[jshjsh]

Hi!

Recently，I have noticed a simple problem about indoor airflow steady-state simulation. I found that monitored point or surface velocity magnitude is difficult to converge (become stable). The velocity magnitude always irregularly changes with the iteration, or it can be "slightly" stable for a long iteration time, though mass balance, heat balance or temperature monitor can achieve fast convergence. In many cases, my physical model of the room is very simple and the mesh quality is not low，the problem still exists. I don't know if this is a common problem in indoor air flow simulation? Or did I set something wrong? can anyone give me some advices?

*Unstructured grid
*Fluent Pressure-Based solver
*Steady-state
*Standard k-e model and Standard wall function
*One velocity-inlet and one outflow/pressure-outlet
*Others keep the default