[lammps-users] Langevin method

To whom it may concern:

I'm trying to cool my system down using Langevin method. In the document, I found

fix ID group-ID langevin Tstart Tstop damp seed keyword values


damp = damping parameter (time units)

The damp parameter is specified in time units and determines how rapidly the temperature is relaxed. For example, a value of 100.0 means to relax the temperature in a timespan of (roughly) 100 time units (tau or fmsec or psec - see the units command).

I tried to set damp = 1. with units metal

I was expecting to see the temperature goes down to Tstop in around 1ps. But it shows after 20ps, the temperature is still far away from Tstop.

How can I control the time span of the procesure precisely?



        In the latest release of LAMMPS(12th July 07) the restart2data
tool does not seem to write the Improper and Dihedral coefficient from
the restart file. To reconfirm i used the 'peptide' example and
generated a restart file which when i tried to convert to text format
from binary using restart2data.ccp in tools directory did not
incorporate the above mentioned coefficients.


Arnab Chakrabarty
Graduate Student
Dept of Chemical Engineering
Texas A & M University

There's probably some other issue with your input script.
E.g. the temperature you are printing and the thermostat
T are not the same. Or your system is far from equilibrium
b/c of some other driving force you are imposing. Or
you are starting at a T very far from T_desired and it
takes a few oscillations to get to it.

Otherwise, the system shouldn't take many psecs to thermally


Thanks all for the help. I just set Tstart=Tstop=300K and it looks fine now.

However, my system seems to start from zero K to 300K.

I just don't understand it does not start from Tstart. It looks confusing to me though.

I know there's an option to start my system with initial velocity by the
command create. I didn't use it.

Thanks again.

if you do not assign initial velocities, they will be zero at the beginning and that’s T=0.
It usually does not matter as it will heat up in a short time, but if you want to start at 300K
you must use the velocity command.

Just posted a patch for this.


Then Tstart doesn't mean the starting velocity? This is the part that confused me. Thanks for the clearifications.

Tstart is the desired temperature at the beginning of the run.
It may or may not be the current temperature of the system.
That depends on how you set the system up.