Waterblock (wb) “C/W”s, actually the combined C/W of the wb and the TIM joint, are generally held to be a characterization of wb cooling effectiveness. This notion, while partially correct, fails to address the pronounced influence of the coolant’s flow rate on the apparent wb “C/W”. Additionally when the TIM joint C/W is subtracted (which is a constant for a given mounting), the influence of the flow rate is even more pronounced.
While it is tempting to compare wbs at equal flow rates, this masks the fact the ‘equal’ flow rates generally produce different pressure drops across the different wbs. From a pumping perspective, which is the reality that all WCing systems deal with, the basis of comparison would more appropriately be the “C/W” vs. Head Loss
Some examples are shown below utilizing two rather different wbs for comparison.
The first is the conventional plot of “C/W” vs. Flow Rate and the second the Head Loss vs. Flow Rate. Combining the two graphs to plot “C/W” vs. Head Loss yields a more pronounced ‘apparent’ difference:
This graph can most easily be understood by hypothesizing that one has a pump in a system with an ‘available’ head of 0.5mH2O (the balance of the pump’s capability being utilized to overcome the other system resistances, rad, tubing, etc.). Then reading the graph vertically one can see that the difference between these wbs in this system would be about 0.02 “C/W”, or 2°C at 100W (nominal per Radiate). Note that while the determination of the ‘available head’ is rather complex due to the interactive resistance changes with the flowrate, such can be ‘ballparked’ sufficiently for our purposes.
In summary; the first two graphs indicate the relative performance of the two principal wb characteristics, while the third graph shows the combined effect of these characteristics. Each of these graphs is useful depending on the wb characteristic being evaluated.