[jhlee9622]

Hi guy.

Now, I want to simulate the indoor air flow by Fluent.

But I have difficulty in coverging. I have used the ICEM for meshing.
It is easy to convergence in small chamber which size was about
0.8 m x 0.4 m x 0.4 m. However, in a real size chamber which size was about 5 m x 5 m x 3 m, It is very hard to convergence in a Fluent.
The small chamber and real size chamber have same physics model.
Is there different way to convergence in real chamber case?
I think this convergence problem was caused by mesh problem.

The tetrahedral mesh was used, and it's size was about 0.1 m.
In addition, inlet and outlet has more fine mesh.

In this simulation, RNG k-e turbulence model was used and
second order upwind scheme was used in Momentum, turbulent kinetic energy, and tulent dissipation rate.

In residual, the continuity has almost equal in 1e-1. Is there any way
to solve this indoor air flow problem? Thank you

[gush]

There can be many different causes. Mesh structure, model selection etc.
I recommend to folllow validated cases from literature first.


Sent from my iPhone using CFD Online Forum mobile app

[Kummi]

Try adopting polyhedral meshing in Fluent...which will reduce the overall cell count by increasing the mesh accuracy as well as reduce the skewness.. its applicable for tetrahedral or hybrid meshes.. hope it helps you