Changing vdwtype leads to change in coulomb potential when coulombtype is RF-Zero
When coulombtype = Reaction-Field-Zero and vdwtype = cutoff is used, the electrostatic interaction shows extremely high energy
(Coulomb(SR) = 2.05078e+07 at step 0. Note the initial structure in an attached file is well-equilibrated with PME.)
Oddly, when vdwtype = Shift, this problem does not occur.
(Coulomb(SR) = -5.27937e+05 at step 0.)
Steps to reproduce:
grompp_d -f rf.mdp -o test mdrun_d -deffnm test
#1 Updated by Erik Lindahl over 5 years ago
- Status changed from New to Closed
- Priority changed from High to Normal
The attached files produce fine results in the current git branches both of 5.0 and 4.6. Since there was no specific version listed I have not bothered checking, but if it occured it must have been fixed by some earlier patch.