NUMERICAL PREDICTION OF PARTICULATE FLOW OVER A BACKWARD-FACING STEP FOLLOWED BY A FILTER MEDIUM

Flow in air filter housings often is characterized by separation upstream of the filter. The effect of the separation on the motion of particles and their distribution at the filter is important to filter performance. The current research investigates these effects by applying CFD modeling to turbulent particulate flows over a backward-facing step followed by a porous medium representing a filter. The two-dimensional step flow was selected as it is an archetype for separated flow with many studies in the literature. The flow examined has a step expansion ratio of 1:2, with an entrance length of 30 step heights to the step followed by a length of 60 step heights. Computations were performed at step Reynolds numbers of 6550 and 10,000 for the step without a porous medium and with the medium placed 4.25 and 6.75 step heights downstream of the step. The mesh was developed in ICEM CFD and modeling was done using the Fluent commercial CFD package. The carrier phase turbulence was modeled using the RNG k-epsilon model. The particles were modeled using the discrete phase model with dispersion modeled using stochastic tracking. The boundary conditions are uniform velocity at the inlet, no-slip at the walls, porous jump at the porous medium, and outflow at the outlet. The particle boundary condition is "reflect" at the walls and "trap" at the filter. The numerical results for the no filter case matched experimental results for recirculation zone length and velocity profiles at 3.75 and 6.25 step heights well. The computed velocity profiles at 3.75 step heights do not match experimental profiles for the filter at 4.25 step heights so well, though the results show a profound effect on the recirculation zone length, matching the experiments. Differences are attributed to different velocity profiles at the step. With the medium 6.75 step heights downstream, the effect on the recirculation zone is negligible, again matching experimental results. The discrete phase model tracks injected particles and provides results which are qualitatively similar to the literature. It is observed that particles with lower Stokes number, and thus lower momentum, tend to follow the flow and enter the recirculation zone while particles with higher Stokes number tend to move directly to the porous medium. When the filter is moved downstream to 6.75 step heights, the increased length of the recirculation zone results in more particles entering the recirculation zone. Results for monodispersed and polydispersed particles agree.