Diagram 6. Temperature readings (temperature of near-socket components)

At the end of this article we publish the noise measurement results (the method of testing is described in the article Cooler noise and the noise measurement method) as well as the efficiency/noise rating.

Diagram 7. Noise characteristics

Note: Background noise level 18 dBA

Diagram 8. Efficiency/noise ratio

Note
The efficiency/noise ratio (ENR) is calculated as:

ENR = DM*(R_t/P¢C)/(NL/R_n), where

R_t — reference temperature (the reference thermal resistance θ_ja of the cooling system - 0.25°C/W), TC — the core temperature with the operating cooling system, R_n — reference noise (the reference noise level is 20 dBA), NL — noise level, generated by the cooling system, DM — denominate multiplier (10).

Contents:

Vitali Crinitsin (vit@ixbt.com)
June 8, 2007

Write a comment below. No registration needed!