[newface5150]

Hello,

I am trying to do a mesh sensitivity study in Star-CCM+ of a CFD model. Currently, I have the result from the current model's mesh, but I am not sure how to proceed with a finer mesh if I want to conduct the mesh sensitivity study on the result.

Does Star-CCM+ automatically interpolate and use the result from the coarser mesh as the initial condition for the finer mesh? (If not done automatically, is there any way to do it in order to use the coarser mesh result as the initial condition for the finer mesh?) Or should I clear the solution and start from the scratch when I run the model?

Please let me know.

Thank you

[FliegenderZirkus]

Does Star-CCM+ automatically interpolate and use the result from the coarser mesh as the initial condition for the finer mesh?

Yes. Unless you click Solution -> Clear Solution, the current fields are kept, even upon remeshing. Whether you want to clear the solution between runs or not depends on what sensitivity your mesh sensitivity measures. Most likely you can keep the old results to speed up the convergence, but there may be scenarios where you would want to start from scratch.

[newface5150]

Originally Posted by FliegenderZirkus

Yes. Unless you click Solution -> Clear Solution, the current fields are kept, even upon remeshing. Whether you want to clear the solution between runs or not depends on what sensitivity your mesh sensitivity measures. Most likely you can keep the old results to speed up the convergence, but there may be scenarios where you would want to start from scratch.

Dear FliegenderZirkus,

Thank you for your explanation. I am working on aeroacoustic simulation and trying to understand the mesh sensitivity based on unsteady CFD runs. When I do a mesh refinement, I usually go back to run RANS without clearing the solution, and then run an unsteady CFD once RANS run reaches the convergence. I wonder if this is the case when I really need to clear the solution. Would my approach require me to clear the previous solution?

Please let me know.

Thank you.

[FliegenderZirkus]

It almost sounds like you wouldn't even need to go back to RANS every time you change the mesh and just continue with the unsteady simulation (is that LES?). But I don't know much about acoustics, so I'm not sure to be honest.

[newface5150]

Thank you so much! It seems that I don't have to worry about it because going back to RANS might be more conservative than continuing with the unsteady simulation. However, it is worthwhile to see if continuing with the unsteady simulation is enough since it will reduce the entire time for the run.

Thank you!

[Aeronautics El. K.]

Originally Posted by FliegenderZirkus

Yes. Unless you click Solution -> Clear Solution, the current fields are kept, even upon remeshing. Whether you want to clear the solution between runs or not depends on what sensitivity your mesh sensitivity measures. Most likely you can keep the old results to speed up the convergence, but there may be scenarios where you would want to start from scratch.

But the Solution Interpolation should be selected in the Physics models as well, right?