CFD- Erosion Mesh convergence

  • Last Post 14 February 2019
Matthew1996 posted this 14 February 2019

Doing a study on the effects of erosion parameters on the erosion rate of a sharp 90 degree pipe bend. Keeping my Y+ value greater than 30 I need to undertake a mesh convergence study to decide a mesh to use. Can I use the turbulent flow solution as validation for the mesh sensitivity or does it need to be done until the erosion converges. Problem is that the number of particles injected needs to converge for each mesh that i study which is proving troublesome and inconsistent. At the moment I am just using a 2D model.

Any help on the matter would be great.

Many thanks,


rwoolhou posted this 14 February 2019

You've got two parameters to consider:

As the mesh is refined the impact of a given particle will cause more erosion (the target facet is smaller).

As the number of particles is increased the erosion damage of a given particle is reduced (mass in the parcel is smaller).


As a result you may want a coarser mesh to ensure the erosion rate is sensible but also inject many parcels to allow the mass to spread more evenly. 

  • Liked by
  • Jackely