[Stargazer]

Hi experts,

I am working on the unsteady simulation of aerodynamics, coupled with Lagrangian phase (water droplets). I use clusters(hundreds of cores) for simulations, with Star ccm+12.06 R8. When the simulation goes on, there gives warning: xxx parcels reached maximum number of sub steps. The simulation becomes quite slow. I then restart the simulation on my laptop, with 12.06(no R8), there is no such warning. I run a few iterations and then upload to the clusters again, the simulation runs without such warning. I am wondering what happened in such a situation?

[Stargazer]

Originally Posted by Stargazer

Hi experts,

I am working on the unsteady simulation of aerodynamics, coupled with Lagrangian phase (water droplets). I use clusters(hundreds of cores) for simulations, with Star ccm+12.06 R8. When the simulation goes on, there gives warning: xxx parcels reached maximum number of sub steps. The simulation becomes quite slow. I then restart the simulation on my laptop, with 12.06(no R8), there is no such warning. I run a few iterations and then upload to the clusters again, the simulation runs without such warning. I am wondering what happened in such a situation?

I think I more or less found it out. To keep a record here: the warning: xxx parcels reached maximum number of sub steps, can be solved by increase the Active Parcel Fraction Cuff-off parameter in the Lagrangian solver control, in my case change from 1E-4 to 5E-4 would work.
The R8 suffix differentiates the mixed and double precision versions. Two versions might give different track results due to different memory and accuracy on particle tracking calculations.