Rerun command

LAMMPS LAMMPS Mailing List Mirror
1

Dear Lammps users,

I have a question regarding the rerun command. I am new to Lammps and as an exercise in learning how to use the rerun command and ave/time fix I have been trying to run a simple LJ fluid simulation and dumping the coordinates with the aim of using the rerun command to calculate the radial distribution function at a longer cut-off.

To the best of my knowledge I seem to be reading in the dump correctly as the thermo data from the rerun appears to be consistent with that of the original run.

The rdf however is zero for all of the vector elements and I can’t work out why this may be as presumably the coordinates have been read in correctly.

Assistance with this would be greatly appreciated. I have attached both the initial run’s script and the rerun script which are named in.ljrun and in.ljrerun respectively.

Thank you in advance, Jeff Armstrong.

in.ljrun (700 Bytes)

in.ljrerun (480 Bytes)

2

The problem is that your rerun script doesn't start with any atoms.
The rerun command creates them at every snapshot it reads, but
the compute rdf commands expects there to be a set of atoms
at the start of the (re)run so that it can pre-compute the # of atoms
of each type.

This is an example of the rerun command not really working correctly
with a compute that was designed without the rerun option in mind.
There are probably several others. I need to think about a general
way to fix that.

But for now you should just create the set of atoms in your rerun
script (link in your 1st script), before overwriting them with the
dump file snapshots.

Steve