Simulation takes a long time to run about splishsplash

[janbender]

I tried this scenario but I didn't have the geometry UnitBox1.obj so I used the UnitBox.obj of SPlisHSPlasH.

The problem of this scenario is that you have a high stack of particles. At the same time almost nothing happens. In this case the CFL condition decides to use large time steps. However, solving a high stack with large time steps is not a good idea since a large number of iterations is required which makes the simulation slow. Probably even the maximum number of iterations is not enough and some particles will slip through the boundary. These particles will then dominate the CFL condition and lead to the lowest possible time step. In summary the simulation will be very slow.

You could try to use a fixed time step size or the CFL-method "CFL - iterations" and a CFL factor of 0.5. This CFL method reduces the time step if too many iterations are required since the largest time step does not lead to the fastest simulation.

[ManosAgelidis]

Great answer, thanks a lot! I was suspecting that it had to do with the CFL condition, but was unsure how to fix it. For the record the UnitBox1.obj is just a box with the top surface missing. Thanks again!