[gmx-users] berendsen vs parrinello-rahman

Justin Lemkul jalemkul at vt.edu
Fri May 16 17:49:52 CEST 2014



On 5/16/14, 11:37 AM, Nilesh Dhumal wrote:
> Hello,
>
> I am running the simulation for ionic liquids. I initially did simulation
> with Gromacs VERSION 4.0.7  with "parrinello-rahman" pressure coupling.
> Here I pasted temperature and pressure coupling part of mdp file.
>
> ; Temperature coupling is on
> Tcoupl = v-rescale
> tau_t = 0.1
> tc-grps  =system
> ref_t =   350
> ; Pressure coupling is  on
> Pcoupl              = parrinello-rahman
> pcoupltype          = isotropic
> tau_p               =  0.5
> compressibility     =  4.5e-5
> ref_p               =  1.0
>
> We updated gromacs version to recent  Gromacs VERSION 4.5.5.  With these
> parameter during equilibration, I got the error "system is blowing up".  I
> changed "parrinello-rahman" pressure coupling to berendsen and able to run
> the simulation.
>
> ; Temperature coupling is on
> ;Tcoupl =  Nose-Hoover
> tau_t = 1.0
> tc-grps  =system
> ref_t = 400.0
> ; Pressure coupling is  on
> Pcoupl              = berendsen;  parrinello-rahman
> pcoupltype          = isotropic
> tau_p               =  0.5
> compressibility     =  4.5e-5
> ref_p               =  1.0
>
> Is there any changes in recent version in pressure coupling?
>

Is there any prior equilibration?  Using P-R for the initial phase of 
equilibration can be unstable.  4.0.7 was a bit more forgiving, and this same 
issue has been reported several times before with the 4.5.x series, with the 
solution being to set nstpcouple = 1 for the initial phase of equilibration with 
P-R.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================


More information about the gromacs.org_gmx-users mailing list