Boundary conditions: Asymmetric plane diffuser, k-omega SST, low-Re
Hey everyone,
I'm trying to simulate the flow through an asymmetric plane diffuser (ERCOFTAC Testcase) to compare the results from the simulation against the experimental results as well as results obtained from a simulation with Ansys CFX. I'm using k-omega SST as turbulence model together with a low Re-mesh in OF 2.1.1. However, simulation results aren't as expected, as the flow separation is largely overpredicted, so I was wondering if I got the boundary conditions right for this kind of simulation (especially for k, omega and nut). Choosing the settings gave me a hard time, maybe there's still something wrong with it. The contents of the files is posted below, any help is much appreciated. Thanks in advance, Krischan k: Code:
#include "include/initialConditions" Code:
#include "include/initialConditions" Code:
dimensions [0 2 -1 0 0 0 0]; Code:
#include "include/initialConditions" Code:
#include "include/initialConditions" Code:
flowVelocity (54.83 0 0); |
Never mind - boundary conditions are correct, after changing the solver parameters in fvSolution (tolereance and reltol), everything works as expected.
|
Hi Krischan,
glad you got your problem sorted! well done! I just had a quick question regarding your set up! I see you need to use a wall function for nut even though you are using a low Re approach for k-omega ... i was wondering whether you knew why exactly you need to do this? i have seen this is the way you need to go, but still cant figure out why you need to use a WF at all if you are using a low Re implementation for the turbulence model and you have specified k / omega as per guidelines for a low Re implementation as you had done? Nut is a calculated from k / omega as we know, so why do we need a BC / wall function for it (nut) anyway? many thanks for any help here! cheers Jonathan |
All times are GMT -4. The time now is 23:12. |