LAMMPS USERS

I’ve been running into a problem with the LAMMPS velocity set command. It appears to only work with LJ units, switching to real units causes the group of particles with velocity to seem to only move with the intended velocity for one timestep, before seemingly resetting their velocities to zero. Is this a LAMMPS bug, or might I be making an error in my understanding of the velocity command?

Thanks so much for your help.

Best Regards,
Zachary Murray

LAMMPS CODE.txt (1.18 KB)

molefraglj.ico

molefragreal.ico

I've been running into a problem with the LAMMPS velocity set command. It
appears to only work with LJ units, switching to real units causes the
group of particles with velocity to seem to only move with the intended
velocity for one timestep, before seemingly resetting their velocities to
zero. Is this a LAMMPS bug, or might I be making an error in my
understanding of the velocity command?

​the problem seems to be more in your understanding of the units command.

if you change the units, *all* parameters that have units need to be
changed accordingly, for instance the lennard jones parameters epsilon and
sigma, velocities, distances, time. if you don't, you are simulating *very*
different systems.

axel.

p.s.: do not use temp/rescale it is *very* bad for any simulation that is
supposed to have meaning.