[2vdba2]

I am using reactingFoam and I noticed that species spread even if I remove diffusive term on YEqn. How to solve it?


I am simulating a simple 1D Plug Flow Reactor by now and species travel much faster than flow speed due to diffusion.


This effect intensifies a specie mass fractions is close to 1. If the problem is due to numeric schemes, which one should I use?





fvm::ddt(rho, Yi)
+ mvConvection->fvmDiv(phi, Yi)
//- fvm::laplacian(turbulence->muEff(), Yi)
==
reaction->R(Yi)
+ fvOptions(rho, Yi)

[Benben]

This should not be occuring. Are you sure that you are calling your modified solver, and that the compilation terminated correctly ?

There can be a bit of numerical diffusion when there is convection, but i doubt that there is enough for you to notice. You should provide the source code for your solver and a simplified test case so that this could be reproduced.

[RobertHB]

Originally Posted by 2vdba2

This effect intensifies a specie mass fractions is close to 1. If the problem is due to numeric schemes, which one should I use?

Holzmann CFD did a study comparing the numerical solution of different schemes, which you can find here.


Please be aware that removing the diffusive term of your transport equation may yield unphysical results. The same way as setting the diffusion coefficient to zero does.


If your simulation is to diffusive, increase the flow speed (see Peclet number)