Learning Forum

[Guojun]

Hello,

I am simulating a case using Fluent 2021R1 on my office computer (Windows system) and HPC(Linux system). However, when the same case was simulated on both system with turbulence model employed, the simulation results are very different. The below picture is the contours of liquid fraction on a surface.

[Rob]

Can you check convergence and whether both are single/double precision. What is the mesh like at the interface as you're not getting liquid in the stationary zone either.

[Guojun]

Dear Rob Thanks for you help. All the cases are convergence and double precision. The red area is a porous media, and its mesh type is hexahedron. For other area, I did not set the mesh type and the grids were automatically generated. Many thanks.

[Rob]

That should give the same result. Can you post contours of velocity etc to see what else is going on? Has anyone else checked your model set up, and are you using any UDF code?