The equipment that FrostyTech uses when evaluating heatsinks and other cooling products has been developed to best gauge the various qualities of today's cooling solutions. To maintain the best possible correlation between heatsinks, all of our tests our synthetic in nature. From our experience, synthetic testing provides the most accurate and reproducible sets of results.

Our methodology currently focuses on the thermal performance of a socket 478 heatsink on a copper interface die, tested at 50W and 100W heat loads. Acoustically, we record the noise signature of a particular fan and heatsink combo and measure the sound levels.

Sound Level Measurements

To measure the level of sound produced by a heatsink in decibels, we use an Omega HHSL1 sound meter. The HHSL1 is highly accurate between a range of 35-130 dB. To measure the level of noise produced, a heatsink is placed on a Melamine foam surface and the meter positioned about 12" above it and off to one side. The highest average level of noise is recorded.

Frosty Acoustic Sampling Chamber:

To record the noise generated by a particular fan and heatsink combination, we use a special enclosure box lined with 2" of foam which is sealed from the outside environment during the recording process. The thick layer of foam helps to ensure that the sound we record is only that generated by the heatsink/fan inside. A standard computer power supply is located outside the enclosure to power the heatsinks' fan.

A monophonic microphone is positioned about 6" from the center of the test chamber so it is out of the direct air flow path generated by the fan. A solid state digital recorder is used record the sound picked up by the nondirectional mic inside the case. The digital recorder has a frequency range of approximately 500Hz to 3500Hz.

The purpose of the enclosure is to record an accurate representation of the noise generated by a heatsink and fan combo - separate from any other external computer noises like hard drives or power supplies.

FrostyTech Synthetic Temperature Test Platform:

The FrostyTech Synthetic Temperature Test Platform consists of an aluminum block heated by up to two 120V 250 Ohm precision resistors. With one resistor engaged, the test platform will output 50W of heat, with both engaged, 100W. The heat is transmitted to the test heatsink via a special copper interface die with an embedded type-K thermocouple.

The test temperatures are recorded from within the copper interface die which approximately replicate the surface area of socket 478 heatspreader covered processor. Since the interface die has the approximate surface area of a heatspreader covered processor, the flow of heat to the base of the heatsink is accurately mimicked.

An Omega HH501DK Type-K digital thermometer is used to record the temperature from the interface die once everything has thermally stabilized. The test is conducted at 50W, and again at 100W for the interface die. All heatsinks are tested using a good quality ceramic-based thermal compound and all original thermal interface materials are removed.

A Note about 1U Reviews

A 1U server case is only 1.75" thick, leaving effectively 30mm or so above the processor core for the entire thermal solution. Noise isn't much of a concern in this environment, but rather sheer cooling performance. Servers are expected to be up 99.9% of the time, and system problems due to an overheating processor simply won't fly.

This brings us to the problem with how to test 1U heatsinks accurately. Since the operating environment of every 1U heatsink is height limited, it would seem ideal to test 1U heatsinks as such. In this example, all 1U heatsinks would be tested with a plate directly above the top of the cooler so that the total distance from heatsink base to the base of this cover plate is 30mm.

Unfortunately this is where the problems arise. Every type of heatsink is a product of the environment to some degree, and in the case of our hard working 1U heatsinks, intake fans, drafting, orientation of heatsink relative to airflow, and countless other factors can influence how the heatsink will perform within each specific 1U chassis.

Therefore unless otherwise noted, we will be reviewing all active heatsinks "removed from the system." The results of our tests on 1U heatsinks will not be under height-limited system conditions. Rather, each heatsink will be tested as if it is removed from the system, yielding more accurate comparisons to the other heatsinks being reviewed. The distinctions are slight, but very important for those of you dealing with decisions on what 1U cooling solutions to consider. Particular systems follow similar guidelines and 1U heatsinks can (and should) be tested for each case environment.

Passive 1U cooling solutions will be tested with the 30mm height restriction in place and use a common source and direction of draft air. The exact conditions of the test environment will be described for those particular 1U heatsink reviews.

Now that we've explained a little about how we test the heatsinks that cross our path, lets move forward and take a closer look at test results for today's heatsink.