[sb612]

Hi Niels,

I have spent quite a lot of time attempting to replicate the surface elevation and undertow results for spilling waves from your paper/thesis using the waves2Foam toolbox with OpenFOAM 2.2.0. I have had a number of problems and any help which you can give would be very much appreciated.

As you pointed out in your thesis, the standard implementation of the k-omega model in OpenFOAM does not perform well. In my experience, the waves do not even seem to break due to excessive damping. Therefore, I have tried adding the changes which you have suggested:

a) The modified production term and new value for alpha
b) The cross diffusion term.
c) Density in each term.

Changes a and b seem to have been implemented correctly but although the simulation runs, the model still does not perform well. On the other hand, change c gives more promising results in the early stages of the simulation but as soon as the first wave breaks, the simulation becomes unstable and soon after crashes. The attached image shows the free surface and turbulent kinetic energy at the start of the first breaking wave (top) and the next write time (bottom). Note that the TKE in the air just above the breaker jumps approximately two orders of magnitude in that time.

Bearing all of this in mind, I have a few of questions:

1. Did you have any stability problems when running the spilling breakers case? If yes, do you have any suggestions on how to improve it?
2. How did you include density in the equation? Did you pass rho and rhoPhi from waveFoam to the turbulence model and then include these in each of the terms in the two transport equations?
3. You say in your thesis that the values for k and omega at the inlet/atmosphere are small fixed values such that nut<<nu. However, I have found that if I choose small fixed values my simulation crashes. To get it to run I have had to use inletOutlet conditions. Is it possible for you to tell me the exact values which you specified on your boundaries?

Many thanks,

Scott

[sb612]

Hi Niels,

I am still trying to reproduce the spilling breaker results from your paper. The problem I am having is that the wave breaks too early, similar to your paper when you used an aspect ratio of 2.

However, my mesh has an aspect ratio of 1 and I have updated the turbulence model and wall functions according to the suggestions in your paper/on this forum. i.e. I have included density and changed the production term to depend on the rotation of the field.

Have you got any further suggestions on how to make the wave break later? Any help that you can provide would be greatly appreciated!

Many Thanks,

Scott