Numerical and Experimental Investigation of Flow Through Water Filters Using Porous Media Approach

Abstract

The aim of the present study is to estimate flow rate through water filters used in refrigerators at various inlet pressures through simulation. Regions of high pressure are identified and design modifications are also suggested to reduce the pressure drop. The baseline design of the water filter was tested in the laboratory and was used to validate the simulation results. As modelling the actual carbon block in simulation is complicated, a porous media approach was used to model the pressure drop through the carbon block of the water filter. Darcy-Forchheimer coefficients are calculated using the pressure drop vs. flow rate characteristics of the carbon block obtained from the manufacturer of the carbon block. For a range of various inlet pressures as per requirement, flow rate and total pressure drop are calculated. The mesh independence study was carried out and mesh size of ~ 2.4 million tetrahedral cells was found to be optimum and used for the final simulations. The commercial code of Ansys-Fluent 20R1 is used for meshing and simulating the problem. Flow rate estimated from simulation for baseline design is within 7% of test results. Regions of high pressure drop are identified - about 11% of the pressure drop is observed due to sharp turns in the baseline design. Suitable design modifications are suggested based on simulation results and simulations on the modified designs showed upto ~ 8% increase in flow rate through the water filter for the same inlet pressure. This study demonstrates the successful use of a porous media approach to estimate flow rate through water filters which matches test results with 7% of difference. Thereby modifying the filter design to improve the flow rate through the water filter is fully based on simulation results.