Boundary layer is a complex phenomenon which is very thin and the solution gradients are very high. To capture this phenomenon correctly, we could use a fine mesh throughout but would be a costly affair. Instead adding up prism layers near the wall would be time and cost effective giving good convergence and fairly accurate results.

Tri/Hex could resolve it but not as efficient. It requires minimization of equiangle skewness and you could refer for the values for each respective case from the software manuals

On Mesh side:The main advantage of prism layer is that they allow high aspect-ratio cells without incurring an excessive stream-wise resolution, which is not the case with other cell types.

On Computation side: PL reduce numerical diffusion near the wall by aligning the sub surface (layer connecting core mesh and PL) with the flow. Again, PL has an advantage over other cell types in doing so.

- Source : STAR-CCM+ 8.04 Documentation

Boundary layer is a complex phenomenon which is very thin and the solution gradients are very high. To capture this phenomenon correctly, we could use a fine mesh throughout but would be a costly affair. Instead adding up prism layers near the wall would be time and cost effective giving good convergence and fairly accurate results.

Tri/Hex could resolve it but not as efficient. It requires minimization of equiangle skewness and you could refer for the values for each respective case from the software manuals

Hi vaibhhav,

Before going to deal with your question i would like to give a brief details about the Boundary layer.

Well to capture the boundary layer effect in usually tetra & hybrid mesh used along with prism mesh while in hexa mesh to do this you have to know about the flow physics before you going to deal with the problem because in hexa mesh for capturing boundary layer you have to provide more number of grids up to boundary layer thickness for capture the flow properly as well as this phenomena also depend on which kind of wall treatment you are going to use for your problem and a particular wall treatment have a desired level of y+ criteria so from there you ill get the first cell height and this is a key factor to decide the grid density.

So this is all about the basic details of how to deal with boundary layer.

Now lets come at your question overall in any kind of mesh whether its hexa , tetra or hybrid all are able to solve with exact accuracy but in case of tetra for achieving that much accuracy you need more computation time while in hexa you achieve with a bit less computation time perhaps to generate a hexa mesh with proper aspect ratio nd quality is bit crucial job for complex geometry.

During my academic work i had same confusion in my mind because in academic research mostly hexa mesh are preferable so mostly new learners thought that tetra mesh not provide a accurate result but the actual thing is, In earlier days we didnt have a high computation power in academic research level thats the reason hexa mesh are preferable but as now after 2010 we have high computation power so its easy to deal with terta or hybrid mesh & in my research work i was compared result with hexa and tetra mesh and i found that both results was almost same precisely an error around 5% in tetra mesh results which are under acceptable range that's the reason all most all industry deal with tetra mesh because they are mostly deal with complex geometries and they not prefer to consume time for creating hexa mesh.

I hope this ill help you to resolve your confusion.

regards

shrish shukla