Re: Atoms mooving too fast ... still the problem

From: Peter Freddolino (
Date: Tue Apr 03 2007 - 20:12:05 CDT

Hi Andrea,
I'd recommend two things:
-Have a look at your trajectory before the crash, and make sure the
volume isn't fluctuating too violently (which usually means an
insufficiently equilibrated system). You may want to write dcds at every
timestep starting immediately before the crash to better observe this
-Try the restart from a recent checkpoint option and see if it helps; if
you get a big change in volume, the initial distribution of atoms into
patches won't be valid, and you need to restart


Andrea Diaz wrote:
> Hello,
> I have been running the equilibration again, minimizing for 5000 steps and increasing the pairlistdist parameter from 18.5 to 30 ( my cut off is 14 ), it didn't work. Now I am getting this error:
> FATAL ERROR: Periodic cell has become too small for original patch grid!
> Possible solutions are to restart from a recent checkpoint,
> increase margin, or disable useFlexibleCell for liquid simulation.
> I am giving the dimensions to the cell vectors using VMD, I used the minmax command and I assigned the value of the length of the box that I measure from vmd to the configuration file. As it is an NPT simulation the volume of the cell is changing with the time.
> I am using NPT simulation, for that reason I don't believe that disable the UseFlexibleCell parameter is a good solution.
> Thanks a lot .. I will be waiting for your suggestions
> Andrea
> ¡Miles de amigos con tus mismas aficiones!

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:46:10 CST