CFD Online Discussion Forums

CFD Online Discussion Forums (
-   STAR-CCM+ (
-   -   Roughness height help (

Remyjnr January 15, 2013 10:08

Roughness height help
Hi, I am a student and have been handed a project using Star CCM+ with zero prior experience, so please excuse any errors, i'll try to be thorough!
I am trying to simulate flow down a compound channel (a smooth deep main channel and a shallow rough flood plain). To calibrate the velocity profile i am trying to use the roughness height of the flood plain, however changing it has no effect, could anyone suggest a reason for this ?
Mesh: 1.7million cells base size: 0.007m
Models: Turbulent (Standard k-e), Liquid (water), segregated, steady, 3D

Any help appreciated ,


abdul099 January 18, 2013 17:25

It can be hard at the very beginning, so it's nothing special to do some mistakes - everybody does this at the beginning.

First, there are a few questions regarding your simulation, since the information you've given here is a bit... little.
Maybe you can post some pictures, that gives us the chance to understand your pain a little better.
Also it's important to know how you judge the velocity profile not to change. Did you create a line probe, point probes, or did you just look at the solution field in a scene? Are there any differences or is there no effect at all? What about your mesh, is it suitable? What about the mesh resolution? What about the resolution of the boundary layer?

When I just think about the last three points, there are several things coming into my mind: A standard k-epsilon model allows to set a roughness height, but this will only affect the boundary condition. So the velocity profile might not change much as long as the roughness is low compared to the first cell thickness. If the roughness height is high compared to the first cell thickness, it might not be valid to go for this approach.

So you see, it might be important for you to post more information, e.g. about the roughness height and your meshing parameters. And of course some pictures (better to see it once than reading it a hundred times).

All times are GMT -4. The time now is 14:16.