[gmx-users] Problem with running crystal simulation

[Elton Carvalho]

Hello, Ben,

Does the system blow up during NVT simulations (pcoupl = no) with either
box size?

Simply increasing the box size without adjusting the system (adding more
atoms if you have a periodic molecule, for example) usually makes matters
worse.

Can you give a more detailed description of your system? I made some
assumptions in my reply that may be wrong, because you said "crystal", I
expected a periodic molecule, now you're mentioning adsorption. Maybe you
have a crystalline surface, periodic in the X and Y direction and have some
sort of vapor on top of it? If so, you can prevent the box collapsing in
the z direction by using a semiisotropic pressure coupling and zero
compressibility in the z direction (pcoupltype = semiisotropic, check
http://manual.gromacs.org/online/mdp_opt.html#pc ).

If you need the crystal to stay in place, are you using position restraints
or freezing (mdp option) in any way?

Cheers from sunny Natal,

On Tue, Nov 24, 2015 at 7:23 AM, Ben Tam <btam125 at hotmail.co.uk> wrote:

> Hi Elton,
> Thank you for your reply. I have increase my box size to 10 10 10. However
> it is still blowing up. Yes I am trying to run NPT simulation, will
> freezing the crystal framework will cause the box to blow up? Because I am
> doing an adsorption simulation hence I need the crystal stay in certain
> place.
>
> Best regards,
> Ben
>
> > From: eltonfc at gmail.com
> > Date: Mon, 23 Nov 2015 17:23:58 -0300
> > To: gmx-users at gromacs.org
> > Subject: Re: [gmx-users] Problem with running crystal simulation
> >
> > Hello, Ben.
> >
> > Looks like your box blew up [
> > http://www.gromacs.org/Documentation/Terminology/Blowing_Up ].
> >
> > It would be useful to know the initial box size and how well the
> > minimization converged (were the forces low enough? How was the box
> > pressure?). Are you sure the box size is adequate for your cristal? Is it
> > adequate for your cutoffs (the shortest box vector should be ~2 times the
> > cutoff AFAIK)?
> >
> > If you are trying to run an NPT simulation right after the EM, I'd advise
> > for an NVT thermalization first, so you can equilibrate the temperature
> and
> > forces before messing up with box dimensions and pressures. If your
> system
> > blows up in NVT, that's a symptom of bad topology. If it doesn't blow up
> in
> > NVT and does in NPT, check your pressure options in the MDP: "pcoupl",
> > "pcoupltype", "tau_p", "ref_p" and "compressibility".
> >
> > Is your topology correctly connecting the atoms through the box wall so
> you
> > have a periodic molecule? Have you set "periodic_molecules = yes" in your
> > MDP files?
> >
> > Cheers,
> > --
> > Elton Carvalho
> >
> > On Mon, Nov 23, 2015 at 3:58 PM, Ben Tam <btam125 at hotmail.co.uk> wrote:
> >
> > > Dear GROMACS user,
> > > I have a problem here when I am trying to run a crystal simulation,
> this
> > > is the warning I am getting from the error file:
> > > Warning: Only triclinic boxes with the first vector parallel to the
> x-axis
> > > and the second vector in the xy-plane are supported.         Box (3x3):
> > >         Box[    0]={         inf,         -nan,         -nan}
> > > Box[    1]={        -inf,          inf,         -nan}            Box[
> > > 2]={        -nan,         -nan,          inf}         Can not fix pbc.
> > > And it terminate the run as segmentation fault.
> > > I was able to gmx grompp it without any error, also I was able to run
> > > minimisation but not md. Therefore can anyone tell me what's wrong
> with my
> > > simulation? I have no idea which file to attach so apologise for that.
> > > Best regards,
> > > Ben
> > > --
> > > 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.
> > >
> > --
> > 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.
>
> --
> 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.
>



-- 
Elton Carvalho