From: Brian Bennion (brian_at_youkai.llnl.gov)
Date: Fri Apr 23 2004 - 10:57:09 CDT
Hi Gianluca
Please get a copy of Leach's "Molecular Modeling" and become comfortable
with it...
Briefly:
The temperature drifts are not a desirable feature but a sign of
integration errors or other problems.
In NVE I gently heat my system to the desired temperature, and then allow
some time for equilibration at this point I use "bad" temperature methods
to stabilize the system if needed. then I use a restart file and the
temperature/energy is no longer under my control.
The temperature is a macroscopic property of the system, no matter what
anyone says, the temperature is not controlled, the energy is. In NVE the
energy is supposed to remain constant by definition not by setting a
variable in the simulation. This requires a well behaved system/setup and
good integration methods. I know immediately when there are system
problems because my energy outputs won't be constant.
Old world stuff by todays thinking, but it is sound....
Brian
On Fri, 23 Apr 2004, Gianluca Interlandi wrote:
> Hello Brian,
>
> Many thanks for your explanations. This means that the time it takes
> untill the temperature drifts to the desired temperature (10ns) can be
> regarded as "equilibration".
>
> But, how do you fix then the temperature in the NVE ensemble? Is it simply
> assigned at the beginning with "temperature" and nothing else?
>
> Thanks,
>
> Gianluca
>
> On Thu, 22 Apr 2004, Brian Bennion wrote:
>
> > Hello Gianluca,
> >
> > if no coupling is asked for none is received. Without any temp bath you
> > are riding the effeciency of the integration scheme in NAMD (verlet).
> >
> > I routinely run without any coupling, ie NVE ensemble, if your system is
> > properly equilibrated then temp baths can be discarded within this
> > ensemble.
> >
> > My training taught me that coupling, especially temperature
> > coupling was "bad", in that you technically start a new simulation
> > everytime you bleed energy off the system.
> >
> > So thats my $0.02
> >
> > Brian
> >
> >
> > On Thu, 22 Apr 2004, Gianluca Interlandi wrote:
> >
> > > Hi all,
> > >
> > > I have to re-post this message in a more detailed fashion, as it is really
> > > worrying me.
> > >
> > > I have done many simulations with NAMD, where I have coupled the system to
> > > a Nosé-Hoover Langevin piston pressure control with the following
> > > parameters:
> > >
> > > langevinPiston on
> > > langevinPistonTarget 1.01325
> > > langevinPistonPeriod 100.
> > > langevinPistonDecay 50.
> > > langevinPistonTemp 400
> > >
> > > I also wanted to couple the system to a heat bath to keep the temperature
> > > constant:
> > >
> > > tCouple on
> > > tCoupleTemp 400K
> > > tCoupleCol B
> > >
> > > I have realized later that the values in the B-column of the
> > > coordinate-file were all set to zero. This means, that I didn't have any
> > > temperature coupling. When I plotted the temperature in function of the
> > > frames, I saw a drift from 350K to 407K during 10ns. What has happened?
> > > What did bring the temperature to the desired value? Is there any hidden
> > > temperature coupling or temperature scaling, when none is specified?
> > >
> > > I would be very glad to get a hand from anyone out there.
> > >
> > > Many thanks,
> > >
> > > Gianluca
> > >
> > > -----------------------------------------------------
> > > Gianluca Interlandi gianluca_at_bioc.unizh.ch
> > > +41 (0) 1 635 55 93
> > > +41 (0)79 655 73 35
> > > http://biocroma.unizh.ch/gianluca/
> > >
> > > PhD-Student of the Department of Biochemistry
> > > at the University of Zurich Irchel, Switzerland
> > > -----------------------------------------------------
> > >
> >
> > *****************************************************************
> > **Brian Bennion, Ph.D. **
> > **Computational and Systems Biology Division **
> > **Biology and Biotechnology Research Program **
> > **Lawrence Livermore National Laboratory **
> > **P.O. Box 808, L-448 bennion1_at_llnl.gov **
> > **7000 East Avenue phone: (925) 422-5722 **
> > **Livermore, CA 94550 fax: (925) 424-6605 **
> > *****************************************************************
> >
> >
> >
>
> -----------------------------------------------------
> Gianluca Interlandi gianluca_at_bioc.unizh.ch
> +41 (0) 1 635 55 93
> +41 (0)79 655 73 35
> http://biocroma.unizh.ch/gianluca/
>
> PhD-Student of the Department of Biochemistry
> at the University of Zurich Irchel, Switzerland
> -----------------------------------------------------
>
*****************************************************************
**Brian Bennion, Ph.D. **
**Computational and Systems Biology Division **
**Biology and Biotechnology Research Program **
**Lawrence Livermore National Laboratory **
**P.O. Box 808, L-448 bennion1_at_llnl.gov **
**7000 East Avenue phone: (925) 422-5722 **
**Livermore, CA 94550 fax: (925) 424-6605 **
*****************************************************************
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:38:36 CST