free flow problem
 

Hi everyone!

I've got a simple (?) 3d-question: I'm going to examine the wind field over a mountain ridge... But I'm not sure about what B.C.s I should choose. Should I expect convergence problems if I define the surrounding (west, east, north and south) boundaries as v_inlets? I was thinking about v_inlets to have the ability to vary the direction of the wind with each time step. For the top boundary ("heaven") I was thinking about a symmetry, but a collegue told me that this would cause big convergence problems.

Can anyone of you give me a hint about how to choose my B.C.s?

Thanks, Thomas

Re: free flow problem
 

Isn't here really nobody out there who can help me?? Please!!!

Re: free flow problem
 

So because of the lack of comments to my original posting, I am going to test it with different B.C.s... I'm also going to comment them here, so if you've got suggestions, feel free to leave a posting ;-)

My first attempt was using a pressure-f-f for the surrounding boundaries, unfortunately my domain is about 6x6x10km (LxWxH) and I need fine resolution. Therefore p-f-f seems to be not well suited for this (I already have about 1 million elements and I am pretty limited what concerns computational resources) because of the FAR field thing. Although, the p-f-f seems not to be compatible with my multiphase mixture model, because it requires the coupled solver (which is not supported my m-m-m, according to the fluent user manual).

So the 4-times-v_inlet idea came into my mind... so that's the point where I got stuck right now...

Hints, suggestions... are welcome.

thanks, thomas

Re: free flow problem
 

I'd neglect time varying velocity conditions at the boundary unless you have a lot of time to wait for the model to run. Apply velocity magnitude at 3 of the boundaries far from the area of interest, closer if you know the profile and can map to the boundary. Make sure the flux across these boundaries sums to a reasonable level for the remaining boundary. Use an outlet or pressure boundary at the downstream boundary. If there are convergence problems due to recirculation, consider decreasing the open boundary area.