[gmx-users] nve energy conservation

[Mark Abraham]

Hi,

That setup is guaranteed to leak. You specified -1 for the tolerance which
means you get to control rlist to choose the buffer size, which is zero
because you chose it to match both rcoulomb and rvdw. Check out the docs of
verlet-buffer-tolerance and try something else!

Mark

On Wed, Mar 14, 2018, 20:08 Jo <jojo412202 at gmail.com> wrote:

> Hello,
>
> I posted earlier regarding running an NVE simulation for a box of 1000 SPCE
> water.  I have tried  many different parameters based on other gmx_users
> email list questions (changing verlet-buffer-tolerance, timesteps, cutoffs,
> coulombtype, vdw-type, etc), but still I am unable to conserve energy. The
> energy shift is catastropic (~300 kJ/mol/ns).
>
> Thanks Mark for referring me earlier to the manual with the energy drift
> figure!  I think the drift I am experiencing is much more significant than
> that depicted in the figure though.
>
> Below are the parameters I am using, I feel like I must be making a major
> mistake in my mdp file.  I assume using Potential-shift for coulombic and
> vdw will satisfy energy conservation for both energies.  I have not tried
> double precision yet but I assume it will not help with the massive energy
> shift I am seeing.
> Please let me know if anyone has any suggestions on what I can try or
> anything I am wrongly assuming!
>
> verlet-buffer-tolerance  = -1
> rlist = 1.3
> coulombtype = PME
> rcoulomb = 1.3
> coulomb-modifier = Potential-shift
> vdw-type = Cut-off
> rvdw = 1.3
> vdw-modifier = Potential-shift
>
> Best,
>
> Jo
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>