[ltiago94]

Greetings to all, I'm trying to do a steady-state simulation of 30 s, of a wave going over a block located in a beach and study the influence this exact block has in the flow. I've done 1 or 2 simulations without problems during those 30s without the beach profile and even with a more basic beach profile.

Since I've tried to do the same but this time with a little bit more complex beach profile, I'm having all kind of problems with the simulations. For instance, one of the simulations crashed around the 26 s, and another one crashed around the 22,5s.

In both cases, the deltaT value started to decrease a lot and eventually it all crashed. I'm using interFoam because it's a water-air simulation.

My questions are that maybe now I have to adapt some values in some files to this new and more complex mesh, opposing to the other ones that were running without problems.

Do you guys have any suggestions about this stuff? This is for a masters thesis and time is closing in for the presentation of the paperwork. I'd really appreciate some help.

Thanks

[gtarang]

Generally this happens when turbulent variables starts reaching unrealistic values, like "k" becoming negative or "epsilon" becoming zero in k-epsilon turbulent flow. Check your log file if you see anything like

 bounding <k or epsilon or omega>

.
If possible run your simulation with lower CFL number lets say after 20 sec or so

[ltiago94]

Originally Posted by gtarang

Generally this happens when turbulent variables starts reaching unrealistic values, like "k" becoming negative or "epsilon" becoming zero in k-epsilon turbulent flow. Check your log file if you see anything like

 bounding <k or epsilon or omega>

.
If possible run your simulation with lower CFL number lets say after 20 sec or so

Yes it started reporting bounding k and epsilon.

Are you suggesting running the simulation after the 20 second but changing the CFL number in the controldict to a smaller one?