OpenFOAM Multiphase Flow Analysis
Theory: tracking interfaces with volume fraction
VOF transports a phase volume fraction, using it to obtain mixture density and viscosity and couple flow to the interface. It can represent breakup, but droplets or films smaller than the mesh are unresolved.
V_i is cell volume and V_l liquid volume. Track deviation from the initial amount in a closed domain; include integrated inlet/outlet volume flux for an open one. Co here is a one-dimensional estimate, not necessarily the solver's face-flux definition.
Identify the distribution first
OpenCFD's interFoam and Foundation releases differ in dictionaries and invocation. The Foundation v13 dam-break example uses foamRun with the incompressibleVoF module. Start from the damBreak tutorial shipped with your installed release rather than copying another distribution's dictionaries.
Worked example: checking time-step scales
These are time-scale calculations from stated ideal models, not OpenFOAM run results or stability guarantees. Separately check experiment agreement, volume conservation, mesh and time-step sensitivity.
A 0.292 m water column and Co = 0.5
| Cell width | Advective step estimate |
|---|---|
| 5 mm | 7.39×10⁻⁴ s |
| 2 mm | 2.95×10⁻⁴ s |
With g = 9.81 m/s², the speed is an ideal dry-bed shallow-water (Ritter) scale, not a front speed measured from an OpenFOAM run. Do not apply it unchanged after wall impact or during strongly vertical initial motion. Monitor actual maximum and interface Co in the solver log.
A surface-tension time scale
| Cell width | Time scale with this coefficient |
|---|---|
| 1 mm | 1.05×10⁻³ s |
| 0.1 mm | 3.33×10⁻⁵ s |
| 0.01 mm | 1.05×10⁻⁶ s |
Assume water at 1,000 kg/m³, air at 1.2 kg/m³ and surface tension 0.072 N/m. This educational scale uses a discretisation-dependent coefficient; it is not a guaranteed stability limit for every VOF scheme. Reducing cell width tenfold reduces it by about 31.6. A maxCo setting alone may not control capillary constraints: inspect the solver implementation and test time-step sensitivity.
Verification workflow
- Run the shipped damBreak unchanged and record release and dictionaries.
- Record alpha extrema, liquid volume and inlet/outlet balance.
- Refine mesh and time step separately; compare front position and pressure.
- Match geometry, initial liquid volume, measurement location and time when comparing experiments.
Common mistakes
| Check | Action |
|---|---|
| Liquid volume decreases | Check boundary flux and alpha transport; review interface Co and step size. |
| Flow around a stationary drop | Possible spurious currents: verify curvature, mesh and capillary balance. |
| A film disappears | Compare film thickness to cell width and refine locally if needed. |
Related: OpenFOAM meshing / OpenFOAM post-processing
Frequently asked questions
What does alpha mean in VOF?
It is the phase volume fraction in a cell: water alpha of 1 means water, 0 means the other phase, and intermediate values identify interface cells. It is a cell average, not a physical interface thickness.
Is maxCo enough to choose the time step?
No. Check the interface Courant limit, surface tension and interface-advection settings too. Names vary by release; follow its tutorials and verify accuracy with a halved-time-step comparison.
What should a dam-break test verify?
Check volume conservation, bounded alpha, front position versus time and pressure or impact. The scale 2√(gH) belongs to an ideal dry-bed shallow-water model; it is not a pass/fail value for a VOF case with walls, obstacles or viscosity.
References and scope of verification
- OpenFOAM Foundation v13 — Dam break
- Brackbill, Kothe & Zemach — Continuum surface force method (1992)
Numbers are educational reference calculations under the stated assumptions. Sources document theory and setup, not proof that these numerical values were reproduced in each package. Engineering decisions require validation for the actual conditions.
Related Topics
Experience the theory firsthand with the interactive simulator for this field
All Simulators