Running dynamics with the environment fixed

GROMACS version: 5.1
GROMACS modification: Yes/No No
Here post your question

Hi all,

I am trying to run the dynamics for the protein in the water box (NVT).
Only protein and the relatively small solvent shell around the protein moves, while the surrounding solvent molecules are fixed.
The dynamics looks great, but the LJ (SR) energy term becomes extremely huge (from 410^4 to 610^9). Why can this happen?

Thanks!