Stability of values of parameter set

[francmst]

I want to use Pybamm to simulate a battery with user-defined constituting equations. I started using one of the existing files (PyBaMM/5-half-cell-model.ipynb at develop · pybamm-team/PyBaMM · GitHub) instead of implementing submodules ex-novo. Before making any modification to the equations, I started by changing some of the parameters (such as particle thickness from 100 to 50 um, or particle radius from 10 to 4 um). Most of the modification results in an error, such as:

SolverError: Could not find consistent states: Could not find acceptable solution: .../casadi/core/rootfinder.cpp:280: rootfinder process failed. Set 'error_on_fail' option to false to ignore this error.
SolverError: Could not find acceptable solution: .../casadi/core/rootfinder.cpp:280: rootfinder process failed. Set 'error_on_fail' option to false to ignore this error.

My understanding is that the solver can not find a suitable starting point for the iteration. Is my understanding correct? What is a possible solution or way around?

[brosaplanella]

You might want to play with the solver settings or the meshes. However, bear in mind that changing the parameters is quite tricky and you might end up in unphysical situations.