From: Todd Trimble (TODD.TRIMBLE_at_asu.edu)
Date: Tue Nov 14 2006 - 10:46:19 CST
Hi All,
Thanks for responding Raffaele. As far as I know the electric field is
not time-dependent. Unless you add explicit time dependence via some
script, the E-field should just be applied instantaneously at the
beginning of the simulation with given magnitude and direction. I tried
changing the first time step anyway and got the same result. The
initial volume of my system is the same as the final volume of the
previous run, which makes sense since I use that configuration for the
restart. But it's very strange that the initial MISC energy is much
smaller than the previous final MISC energy, and that the volume quickly
(over several Langevin piston periods) decays to a value about equal to
the initial volume of the previous run. Could this be related to the
reinitialization of the PME grid?
Todd
-----Original Message-----
From: curcio [mailto:r.curcio_at_bioc.unizh.ch]
Sent: Tuesday, November 14, 2006 1:55 AM
To: Todd Trimble
Cc: namd-l_at_ks.uiuc.edu
Subject: Re: namd-l: electric field and MISC energy
> However, when I start a new
> simulation from the restart values of the previous run, the MISC
energy
> immediately jumps up and the volume quickly decays to values close to
> the starting values of the previous run.
Hi,
when using an additional time-dependent potential (E(t)) in a simulation
(such
as a time-dependent electric-field) it's important to set
"firsttimestep" and
to set it correctly whenever restarting a simulation.
By default "firsttimestep" is set to zero and so E(t)=E(t=0) at each
restart.
If you haven't already, try setting "firsttimestep" to the total number
of
previous (integration)steps at any restart - this might solve the
problem.
Best,
Raffaele
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:49 CST