Skip to main content
en

Geometry preparation: what to remove and what to keep

Simplification takes the largest share of the time and produces the most silent errors. The criterion is the detail's effect on the flow or the heat path.

Most of the time spent on a CFD project goes not into solving but into preparing the geometry. The step looks mechanical; it is in fact a chain of engineering decisions, and wrong ones corrupt the result silently.

What can be removed

  • Cosmetic detail. Logos, lettering, decorative chamfers and fillets.
  • Internal structures closed to the flow. Volumes the fluid cannot reach.
  • Small, thermally insignificant fasteners — provided they are not on the heat path.
  • Very small fillets. If the flow does not separate there anyway, they can be sharpened.

What must not be removed

  • Sharp edges that fix separation. Adding a fillet moves the separation point. See the separation note.
  • Metal fittings that form thermal bridges. Small but conductive. See the thermal bridge note.
  • Sealing clearances. A gap set to zero removes the leakage flow and shows the distribution incorrectly.
  • Surface roughness and coatings. Not as geometry, but they must be represented as boundary conditions.

The criterion

A practical question for the decision: would the flow or the heat path be different without this detail? If you are unsure, running twice and measuring the difference is the cheapest way to settle the argument — and that measurement belongs in the report.

Watertightness

The flow volume has to be closed. Gaps between surfaces, overlaps and free edges either stop the mesh generation or silently produce a wrong volume — the model lets fluid leak into a place it should not.

Checking that the volume is closed before meshing prevents a significant share of convergence problems that are never otherwise understood. See the solver settings note.

Record

Which detail was removed and why should be written down. Months later, when the result is discussed, that list is what settles the discussion.

References

  1. Ferziger, J. H., Perić, M. & Street, R. L. (2020). Computational Methods for Fluid Dynamics, 4th ed. Springer. DOI: 10.1007/978-3-319-99693-6
  2. Celik, I. B., Ghia, U., Roache, P. J., Freitas, C. J., Coleman, H. & Raad, P. E. (2008). Procedure for estimation and reporting of uncertainty due to discretization in CFD applications. Journal of Fluids Engineering, 130(7), 078001. DOI: 10.1115/1.2960953

Let's discuss your project

Let us pin down together the system to be analysed and the question that needs answering.

Let's discuss your project