[lammps-users] "Energy not a number" error from REAX

LAMMPS LAMMPS Mailing List Mirror
1

Hi all:
   I am simulating a C-H-O system with 10000 atoms using reaxff as implemented in lammps. I got the error "energy not a number" and the energy terms
   ev, et and eco
   are NaN in the calculation.
   This error doesn't show up for a 2000 atoms system. Could anyone give me some suggestions? Thank you
Best,
Zhiping

2

This indicates that one of the ReaxFF math functions received an invalid argument. This could occur for a very unphysical geometry (due to too large a timestep, etc), but could also be caused by array overflows. I would suspect the latter in this case. We recently (4Jul) posted a patch for two potential memory overflows. You should make sure you are using it.