In some (realistic) scenarios the pde solver still chooses a too large time step resulting in exploding solute concentrations

kreft / iDynoMiCS-2

Other

7 stars 1 forks source link

In some (realistic) scenarios the pde solver still chooses a too large time step resulting in exploding solute concentrations #134

Closed bascee closed 5 years ago

bascee commented 8 years ago

For now halving the mini time step seems to result in a more stable simulation

bascee commented 5 years ago

Recent work have solved multiple issues that stopped multigrid before convergence, it appears that this was also causing high solute concentration spikes.