CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Channel flow- How to apply "Translational Periodicity" (

Giorgitseli March 5, 2011 11:05

Channel flow- How to apply "Translational Periodicity"
Description of my problem:
The geometry of the project I am currently working on includes an open vegetated rectangular channel, 19,5 m long. However, the number of volumes created in CFX-meshing makes it impossible to deal with. That is because the density of the cylinders simulating the vegetation is very high.

As a result I would like to desing just a fraction of my channel (1-2 meters for example) and apply "Translational periodicity" between the inlet and the outlet face, so as to ensure that the flow can be fully developed in the part of the channel designed.

Already applied boundary conditions:
I choose two parallel faces (inlet-outlet) to be the domain interfaces and - in order to define the discharge- I apply "Mass Flow Rate". I also apply wall boundary conditions for the sides and the bed, and free slip wall boundary condition for the free surface (single phase problem).
Nevertheless, it doesn't work.

How do I apply "translational periodicity"? Does this work in channel flows?

I am relatively new to cfx and any help you can offer would be highly appreciated, as I have to present my project in two weeks from now.
Thank you a priori !!

ghorrocks March 6, 2011 18:20

If you have so many obtructions to the flow can you consider a porous medium approach and ignore them all and replace them with a resistance?

Giorgitseli May 30, 2011 05:22

What I am actually looking for-and the conclusion of my project-is the resistance due to the presence of vegetation, so I can't define it. Maybe the porous approach is good enough. I will try it!

Thank you very much!

ghorrocks May 30, 2011 06:17

If you are doing atmospheric scale modelling there are plenty of literature articles of how to model the atmospheric boundary layer, including the effects of things like trees. You can use CFX's built-in surface roughness model as a starting point, but I suspect you will quickly find the limitations of that for atmospheric modelling.

All times are GMT -4. The time now is 15:08.