RE: Periodic cell size

From: Blake Charlebois (
Date: Wed Dec 08 2004 - 23:54:04 CST


I don't have any experience with this kind of thing, but here are my

Does your periodic cell have a net charge of zero? I am under the impression
that simulations of periodic systems are not really valid unless the cell
has a net charge of zero. Otherwise, you are simulating an infinite lattice
of mutually-repulsive cells.

What is the length of your time step? It sounds as though your particles
experience high forces, and therefore high accelerations, so the velocity
accumulated over, say, 1 fs, might be high. I would expect that a
many-particle system needs to be simulated with a temporal resolution high
enough to resolve all particle collisions, since these collisions would
prevent huge velocity build-ups.


-----Original Message-----
From: [] On Behalf
Of Brian Bennion
Sent: December 8, 2004 12:43 PM
To: Suxing Hu
Subject: Re: namd-l: Periodic cell size


Is your cell size big enough? It may be that a few of the particles are
outside the boundary box.

A more thoughtful answer though would include a discussion about the
applicability of PME to massively charged systems. However, I don't have
the experience discuss that...


On Wed, 8 Dec 2004, Suxing Hu wrote:

> Hi all:
> I am using the NAMD to simulate ultracold plasmas, which basically consist
> all charged particles. Such a system is quite extended, normally have a
> size of 50umx50umx100um. I am testing for ~500 particles starting almost
> zero velocity, but it shows at the first time step the particle velocities
> huge (>10^15), and even nan (not a number).
> Are there any limits on the size of periodic cell in NAMD? PME is turned
on in
> my simulation tests.
> Thanks,
> Suxing Hu

**Brian Bennion, Ph.D. **
**Computational and Systems Biology Division **
**Biology and Biotechnology Research Program **
**Lawrence Livermore National Laboratory **
**P.O. Box 808, L-448 **
**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 - 15:38:03 CST