[lammps-users] Energy not conserved in an NVE sim .. Included Cu simulation results too..

Dear all,

First of all apologies for the several posts (with incremental
corrections) in one day.

I have carried out NVE simulations of a single RDX molecule (by
slightly modifying the reax example with lammps Jul 21st, 2010
version). The total energy is not conserved during the NVE simulations
and shows a trend. The RDX molecule does not disintegrate and even if
it does I do not think energy conservation should be affected. I am
attaching the input file and the image file of the total energy as a
function of time.

To check if this is a problem with all potentials, I repeated the NVE
simulation with a 5x5x5 unit cell of Cu with the EAM potential
(Cu_u3.eam). I am attaching the image file of total energy of Cu as a
function of time with this mail and also my input file for the Cu
simulation. The total energy fluctuates in the "fourth decimal" in
this case and I am not sure if one can say there is a trend unless one
does longer simulations. Is it OK to have this much fluctuation, even
though the energy conservation is much better than the reaxff
potential cases?

Than you in advance.

in.reax.rdx (508 Bytes)


in.cu_econserve (579 Bytes)


Hi Manoj,

I think it would be more informative if you plot the temperature
rather than the total energy which has a potential-dependent 0K
potential energy component.