net translation in long simulations

From: Mehdi Bagheri-Hamaneh (
Date: Tue Dec 19 2006 - 17:00:59 CST

Hi all,

I have posted my question before, but nobody replied, so I thought I was
not very clear. That is why I am sending this email. I am trying to
study the diffusion of a protein using MD simulations. According to the
NAMD documentation, using the Particle-Mesh-Ewald (PME) method, will
result in a net total momentum for the whole system, and there is a
command (ZeroMomentum) in NAMD that removes this net translation. But
the default value for this command is NO, which means by default the
total momentum is not zeroed. I have already run long simulations with
this option, and want to continue the simulations. My question is that:
Considering the fact that I am going to study the diffusion of the
protein, should the value for ZeroMomentum be set to "yes"? If the
answer is yes, how important this is going to be? Does anybody have an
estimate of the error that this may cause? If this is really important I
will have to run my simulations again.

Thank you so much,


This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:58 CST