Methodology – Joe

I received a number of requests to do some rigorous objective testing of watercooling radiators. After doing some testing on approaches, I decided to use the following methodology:

Bill Adams’ seminal work on Radiator Heat Dissipation Testing developed the basic approach which can be summarized in the following equation:

Heat Extracted = Btu/hr = 499 x flow rate in gpm x (Tdiff)

Where Tdiff = Inlet Temperature to the Radiator – Outlet Temperature

The equation is the easy part – making it work is a bit more difficult.

I finally settled on the following approach:

In order to make it an apples-to-apples comparison, the only variable in the setup is the temperature difference between the inlet and outlet water temps. To keep airflow through the tested radiators constant, each radiator is measured for airflow using an Airflow Chamber as outlined HERE (Earlier I measured a number of radiators in the article Radiator System Resistance).

For these tests, each radiator is calibrated with a fan at 30 and 45 cfm. This setting is then used in the test rig so that airflow through the radiator is the same for each one. The Neslab heater keeps water temps in the reservoir (and radiator inlet temp) at a constant temp to within 0.1ºC.

As for waterblock tests, water flow is measured by a King Flowmeter so that flowrates are equal for each tested radiator; a valve is used to adjust flow to the desired setting.

Not shown in the diagram are pressure taps to measure the radiator’s pressure drop; these are located at the same points as the temperature taps.

This testing allows an objective, relative ranking of radiators. With this data, consumers may determine the cooling capacity of each radiator and its pressure drop relative to others tested. Pressure drop is an important variable in watercooling systems when coupled with a waterblock’s pressure drop and a waterpump’s head rating.

One thing that should be mentioned is that size matters. Even though it might be that a radiator that is twice as large as another may rank lower in relative performance to one half its size, the larger radiator will dissipate more heat than the smaller by virtue of its size, albeit perhaps less efficiently. In cooling, it’s surface area that counts.

Stay tuned for results!