[ivan-s]

The geometry required for our in-house product is large, on the order of several tens of meters. It would be described using a turbulent flow model. It also contains many smaller elements on the order of 10cm in diameter.

Recently I attempted to model it using specific well-known internationally renowned software, which is an FEM software and contains many many different kinds of physics (yet I will not name it here) and I was unsuccessful on the computer that I have on my desk (4GB RAM, 4 core CPU) to have it converge. I sent it to the headquarters of this company and it converged to a solution within a day. The issue was computational resources. In retrospect, the meshing was probably too fine, perhaps twenty to thirty cells across each smaller element.

Later we needed some modelling and we requested the modelling be performed by an outside company. This problem was routine for them. They chose meshing which had about ten elements across the diameter of each smaller element.

Now I would like to evaluate this company's software again, using finer mesh and compare the results with the results which were obtained by the outside company. The company has told me that they will offer another trial software. However, they have told me that they feel that their software is not ready for the kind of CFD modelling that we need.

Does this seem realistic to you? Which software would you recommend? I have some experience with OpenFOAM but the learning curve is a little steep compared to this other software. I think I can get OpenFOAM to work but it might take a bit longer.

Thank you for your help.