Boltzmann distribution over initial conditions

I think you are mixing up some terminology. I would think equilibrated means that both velocity and position have become independent of each other and satisfy equipartition.

In your case it seems more like the (initial) total energy of each particle should be conserved.
It should be relatively simple to produce a data file that has the properties you need with some script or small program. It is probably possible in LAMMPS as well with a combination of per-atom variables, but I don’t know how exactly.

Note that the initial data file always defines
atom positions, and can optionally define atom
velocities. So as Stefan says, you
can create a data file to define exactly
the x,v of every atom if you need to.


Thanks guys! I actually came up with another alternative: I let my particles equilibrate with a terribly high damping.

Do you know if there’s an easy way to change the damping after a certain number of runs? Or do I have

to run the equilibration with a high damping, save my results and do another file where I run with my real damping?

How did you set the damping? You can
just unset it, or set it to something else.