problem with langevin and temp/com

LAMMPS LAMMPS Mailing List Mirror
1

Hi,
I would like to correctly subtract the non-zero velocity of the center
of mass in a langevin dynamics by doing the following:

run_style verlet
fix mynve all nve
fix mylangevin all langevin \{temp\} {temp} 1.0 1905655
compute mytemp all temp/com
fix_modify mylangevin temp mytemp
run 1000

but it does not work, the temperature of the system slowly explodes.
Where is my mistake?
Thanks
Stefano

2

Hi,
I would like to correctly subtract the non-zero velocity of the center
of mass in a langevin dynamics by doing the following:

run_style verlet
fix mynve all nve
fix mylangevin all langevin \{temp\} {temp} 1.0 1905655
compute mytemp all temp/com
fix_modify mylangevin temp mytemp
run 1000

but it does not work, the temperature of the system slowly explodes.
Where is my mistake?

stefano,

please try updating to the very latest version of LAMMPS
and use the corresponding feature that was just added
to fix langevin by aidan thompson a couple days ago.

cheers,
    axel.

3

What units are you using? A damping coeff of
1.0 may be bad. The coeff should be equivalent
to 100+ timesteps.

Steve

Steve