search

PMEAL / OpenPNM

A Python package for performing pore network modeling of porous media
http://openpnm.org
MIT License
456 stars 174 forks source link

Problem with how NernstPlanckMultiphysicsSolver passes settings (e.g. time step) to coupled algorithms #1788

Closed mkaguer closed 2 years ago

mkaguer commented 4 years ago

Before running the TransientNernstPlanckMultiphysicsSolver I set the time step in settings to 0.01 and believed that this change would automatically change the corresponding settings in charge and mass algorithms. However, the change did not take effect until after the first time step. That is, the first time step was calculated using 0.1 (the default) and then after was mysteriously changed to 0.01 which is what I set. To fix this issue I had to set the time step for mass and charge conservation algorithms as well as for the multiphysics solver before running the solver.

Note: I was using a multiphysics solver that I made some custom changes in. None of my changes I thought would have caused this error. So I am noting this issue here for now.

ma-sadeghi commented 3 years ago

Hey @mkaguer, what's the status on this? Did you find a fix for this issue?

mkaguer commented 3 years ago

Hi @ma-sadeghi no I did not find a fix for this. I don't even know yet where the issue or bug is exactly. My way around it at the moment is manually setting the time step of each algorithm as opposed to only settings the time step in multiphysics solver.

ma-sadeghi commented 2 years ago

We're going to implement the NernstPlanck algorithm using the yet-to-be-implemented multiphysics solver.