[benk]

Hi, I'm in the process of learning OpenFOAM (been at it for a week now) and I'm trying out some of my own simple models (more like thought experiments to grasp basic concepts of OpenFOAM).

I'm comparing these models to the same model in comsol to make sure that I'm doing everything properly in OpenFOAM. My eventual application will involve highly coupled equations.


One of the tasks that I'm currently trying involves two coupled transport equations on a simple 1m x 1m grid (I'm modelling this as 1D though with zero gradient boundary conditions everywhere except for the left, x=0, and right walls, x=1). My model implementation in OpenFOAM converges only for certain situations (although comsol has no problem with it) and I'm looking for any tips or suggestions on how to improve this or make my model more efficient.

I've attached my solver, but to summarize, the model I'm trying to solve has the following equations:

TEqn: laplacian(DT,T) - beta*c
CEqn: div(phi,c) - laplacian(gamma*c,c) - alpha*T

with boundary conditions:
x=0:
c=10
T=298

x=1:
c=5
dT/dx=-10*c

and other parameters:
DT=4e-3
beta = 1.0
alpha = 0.001
gamma = 0.01
and phi is just a constant velocity field of (0.01 0 0)

My OpenFOAM implementation works well for values of alpha=0.001 but when I try alpha=0.01, it will not converge (and yet comsol has no problem with it). I've attached my solver (ct2.C). My suspicion is that the problem is with the relaxation factors but they don't seem to help me.

Any tips/suggestions?