[Miguel Hernandez]

I'm continuing my studies with interFoam (openFoam 2106) for open channel simulation.
I'm currently working on a simulation of a small river and a concrete bridge.
I'm wondering how to consider the roughness of this two different surfaces.

I was thinking to use the nutRoughWallFunction for both the bridge and the riverbed in the nut BC.
Is this enough? And, what about the Ks values?

For concrete I had supposed Ks = 0.002, but for the riverbed - using a formula I found on some past posts - Ks = 0.55 (Manning n = 0.035).

And last but not least, what is the importance of y+ , and so the last wall cell dimension, in a simulation of this kind?

Thanks in advance...

[David*]

Assuming you mean nutkRoughWallFunction and you want to use the kOmegaSST turbulence model, you'll have to put:

• nut: nutkRoughWallFunction

• k: kqRWallFunction (~zeroGradient, uniform 0)

• omega: omegaWallFunction (uniform 0)

In the nut file, you'll want to define Cs = 0.5 (default value) and Ks in meters as the equivalent sand-grain roughness. 0.55 m definitely seems too large for a riverbed. If you really have rocks like this, you should represent them by their form, not by roughness. If not, try using your imagination/expertise instead of a formula

Regarding y+: as long as you are using those wall functions (not low-Re turbulence modeling) and as long as you do not care about wall shear stress or exact velocity in the wall-adjacent cells, you can largely ignore it from my experience for water in open channels.

Cheers,
David