[lammps-users] E_long too high for SPC/E water?

Dear List,

I'm still trying to simulate SPC/E water in Lammps :slight_smile:

I'm now finding that the E_long value seems too high. Using NAMD I
get the configurational energy of my water to be about -45 kJ/mol,
which is what I expect. In LAMMPS the VDW energy seems correct, and
the Coulombic energy is in the ballpark, but the E_Long is much higher
than I would expect it to be. I'm attempting to use PPPM

I've put my input files here:

http://gubbins.ncsu.edu/users/joshua/lammps_question

Am I making a mistake in the implementation of the EWALD/PPPM?

Thanks in advance. Below is a comparison of values between NAMD and LAMMPS.

Josh

E_Coul (LAMMPS): -659.8 eV
E_Long (LAMMPS): -3,069.6 eV
E_VDW (LAMMPS): 105.0 eV
PotEng: (LAMMPS): -3,624.3 eV

Configurational Energy (per 1267 water molecules): -2.86 eV or -276 kJ/mol

E_Coul (NAMD): -694.2 eV (with PME)
E_VDW (NAMD): 105.0 eV
PotEng (NAMD): -589.2 eV

Configurational Energy (per 1267 water molecules): -0.465 eV or -44.9 kJ/mol