Learning OpenFOAM 14 – Step-by-Step CFD Guides

A path for learning OpenFOAM 14 (Foundation / CFD Direct) from scratch, one case at a time. Each guide builds a complete, reproducible case, solves it and checks the result with numbers: where an analytical solution exists, the comparison is quantitative, not just visual.

The first block uses a single physical problem, laminar flow between two parallel plates, which has the exact Poiseuille solution. Keeping the physics fixed makes it possible to change one thing at a time and see its effect: the mesh, the mesh generator, Python automation, parallel runs, boundary conditions.

1. Case 01 – laminar flow between parallel plates, validated against Poiseuille
2. Case 02 – mesh convergence study
3. Case 03 – importing a Gmsh mesh with gmshToFoam
4. Case 04 – parametric channel with the Gmsh Python API
5. Case 05 – parallel runs with MPI
6. Case 06 – sensitivity to the inlet boundary condition

The tools are OpenFOAM 14 with the incompressibleFluid solver via foamRun, blockMesh and Gmsh for meshing, ParaView 5.13.2 in client-server mode for visualization, Python 3 for parametric generation and OpenMPI for parallel runs.

These are the first six introductory guides.

Building a case, file by file

The second stage of the path changes the question: no longer how to run a ready-made case, but why it contains those files and how to write them yourself. It is an eight-part series, one layer of the case at a time.

1. Part 1 – case anatomy and dictionary language
2. Part 2 – building a mesh with blockMeshDict
3. Part 3 – fields, initial conditions and boundary conditions
4. Part 4 – controlDict and simulation control
5. Part 5 – fvSchemes, from equations to discretisation
6. Part 6 – fvSolution, linear solvers and solution algorithms
7. Part 7 – physical properties and models
8. Part 8 – reading and creating cases independently

Read in order, the eight parts lead from the structure of a case to reading an unfamiliar one and building a new one independently: fields and boundary conditions, execution control, discretisation schemes, linear solvers and algorithms, physical properties and models.

Working on a Linux server from Windows

Four operating guides accompany the path for anyone working on a Linux server from a Windows PC. The first covers the whole workflow: editing dictionaries in Notepad++, SFTP transfers with FileZilla, running over SSH from PowerShell with tmux, inspecting logs with klogg and visualizing results with ParaView connected to pvserver through an SSH tunnel. The PDF is included for download. The second covers what comes after launch: reading the log, residuals, function objects, pressure drop and flow rate, mass balance and a reproducible convergence assessment. The third uses ParaView to measure: profiles, section averages, flow rates, pressure drop, Poiseuille comparison and recirculation regions. The fourth establishes whether a result is reliable: quantity of interest, a family of three meshes, mass balance, comparison with Poiseuille, observed order, Richardson extrapolation and GCI, time-step study and physical validation.

• OpenFOAM on a Linux server, working from Windows: SSH, FileZilla, PowerShell, logs and ParaView client/server
• OpenFOAM: monitoring, troubleshooting and convergence assessment
• ParaView for OpenFOAM: quantitative analysis and scientific post-processing
• OpenFOAM: verification, validation and mesh independence