I want to calculate the flowfield of one nozzle for certain boundary conditions. In fact, after 30000 interations, the residual fo velocity, energy came to stay the same.But the residual is too big to be considered to be converged. How can I refine the grid to improve the result?
Re: Convergence problem
Did you set your vicous model properly? I wonder if it was set reasonbly. How about getting k & e value(or turbulence intensity) from papers dealing with simlimar geometry and b.c.?
For grid refinement, as I know, there are two ways. First, build your grid system again in GAMBIT. Second, there is an option "Adapt" in solver.
P.S. : In my experience, in case of turbulent flow problem, there were several cases solved by changing only momentum relaxation factor. Of course, setting a proper relaxation factor is totally depedent on trial and error.
|All times are GMT -4. The time now is 15:55.|