Gas plume or flare simulation

ahourri

Hello: My probleme is: I have a domaine with dimensions 27x27x27 m. In the middle of the domaine I have a stack of 3.5m height and 1 inch diameter. A sonic compressible gas is coming out from the stack vertically. I have also a wind with a constant velocity (9 m/s)that's blowing trought the domaine through the x direction. My boundary conditions are: face left is set as mass flow rate for the wind because for a compressible model Fluent doesn't accept velocity inlet as a boundary condition, mass flow inlet is set for the gas inlet to the domaine, all the other faces are set as pressure outlets except the ground of course. My model is turbulent (RNG-ke. After I initialize my domaine, I patch the whole domaine with an x=9 m/s velocity. Is-it right to do this? My domaine has 1 200 000 cells elements.the mesh is unstructed. When I compare my steady state solution with a solution from Phoenics, the gas concentration contours from my case are very much longer (300% longer) along the x direction than that from Phoenics. For the phoenics application they have only 19 000 structured cells. What do you think about this? Which one do you think is Should I patch the whole domaine with a 9 m/s x velocity? I did the simulation without patching, my result is more or less similar. Do you have any ideas on simulating gas plume or flare situation in windy environment. Am I doing right or something must be wrong in my setting. Please help me.