[DmitryS]

Hello dear colleagues!

There is wall with slip condition and two fluid flow on the opposite sides of this wall. Need to equate flow velocities on opposite sides of the wall. I guess for this purpose need make UDF. Could you help me with it?

Thanks!

[pakk]

You should completely specify your problem...

You say that you have slip conditions on both sides of the wall, but what do you exactly mean by that? Free slip? Low pressure boundary slip?

And the bigger problem: what do you mean by "equate flow velocities on opposite of the wall"? If you mean what I think (set the flows to the same value), you have conflicting requirements wrt to slip.

Suppose that without your udf, you have a flow on the left side of 1 m/s, and on the right side of 10 m/s, calculated by Fluent using the slip condition that you supplied. What should your udf do in that case? Make them both 1 m/s? Make them both 10 m/s? Make them both 5.5 m/s? Whatever the udf does, the slip condition on at least one side is no longer correct.

Before you can make a udf (or use a different solution), you should know exactly what you want. Your question now is like asking if somebody can help you to draw red lines, using green ink.

[DmitryS]

Description of the problem.

Need to calculate two-phase flow in horizontal pipe due to pressure drop. One approach is to use VOF model. But this solution is very expensive from the point of view of computing resources.

At the same time a flow regime in the task leads to the flat (waveless) phase boundary. It's meant what possible to calculate separate flows with some interaction. Need to define and implement this interaction.

I guess, in a first approximation, can assume that flows interact with each other through the movable wall. It means that the wall shear stress should be equal on both sides of the wall.

Isn't it?

[pakk]

Originally Posted by DmitryS

Description of the problem.

Need to calculate two-phase flow in horizontal pipe due to pressure drop. One approach is to use VOF model. But this solution is very expensive from the point of view of computing resources.

At the same time a flow regime in the task leads to the flat (waveless) phase boundary. It's meant what possible to calculate separate flows with some interaction. Need to define and implement this interaction.

I guess, in a first approximation, can assume that flows interact with each other through the movable wall. It means that the wall shear stress should be equal on both sides of the wall.

Isn't it?

Now you have given three possible boundary conditions:
1. Use slip conditions on each side
2. Set velocities equal on both sides of the wall
3. Set shear stresses equal on both sides of the wall

You started by asking a question on how to implement things in a UDF, but it appears like you should take a step back, look at the problem, and think what you really want to know. If you skip this step, you will end up spending a lot of time in implementing something that you don't need.

I am not going to give you the answer to the question which boundary condition you should use. Not only because that is your work, and I am not going to do your work, but also because I don't know the answer.

[DmitryS]

Originally Posted by pakk

Now you have given three possible boundary conditions:
1. Use slip conditions on each side
2. Set velocities equal on both sides of the wall
3. Set shear stresses equal on both sides of the wall

1) I think first one is not correct for described task. I've understood it.

2) Second case consist in separate calculation of two parts of pipe (wet and dry). BC on phase boundary should be set velocity. And profile of velocity from wet side should be set to wet side and vice versa. These actions should be on each iteration. After a few iterations velocity on different sides should be the same.

3.1) BC on phase boundary is movable wall. Vary the speed of wall until the shear stresses will not be equal on each sides. But it is not a simple movable wall - cell's faces located on phase boundary should has own speed.

3.2) BC on phase boundary is stationary wall. Used approach like in case 2, but regarding to the shear stresses.