|October 4, 2013, 08:00||
Diffusion term in OpenFOAM
Join Date: Oct 2010
Posts: 8Rep Power: 8
checking the term turbulence->divDevReff(U) I came up with the following question:
The diffusion term for incompressible flows should look somehow like this (here, I ignored the viscosity):
fvc::div(fvc::grad(U)+fvc::grad(U)().T())And this can be transformed to this:
fvc::laplacian(U)But when I implement that into OpenFOAM, those terms give different results. I assume it is connected to the discretization schemes, but even when I use linear for all schemes, it gives me different results.
So, what do I do wrong? What do I miss? Is it because of wall-functions or boundary conditions?
Any comment is appreciated...
|February 12, 2014, 10:23||
Join Date: Sep 2010
Posts: 387Rep Power: 11
I think this wiki page might explain what is going on here:
Also have a look at this thread (the difference between compressible and incompressible models goes away if the continuity equation is fulfilled after convergence of the solution):
|February 12, 2014, 10:43||
Join Date: Sep 2009
Posts: 790Rep Power: 14
|Thread||Thread Starter||Forum||Replies||Last Post|
|Moving mesh||Niklas Wikstrom (Wikstrom)||OpenFOAM Running, Solving & CFD||122||June 15, 2014 06:20|
|Numerical Diffusion in OpenFOAM||srikanth_b||OpenFOAM||0||September 28, 2011 04:50|
|Using source term in OpenFOAM||user369||OpenFOAM||13||August 7, 2011 08:51|
|diffusion Term for add. Variable||Zaktatir||CFX||9||July 16, 2011 09:51|
|Numerical diffusion error||Z.Zeng||Main CFD Forum||8||October 22, 1999 09:06|