output of fix langevin

LAMMPS LAMMPS Mailing List Mirror
1

Dear Sir/Madam,

I’m using fix langevin to thermostat a slab of 3C-SiC with its (111) surface facing to the vacuum. I set the fix commands as:

fix 1 all nve
fix lange all langevin 300.0 300.0 0.025 598043 tally yes

I run the system for 1 ns, and output the return scalar of fix langevin (f_lange). I found that the value of f_lange is always increasing linearly throughout the simulation (from zero to about 50 eV). Is this normal? I think f_lange should stay constant after the system reaches the equilibrium since no energy is added in or removed from the system, right?
I appreciate your help!

thanks,
Jibao

2

Dear Sir/Madam,

I’m using fix langevin to thermostat a slab of 3C-SiC with its (111) surface facing to the vacuum. I set the fix commands as:

fix 1 all nve
fix lange all langevin 300.0 300.0 0.025 598043 tally yes

I run the system for 1 ns, and output the return scalar of fix langevin (f_lange). I found that the value of f_lange is always increasing linearly throughout the simulation (from zero to about 50 eV). Is this normal? I think f_lange should stay constant after the system reaches the equilibrium since no energy is added in or removed from the system, right?

​
​for that to be true, you have to have a simulation that conserves energy sufficiently well, when running with fix nve alone without fix langevin after the system is sufficiently well equilibrated. a thermostat can hide this, and your observation suggests, that this may be the case here.

​axel.​

3

Dear Axel,

I got it. Thanks for the tip!

Best,
Jibao