I am looking to remove a considerable amount of power from an rf cavity structure. Some of the cooling will require water flow in the transverse direction where the path length is about 8 inches. There will be a total of four paths. The paths have two curves with a minimum radius of curvature of 20 mm. I barely have room to get in 7 mm diameter paths because of geometrical limitations. The paths will be machined into OFE copper with a CNC machine so the surface finish can be quite good, 128 microinches or better is feasible. It looks like 1.3 gpm is an adequate flow rate to get into the turbulent flow region. The calculated Reynolds No. is about 15,000. The problem is the calculated flow velocity is about 2.1 m/sec. From what I have read this seems to be a high velocity for a long channel life. The structure is expensive to fabricate and needs to have a lifetime of 20 years or more. Let's see what else? The minimum channel wall thickness is 0.5 mm in some places. So my question is, is a flow velocity in excess of 2 m/s too much?
I am doing a feasibility study to see if it is possible to operate the structure at high average power levels. I haven't done detailed thermal calculations yet but I will need to remove about 25 kW with these four channels.
Thanks, Jim