HI,Every one.I have a question, As I simulate a river(open channel flow) with Fluent,I don't know how to specify the Boundary condition of the top face(the free-surface of the river exposed to the air). I think the the surface is not wall and symmetry face.Must be use Multiphase Flows model here?Now I use it just with single turbulent flow model.

You have two approach.

1. Detail Simulation : If you want to include the effect of the change of free surface, you should use two phase model. For your case, VOF is the best one.

2. Approximation : For most case, the variation of the shape of the free surface is fairly weak. That is to say, they can approximate the free surface as given shape, in grneral, straight face for 3D(or line for 2D). In this case, shear force at the free surface is negligible, so that symmetry condition is very good approximation(many CFD experts does do).

Sincerely, Jinwook

thank you for your help. I want to simulate Approximate surface such as the latter item you have said. I have test the simulation on a rectangular open channel, I specify the center face as symmetry face so as to miror the right part. The left and bottom wree specified as walls. the top face was specified as symmetry condition. The front- and back- face are velocity inlet and outflow conditions respectively. Using RSM model, I simulate the secondary current. the result is fine,but it was the secondary flow like the flow in a duct(confined by walls around flow, not the open channel flow).I think the two symmetry face at center and top make the simulated domain as quarter of a duct. How can I specify the top symmetry condition like a real "free surface" but not a face to miror the bottom. How can I do? thenks a lot.

later, I specify the top face as pressure inlet condition, and make its velocity as 0 for all directions, but the result is wrong.

Jinwook, I think if I need to add the gravity on the fluid in the model which like a quarter of duct, it will like a open channel with approximate free surface?? if yes ,how to add gravity.

I could not exactly understand your problem. But I can say that symmetry BC is BEST condition for the free surface unless you try to two-phase flow(VOF).

Pressure inlet never be used for your problem. Then you may have flow-in or flow-out through the surface. Is that your situation ? I don't think so.

Your problem is not so related to the gravity. If you want to include the gravity effect, you can use the gravity opin in the 'DEFINE --> OPERATING CONDITION'. But I am sure that gravity option can not clear your problem.

Sincerely, Jinwook

Dear Jinwook: It's so kind of you, my project is just to simulate the secondary flow of open channel. could you give me some more information about it? Please give me some advise how to specify the BC, I use the RSM model. I have use the symmary condition as free surface, but the resilt is like in a quarter of duct but not in half of open channel. Actually, There is a "g*sin(a)" in the streamwise momentum equation of the mathematical model(a is the slope angle of channel), which is not appear in basic Reynolds-averaged Navier-Stokes (RANS) equations. Must I add the "g*sin(a)" in the source term? how to add it? if not, How to operate it in the simulation.

You may use wall with zero shear stress at top free surface as the Boundary condition. Hope this works.