[kennymhx]

Hi all,

In my case, I want a turbulent inlet boundary condition for my simulation. I tried the turbulentinletPatch in Openfoam 7, and I tried the FilterNoise method. But they can not give me a right solution for my pressure field. I found the DFSEM (Divergence-free synthetic eddy method) could be used as the boundary condition at the inlet for LES. But it is just available in OpenFOAM v1906 (the one downloaded from www.openfoam.com).

It seems like this BC is not provided in OpenFOAM 7 downloaded from www.openfoam.org. So how can I get the DFSEM boundary condition for OF 7?

Thanks

[HPE]

Hi,

- Would you mind to tell us what 'FilterNoise' method is (just curiosity?
- Have you ever compiled a BC? Basically, copy-paste the OF.com BC into OF7, and recompilation of the BCs only should be enough.
- If there is no particular reason for using OF7, might consider to switch to OF.com version since they may update the content of this BC in the future, and then you would need to do the same transfer procedure again ( I think this option is the easiest).

[cryabroad]

Hi kennymhx,

I've had similar questions as yours. Note that .org versions of FOAM is a littble bit different from .com ones. If I remember correctly, their definition and use of random number generators are different, so it would be hard to directly compile the DFSEM boundary condition from .com FOAM in .org FOAM. I end up using the inflowGenerator from Rostock, search in the forum or google and you will find it. I've compiled it in OF-4.x with minor effort, and it works well.

A little bit more of the performance of turbulent boundary condtions, I think the DFSEM one gives better pressure behaviours at the inlet, because it is divergance free. The inflowGenerator from Rostock is not strictly divergence free (I'm not 100% sure on this). By the way, depending on what you are looking at, the turbulentInlet in FOAM should be good too. If you are considering an unconfined flow, the turbulence should be created by itself (by mixing for example) instead of being transported from the inlet, in this case turbulentInlet works fine. But if you are considering for example a confined flow (flow in a tube or something similar), a correct inflow turbulence is desired.

[HPE]

I kindly disagree. Never use the turbulentInlet boundary condition for inlet turbulence. It is just a bad naming, I am afraid. The turbulentInlet condition has in reality nothing to do with resolved turbulent fluctuations.