[rgravoue]

Problem Description:
I am attempting to determine the pressure difference and outlet velocity of a simple CAD part, with the inlet velocity and diameter as known quantities. I am using Simflow, a GUI for OpenFoam. The inlet velocity is 0.69 m/s and diameter is 12.418 mm. The fluid being used in the simulation is air.

I have been able to obtain a result however, when compared to the known value, there is a 56.23% difference. I checked the mesh & physics of the model, but remain unsure why such a large difference exists.

Can anyone suggest possible sources of error, things I should check?

Will a finer mesh improve the result significantly?

Is numerical diffusion a concern with this simulation?

Below is a description of what I have done so far.

Attempted Solution:
Mesh:
I have cut the model in half to limit the amount of nodes in the mesh. The mesh checker cannot find any problems but, there seems to be a few skewed regions. The meshing process also created another boundary, which was originally part of a wall. For some reason, this feature doesn’t appear to affect the solution when viewed in Paraview. Simflow seems to be treating it as a wall, even though it is grouped under symmetry and located outside of the symmetry plane.
I have used the “Pipe Tee Meshing” and “Catalytic Converter” tutorials to create my mesh, but there are still many meshing features I don’t fully understand.

Solver:
When setting-up the solver, I followed the “Catalytic Converter” tutorial, ignoring porous material sections of the set-up. Since most of the physics I see in the solver has not yet been formally introduced in classes, I am entirely dependent on the tutorial. For example, the mixing length was not explained in detail during the tutorial. Although I have used what seems to be a reasonable value (given what I could find online), I am unsure if it is correct.

Post-Processing:
When running Paraview, the fluid velocities within the part seemed to coincide with previous simulations using other simulation software.

Please let me know if you require any further information. Attached are some pictures of my simulation: