[gmx-users] water molecules crashing system

Shayna Hilburg shilburg at mit.edu
Tue Apr 24 16:52:27 CEST 2018


Hi,

I have a a couple of SPC/E water molecules restrained within a low
dielectric medium in a system that also contains ions (restrained to a
different region of the box). I was able to run the simulation for several
nanoseconds before it crashed due to "water molecule ... can not be
settled". The system can run for a very long time without error in the
absence of the water molecules.
I thought it was due to my slight background fields from floating point
math but Justin helped me understand that is not the case. So, my best
guess is this could have to do with domain decomposition or PME grid size
since there are so few charges in that region, but I'm not sure how to go
about fixing it.  Also, I'm wondering if I should be playing with any of
the solvent optimization since perhaps that is intended for bulk solvent
rather than individual water molecules (such as turning off water
innerloops?).

Thank you!
-Shayna


More information about the gromacs.org_gmx-users mailing list