[gmx-users] Problem with COM movement

[Tomy van Batis]

Dear all

I have the following system:

Two substrates (which are FCC lattices consisting of LJ particles) in a
distance of 50nm and between these substrates I put completely stretched
polymer chains. The substrates are parallel to xy plane and the chains are
vertical. The length of the stretched chains is 30 nm and I put them in the
middle, so the distance of the each substrate to be 10nm.  The interaction
between the substrates and the polymers is repulsive and it’s the same for
each substrate.

I want my substrates to be stable during the simulation, but to be able also
to move under compression (in the z axis).

For that purpose in my .mdp file I define the following:

*nstcomm                  = 1*

*comm-grps                = ZZZ   **,where ZZZ is the name of the residue of
the substrates*

* *

I have 2 problems:

1)      At the beginning, I run NVT for 20ns and the chains are moving UP
during the simulation.

Since my system is completely symmetric, I would expect to have symmetric
movement of the chains, i.e. symmetric density profile.



2)      In my .log file the following message appears:

*Large VCM(group rest):      0.00000,     -0.00000,      0.00001,
T-cm:
inf*

It doesn’t appear as an error message, but it appears thousand times in my
log file. I had a look in the mailing list about that, but I did not find a
proper answer.



Can anyone help me?



I also have one question:

When I remove that COM motion of one residue, i.e. from the ZZZ (by adding
comm grps: ZZZ in my mdp file), what happening with the COM motion of the
rest of the system?



I hope I was clear!



Thanks in advance, Chrysostomos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110111/4680a070/attachment.html>