UC Berkeley

The goal of this project is to design, develop, and validate a customized high performance computing (HPC) numerical framework to study rotary bell atomization, a key manufacturing process used in industry to spray and apply paints. Optimizing the atomization process for higher paint flow rates would allow industry to accelerate the assembly line, reduce energy consumption, and increase throughput while maintaining coat thickness and high quality appearance. Accurately modeling this process requires a combination of advanced HPC and computational fluid dynamics to significantly advance the state of the art. The developed numerical framework will be used to study the rotary bell atomization process as a function of paint fluid properties, film thickness, and bell geometry at scales and parameters relevant to industrial conditions.The combined experimental findings and insight from 2D and 3D numerical modeling enabled progress to be made in the understanding of the key physics driving the rotary bell atomization process. Accomplishments include improved understanding of: (i) the relative strengths of inertial, viscous, centrifugal, and surface tension forces in the overall atomization process; (ii) the source and role of non-uniform film thicknesses on the inside surface of the cup; (iii) dependence of sheeting behavior on film thicknesses at the edge of the cup; and (iv) formation of tendrils and their subsequent breakup into droplets.