Dear all, I have a question regarding grid generation for swirling flows. It seems impossible to build a grid which is aligned with the flow all the times. So my question is:

Is it a good choice to go for an unstructered tetra-grid and refine it where necessary. Or would it be better to build a hexa-grid, which can have high aspect ratios on boundaries but is harder to refine in regions of strong swirl. Regarding the modern codes (and node reconstruction etc), is it still true that hexa meshes are of greater accuracy? Maybe someone of you has gained experience dealing with this issue and can give some advice. Thanks and regards Stephen

We did a swirling flow simulation for jets with a single-block structured mesh. The results agreed pretty well with the experiments. There are few things which you might want to ponder about like the accuracy of the discretization available on structured and unstructured grids with the solver that you plan to employ.

http://www.cd-adapco.com/press_room/...estVspoly.html

I have not obtained good results with unstructered grids. I used two block structured grid. If you are working with an high swirl flow you'll have problems with secondary flows because they induce scales-separations and a crucial issue is also turbulence modelling approach. good luck

The article describes the advantages of polyhedral cells compared to tets and that the polys are even better than hexas for recirculating flows.

Block structured hexahedral grids are good as long an the flow is aligned with the grid and you are able to resolve your geometry with the grid, which can be quite difficult for industrial cases.

The article does not give any clear demonstration on the advantages. It just compares the results on the two different meshes and how can we assume that one result is better than the other.

1) The text also gives an explanation why polys can be even better than hexas:

Polyhedral cells are especially beneficial for handling recirculating flows. Tests have shown that, for example, in the cubic lid-driven cavity flow, many fewer polyhedra are needed to achieve a specified accuracy than even Cartesian hexahedra (which one would expect to be optimal for rectangular solution domains). This can be explained by the fact that, for a hexahedral cell, there are three optimal flow directions which lead to the maximum accuracy (normal to each of the three sets of parallel faces); for a polyhedron with 12 faces, there are six optimal directions which, together with the larger number of neighbors, leads to a more accurate solution with a lower cell count. A more detailed analysis of properties of various mesh types and some results from test cases are published in an article by Periç (2004)1.

Reference 1 M. Peric: Flow simulation using control volumes of arbitrary polyhedral shape, ERCOFTAC Bulletin, No. 62, September 2004.

2) Take one of the polyhedral codes (OpenFOAM, Star) and run a series of tests.