Conference Year
2016
Keywords
louver fin, CFD, correlarion
Abstract
In recent years, small diameter (less than 5 mm) microfin copper tubes have emerged as a better alternative to large diameter (7 mm and above) copper tubes. The superior performance of these small diameter tubes comes from their lower drag, and increased heat transfer coefficient. This results in heat exchangers with lightweight, compact designs, lower material cost, and reduced refrigerant charge compared to their large diameter counterparts. One of the prerequisites before one can design any heat exchanger is the characterization of fin and tube performance in terms of heat transfer and pressure drop correlations. The present literature lacks these correlations, especially the ones that can be used over a wide range of parameter space, for heat exchangers using louver fins and tube outer diameters of 5 mm and below. In this work we address that by undertaking a Computational Fluid Dynamics (CFD) study wherein we parametrize the louver fin design and evaluate the performance of these designs by running multiple CFD simulations in parallel. The study focuses on 3 mm to 5 mm diameter tubes with varying design parameters such as transverse and longitudinal tube pitch, number of tube banks, number of louvers, air velocity. We utilize Design of Experiments (DOE) methodologies including two level full factorial and latin hypercube sampling to efficiently sample the design space. Using the data generated from more than 1000 simulations, air side heat transfer and pressure drop correlations are developed. Multiple linear regression analysis is performed to develop Colburn  j factor and Darcy f factor correlations. The new correlation reproduces 90% of the CFD data within 85% accuracy for heat transfer coefficient. For pressure drop the new correlation reproduces 90% of the CFD data within 90% accuracy. Work is underway to physically test the small diameter tube coils with louvered fins. The heat transfer and pressure drop data obtained through this physical testing will be used to further refine the aforementioned correlations.