[gmx-users] Coulomb energy with cutoff-schemes verlet and group

[Donnini, Serena]

Hallo,

I am observing a strange behaviour with gromacs 2016.1.
I calculate the potential energy for a charge in a cubic box of 50 nm in vacuum with

coulombtype = Cut-off
vdwtype = Cut-off
nstlist = 10
ns_type = grid
rlist = 1.0
rcoulomb = 1.0
rvdw = 1.0
pbc = xyz

if I set
cutoff-scheme       =  group

I get (just one step mdrun) a Coulomb (SR) energy term of zero as expected.
If, however,

cutoff-scheme       =  verlet

the Coulomb (SR) energy is different from zero (in the order of tens of kJ/mol). This term
becomes smaller if I increase the rcoulomb.

Has anybody dealt with similar behaviour? How could I set the mdp options with verlet to avoid this?

Thanks!

Serena