False Time Step..
 

Hi CFD colleagues:

I am currently working on a indoor air flow problem, in which I have experienced difficulty in obtaining a "converged" solution while modeling the buoyant flow using the Boussinesq model.....

(1) I come across from some papers that using false time step in promoting convergence. What is its difference as compared to the linear relaxation factor in fact??? What is its merit against the linear relaxation factor while solving the matrix??? Can someone pls enlighten???

(2)**** Appreciate if someone could guide me on how to implement false time step in a CFD code.

(3) For your information, I have experienced convergence trouble while trying to obtain the steady-state buoyant flow solution by using one of the commercial codes available nowadays.. However, from what I have seen from some papers, the authors are managed to obtain a so-called 'steady' solution even in the room-flow case where its Rayleigh number is relatively high....Therefore, my question is: in CFD, what criterion to say that the flow is steady? Obtaining a non-varying result without caring about the error level of the associated residuals?? Any ideas??

Thanks in advance..