Evaluation of Pressure, Surface Characteristics, and Fluid Properties Effect on Pool Boiling Heat Transfer Over Plain and External Micro-Finned Cylindrical Surfaces

Abstract

Abstract Pool boiling is extensively used in high- and low-temperature heat exchangers as it results in a high heat transfer coefficient compared to natural and single-phase forced convection. Pool boiling experimental study conducted over a plain cylindrical surface (PS) and four external micro-finned cylindrical surfaces (MFCSs), with R123 and R141b at different pressures in the heat flux range 20–100 kW/m2, is presented in this paper. The objective of the present study is to explore the effect of pressure, surface characteristics, and fluid properties on pool boiling heat transfer over plain and micro-finned cylindrical surfaces. The boiling performance improved at a higher pressure, irrespective of the working fluid used for all the test surfaces. It was found that, with the rise in pressure, the boiling heat transfer coefficient (BHTC) for the MFCSs increases at a higher rate than the PS. In comparison with PS, the average rise in the BHTC with pressure, for the MFCS-1, MFCS-2, MFCS-3, and MFCS-4 with R123 were 69.3% to 84.3%, 3.3% to 9.9%, 16.9% to 22.4%, and 29.4% to 40.2%, respectively. The higher BHTC over micro-finned cylindrical surfaces results due to more nucleation site results from lower surface wettability and micro-finned geometry. It was observed that the pool boiling over the plain surface with R123 results in higher BHTC compared with R141b at all tested pressures, whereas the pool boiling characteristics over MFCSs varied based on the combined effect of micro-finned surface geometry, surface wettability, heat flux, pressure and fluid properties. The bubble departure diameters over all the surfaces were measured at 30 W, 60 W, and 90 W at different pressures, and a new model of bubble departure diameter was proposed based on dimensionless terms. The total mean absolute error (MAE) of the proposed bubble departure diameter model was about 6.79% for the whole range of data points.