| This investigation assesses the utility of fine filament screen-laminate enhanced surfaces as high heat flux boiling surfaces and effective bubble nucleation sites. Experiments were conducted on vertically oriented, multi-layer laminates in distilled water at pressures of 0.2 atm--1.0 atm. The first phase of this work focuses on saturated pool boiling, where the performance of twelve different copper-filament surfaces, having pore hydraulic diameters ranging from 14 to 172 microns, is documented. Experimental results show that boiling performance is a strong function of screen laminate geometry. Enhancement of up to 27 times that of an unenhanced surface was obtained, and dimensional analysis and multi-parameter regression are used to develop a heat transfer correlation that relates the boiling heat transfer coefficient to the lamination geometry and system pressure.;The second phase of this work documents the performance of these structured-porous screen laminate enhancements within the subcooled, vertical, up-flow boiling environment. Experiments with distilled water at 0.2atm pressure assess the effect of: enhancement geometry, subcooling, and Reynolds number; and their relationship to each other. The subcooling ranges from 2K to 35K; and, channel Reynolds number is varied from 2,000 to 20,000. Boiling performance is documented for ten different surfaces having pore hydraulic diameters ranging from 39mum to 105mum, and surface area enhancement ratios ranging from 5 to 37. Heat flux of up to 453W/cm2 is achieved at 35K sub-cooling at a Reynolds number of 6,000, which represents a 3.5-fold increase in Critical Heat Flux (CHF) over that of saturated pool boiling on the same surface. It is found that CHF enhancement due to sub-cooling and Reynolds number is intrinsically linked to the surface area enhancement ratio, which has an optimum that depends on the degree of subcooling. High speed video imagery (1200fps) and long range microscopy are used to document bubble dynamics. Boiling mechanisms inherent to sub-cooling, enhanced surface geometry, and CHF are discussed, and a mechanistic model for CHF is developed, which predicts CHF with a mean absolute error of 10.3%. |
