[lammps-users] temperature drifting towards zero

Dear Steve, I wrote this query on lammps-user list but did not go through. So, i am re-sending it and CCing it to you so that you could put it up in case it does not go through again.




The lines from your script seem strange. LAMMPS does not have a “fix cold”. And you’ve got two fixes labelled as number 4. So either you’re using a modified version of LAMMPS or these are not the input script lines you used in the actual simulation.

If energy is being removed from part1 instead of added, what happens if you switch the sign on the heat flux? I’d recommend trying this without the momentum fix. I’d be surprised if you see a lot of drift or rotation — you probably don’t need (or really want) to use the momentum fix.


Dear Paul,
I miss wrote the command, I dont know what I ws thinking at that time. Here are the correct ones

fix 4 all momemtum 1 linear 1 1 1 angular
fix 2 part1 heat 10 0.0001
fix 3 part2 heat 10 -0.0001
fix 5 all nve

I am going to try it without momentum fix. I am quite sure that it will work but I am a little cautious about rotation as part1 and part2 are nanotubes and I dont want the energy (heat command) I am providing to go to rotational energy of nanotube. I will let you know what happens.


P.S. From my account, I am getting delivery failure on user-list email address. I am not sure if there is a problem currently with sorgeforce server.