too many walls??
 

We have a little problem: Some of our geometries contains a hugh number of walls, eg. 3mio wall faces for 12 mio fluid cells. With some tricks we got the problem to run, but: the mass error grows, inlet and outlet are ok, but a few cells away from inlet all velocities are very small (~10e-6). When checking the results on the screen, the fluid seems flowing through the wall. Who knows this problem? Who knows who to fix it? Thanks to all,

Matze

Re: too many walls??
 

Did you merge vertices in the model? You can issue "Vmerge ALl" in PROSTAR, then write out geometry again.

Re: too many walls??
 

The vertices are merged. I had a correct running model. Then I deleted some cells in the fluid volume to have some walls inside. Only this few (~50000) deleted cells and extra walls cause the problem.

matze