[sinatahmooresi]

Hello FOAMers,


I am currently working on generating a fully developed turbulent velocity profile at the beginning of a jet nozzle to be discharged into a tank using LES simulation in OpenFOAM. To achieve this, I have tried incorporating a precursor simulation of pipe flow before the tank inlet. Among the various methods available, I have utilized the mapped inlet for velocity. However, I have noticed that the resulting velocity profile appears smooth and averaged, rather than the expected fluctuating instantaneous profile typical of LES simulations. I would appreciate any advice or suggestions on this matter. Attached are boundary, U, p, k, nut dictionaries and the resulted velocity profile. The simulation is conducted in 2D and I built the model for the most part based on the pitzDailyMapped tutorial. simulation is non-dimensional and thus the U=(1 0 0). The number of generated elements in horizontal and vertical directions are 2000 and 200 respectively. The non-dimensional length is 30 and the mapping source face is located at x=18 which is after the entrance length required for the flow to be fully developed (Re=3000).



Thank you,
Sina

[sinatahmooresi]

Greetings all,


I'm seeking advice from anyone who might have some suggestions. I've attempted to use both cyclic and mapped boundary conditions, as well as the perturbU utility for initialization, but have not achieved any success. It appears that the primary issue is that the turbulence does not transition from the wall to the center. Additionally, yplus remains below 5 to ensure proper wall capture.