[alexandros80]

Dear fellows: I'm currently working in the simulation of a turbulence fluid contained in a mixing tank. I obtained the mesh in Gambit and successfully imported by using fluent3DMeshToFoam command, once checked all the aspects are reported as OK. When I run the simulation employing the MRFSimpleFoam solver, the bounding epsilon starts to increase and in around 10 iterations it's in the order of 1e-54 to 1e+55 and of course exceeds the double precision of the methods. I attach an image with the results showed in FOAM console. Could somebody please illustrate me about that issue, I obtained convergent solutions in Fluent for the same problem so I'm completely sure it's perfectly possible in OpenFOAM too. With best regards