[lammps-users] Langevin thermostat

LAMMPS LAMMPS Mailing List Mirror
1

Dear users,

does anybody know well if people want to equilibrate a homogeneous system by means of langevin thermostat, in order to get system stablized at Tstop, what would be the estimated total simulation time based on a specified damping parameter (usually the 6th parameter in fix langevin command)? I guess the damping parameter is something similar to tau = RC in the RC circuit, but I did not achieve to get system stablized at Tstop after 5-7 times the damping parameter of simulation time. Could anyone explain the relation between the damping parameter and total simulation time?

Thanks and regards!

Xiang Gu

2

Normally, 5-7x the damping time should be OK. If you started
far away and it is oscillating, maybe you need to run longer.
Since the damping time is normally only a few hundred timesteps,
it shouldn' t be a problem to run for 10K or more steps
to equilibrate.

Steve