[lammps-users] pressure

Hi:
I used LAMMPS and made a simulation on a 500 particle system using NVE with a potential(Lj) cutoff distance of 2.5σ . The neighborhood lists distance was 0.3 σ and lists were updated every 5 time-steps with
a time-step of 2 × 10–15 s. Initial configurations were generated randomly according to FCC lattice.
Initial velocities were generated randomly from a Gaussian distribution. The results of simulation for MSD and temperature and volume ( density) are very good but for pressure I get unreasonably high pressure!
when I use the results of LAMMPS and calculate pressure from the first equation in p.216 of manual, I can get more reasonable results for reduced pressure.I am using thermo_style custom command to print reduced pressure.
I am looking forward to hearing back and receive comments to find the reason for my problem.
Thank you so much,
Fateme

It is my understanding, small system, More pressure fluctuations, large system, less pressure fluctuations. 500 particles is too small, I think

2010/5/6 Esfandiarinia, Fatemeh N <fatemeh.nia@…1514…>

Thanks, but I increased the number of atoms to 4000 and still same results! I calculate the reduced number density as (N=number of atoms)/(reduced volume). and I calculate reduced volume from the reduced mass density. is it wrong?!
on the other hand when I use the results of LAMMPS and calculate the pressure from the first equation in p.216 of manual I can get more reasonable results for pressure!
am I doing something wrong?

Hi:
I had a problem with the results of LAMMPS for the pressure. I got a reply( Vikas) about the number of atoms, I increased the number of atoms to 4000 and still I get the same results. I am wondering, maybe there is a problem in printing the results since every other thing is fine! I guess ,instead of reduced pressure, LAMMPS is printing another parameter(?)! specially, when I calculate pressure from the results of LAMMPS for r*i,f*i,T*,V*( reduced coordinates and force and temperatures and volume), I can get more reasonable results for P*( reduced pressure).
I hope to hear back from you.
thanks in advance,
Fatemeh

Pressure, as output by LAMMPS, is always normalized by the
number of atoms. Like temperature, it's an intensive quantity.
If you are using LJ units, then it's also a reduce pressure (unitless)
if that's what you mean.

Steve

2010/5/14 Esfandiarinia, Fatemeh N <[email protected]...>: