[lammps-users] Problem with temperature

LAMMPS LAMMPS Mailing List Mirror
1

Hello users,

I am running a simulation on CNT using LAMMPS. I tried to set up a temperatre of 300K using NVT but i could not. It’s only around 180K after 25000 timesteps with 1fs timestep. I am attaching the log file. Could some one help me with this? There are 400 atoms in my structure i input the coordinates of all these atoms using read data.

THANKS
Ajit

units metal
dimension 3
boundary s s p

atom_style atomic

pair_style airebo 3.0

read_data coordcntcoordinates.d

pair_coeff * * CH.airebo C

#read data command for coordinates of atoms

#create velocities on all atoms, using T = 300, with a

velocity all create 300.0 873586443 dist gaussian

fix 1 all nvt 300.0 300.0 100.0

#output of varaibles
timestep 0.0001

thermo 1000
dump 1 all atom 250 positionr.metal
run 25000

2

These lines:

fix 1 all nvt 300.0 300.0 100.0
timestep 0.0001

in metal units mean you are using a 0.1
fmsec timestep, and the time period on which
NVT is operating is 100 psec (million steps).
So you'd have to run for multiples of 1 million timesteps to
equilibrate the temperature.

Steve