fluid velocity and time step

[Flopi]

Dear forum,
I am trying to simulate the evolution of a core subjected to a flow a fluid. In my current situation, I would like to have cells of 0.002m length and a velocity of fluid of 1e-5 m/s.
If I understand correctly the behavior of the TRANSPORT keyword, this forces me to take a time step of 200s.

However, I would like to simulate the behavior of such system over several weeks or even longer and with such a time step this means extremely long calculations. Is there any way to increase the time step while not modifying the cell size and the fluid velocity?

Thanks for your help,

[dlparkhurst]

In short, no. Once you pick the cell size, the velocity sets the required time step with TRANSPORT. It is chosen this way to minimize numerical dispersion.

PHAST has no time/cell size requirements when using implicit time and distance weighting, so you can adjust either or both time and distance discretization. However, you will be adding numerical dispersion as you increase discretization. Other transport models have TVD (total variation diminishing) capabilities to automatically adjust time stepping to minimize numerical dispersion.

[Flopi]

Well, I was half expecting this answer... I guess I'll be doing long simulations then !
Thanks

[dlparkhurst]

If possible, start with coarser discretizations, at least until you are sure you have the system working the way you expect. Then sequentially refine the grid to see if the finer discretization is making a difference.

If you are talking about weeks of calculations, I would consider PHAST. It may not have the features you need, but if it does, it is simpler to evaluate grid/time step convergence, but most importantly, it is parallelized.