[nnprashant]

Hallo Foamers,

I want to explain in detail, the task at hand and the outcome of results using OpenFOAM v11 and comparing it with ANSYS Fluent

The task is a classical T-Junction Pipe with 2 inlets and a single outlet. The flow media is a multicomponent (O2, N2, H2Og, H2) non-reacting compressible gas with different mass flow rates. The main goal is to investigate the H2 Diffusion in Pipe and at outlet.

A total of 4 meshes (Hexa, Poly, PolyHexaCore and Tetrahedral) were used to monitor the impact of discretization error in OpenFOAM.

To properly capture turbulence, wall resolved meshes were created and Realizable K-Epsilon turbulence model was chosen. To avoid abrupt fall of K residuals for Realizable K Epsilon model, the simulations were started with Standard KEpsilon and were dynamically changed to Realizable K Epsilon.

Different parameters like Wall Yplus, H2 Diffusion, WallShearStress were chosen to compare the effect of mesh type and find the differences between OpenFOAM and Ansys Fluent.

In OpenFOAM, all schemes are set to Second order and a robust FvSolution was defined which was suitable for all mesh types.

What are the findings :
[LIST] Except for Hexa mesh, the H2 diffusion in the pipe is really large even when using second order schemes. Is it a known issue or is there any solution to it.

• The result of Poly Hexa Core mesh in OpenFOAM matches good with Ansys Fluent with the same mesh, when 1st Order Upwind scheme for species transport is selected in Ansys Fluent.

The Hexa result in OpenFOAM matches good with the Poly Hexa Core with Ansys Fluent using 2nd Order Upwind scheme for species transport is used.

Please find the pictures attached and any help or suggestion is highly appreciated.


Physical Properties:
FoamFile
{
format ascii;
class dictionary;
location "constant";
object physicalProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
thermoType
{
type heRhoThermo;
mixture multicomponentMixture; // coefficientWilkeMulticomponentMixture ;
transport sutherland;
thermo janaf;
equationOfState perfectGas;
specie specie;
energy sensibleEnthalpy;
}


defaultSpecie N2;

#include "thermo.compressibleGas"



I have tried both the mixture properties mentioned above and no change in the final result of H2 diffusion was observed.

Main question is as how to set the gradient of the species solver in OpenFOAM v11 to second order ?

[nnprashant]

Main question is as how to set the gradient of the species solver in OpenFOAM v11 to second order ?