dragOverSphere-PyFoam

This is a quick project to determine the drag over a sphere using OpenFOAM and the parameter variation capabililties of the PyFoam Python library.

Prerequisites

• OpenFOAM 9
• PyFoam

Background

The drag coefficient of a sphere varies with Reynolds number, and this data is tabulated and correlated below (Schlichting, 1955; Morrison, 2013):

Methodology

• Mesher: blockMesh
• Solver: simpleFoam
• Turbulnce model: laminar

The 0/ and system/ directories from the simpleFoam/motorBike tutorial were adapted for the simulations, as the tutorial was similar to a wind tunnel.

(Preliminary) Results

The simulated drag coefficient aligns well with the data correlation throughout the Stokes region ($Re<1$) until $Re=100$, and a slight deviation is observed for higher Reynolds numbers. The drop in drag coefficient at $Re\approx200,000$, known as the “drag crisis” or “Eiffel paradox” phenomenon is not displayed in the simulated results.

The drag crisis is caused by a transition from a laminar boundary layer flow to a turbulent boundary layer flow, which results in a change from periodic vortex shedding to randomized vortex shedding. Since no turbulence model is used for these results, it is natural that this is not observed.

To-do

• Plot automation

• [] Mesh convergence study