From: JC Gumbart (gumbart_at_ks.uiuc.edu)
Date: Sun Jun 18 2006 - 20:55:11 CDT
A couple things I noticed:
If you are using a *.restart.xsc file, you don't need to define the
cellOrigin again. For similar reasons, you should remove the
"reinitvels" option which might override the binVelocities.
Also, make sure your PME GridSizes aren't too small. They should be
at least equal to the dimensions of your box (keeping the 2,3,5 rule
in mind of course).
Is the drift random across multiple simulations? Or is it always in
the same direction? Check how the duplicate cells look in VMD; you
can check "z" under the periodic tab.
On Jun 18, 2006, at 10:10 AM, ghg_at_xalt.am wrote:
> I have lipid bilayer - water system, which consist from 128 lipid
> molecules, 64 molecules at each side. Starting system equilibrated
> and from each side of lipid
> bilayer I have same quantity of water.
> During simulation center of mass of whole system not changing(its
> normall), but lipid bilayer center of mass slowly going down in z
> direction, and on 20ns of my simulation on upper side there 2000
> water molecules and on lower side 970 molecules.
> I understand that WrapAll function enabled and water molecules
> transfering over Periodic Boundary Box. But what is the physical
> reason for that type of transfer of lipid bilayer?
> Additionally I attach simulation configuration file.
>
> Thank's
> Hrant Gharabekyan
> Bioinformatic Group of National Academy of Science of Armenia
> <NPT.conf>
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:13 CST