I'm simulating a cold flow analysis (methane-air) on fluent, using species transport model. The model has around 5 millions meshes with tetrahedrons and swept bodies. I use "realizable k-epsilon" turbulence model with non-equilibrium wall function. Though the solution is converged, the results look nonphysical in the combustor. Any body who can give me some tips?
The problem could be using non-equilibrium wall function; in some computations on matter of ventilating flow in a volute, wall-functions applied to volute walls have given unrealistic results. Gianluca
Check your B.C's as well. If warranted , can you run a smaller version (less mesh) and analysis the results. Good luck.
I started from 3 millions of meshes and i am now at the edge of computer's capacity. So i don't know if the results are independent of mesh size .
Hello
I think you should:
1- recheck the boundary conditions
2- conduct a mesh sensitivity study (the results should not be influenced by the nature of the mesh)
3- increase the precision of the residual
4- check the wall law (check y + and the total thickness at the inflation option)
RENANE
Hello,
My geometry doesn't allow me to create inflation layer mesh. So the average y+ value is 40 for my case. That is why i use non-equilibrium wall treatment, which is recommended for geometries without inflation layer.
- Badges
- Science topic