I'm trying to figure out if using the SW Flow heatsink idealization is applicable to my situation. I've currently got a good, working thermal sub-model of a fan/heatsink combination (meshes to about 4 million cells in order to capture all of the channel geometry effectively).

I now need to upscale this sub-model (aka. run a bunch of these fans/heatsinks in parallel) to make sure that my overall flow in the system doesn't get wonky (not as concerned about flow through the heatsink channels itself anymore). Problem is, this mesh will require like 30 million+ cells to achieve.

Here is where I figure the heatsink idealization might be useful at reducing my overall mesh by idealizing all the flow areas I'm not as concerned about at a small scale any more? I'm just interested in making sure that the right flow/pressure drop makes it's way through the heat sink channels according to the fan/system curves.

Two main questions I have are in regards to the pressure drop & thermal resistance tables required at the heatsink input page in the engineering database:

1) Pressure drop curve: I've got different runs of volumetric flow vs pressure drop across the sub-model, but the heatsink database wants velocity vs. pressure drop. My heatsink submodel doesn't have constand velocity, as the channel area changes width over the length of the heatsink (it's a radial heatsink). Is there a way to change the input curve to volumetric flow vs. pressure drop instead? If not, is there a good way to attempt to brute force a working curve in (hopefully the answer is not trial & error)?

2) Thermal resistance curve: Same issue here -- not sure how to tackle the required input velocity as my model has a changing fluid velocity field over the course of the heatsink.

Thanks.