# [lammps-users] No fix, No move?

Hi,

But I think the volume of the system is constraint by the fix NVE. Is this no business with the position/velocity update?
Another question: If I want to add some energy to the system in a period time and then let the system go, what can I do? It’s seemly that there is no ensemble I can use.

Thanks

Hi,

But I think the volume of the system is constraint by the fix NVE. Is this no business with the position/velocity update?

there is no constraint. you just run. please have a deeper look
into one of the textbooks on molecular simulations.

Another question: If I want to add some energy to the system in a period time and then let the system go, what can I do? It's seemly that there is no ensemble I can use.

NVE ensemble doesn't change the energy, if you add a fix
total energy will go up.

in fact, conservation of energy in NVE ensemble is a measure
of how well a simulation is set up (provided no energy is added,
etc.). with too large a time step, bad cutoffs, the energy
will go up, which would be unphysical.

axel.

If you want to add energy to the system, rescale
the velocities with the velocity command.

Steve

2008/10/22 Xiaopeng Huang <[email protected]...>:

``````在2008-10-23，"Steve Plimpton" [[email protected]> 写道：
>If you want to add energy to the system, rescale
>the velocities with the velocity command.
>
>Steve
](mailto:[email protected]> 写道：>If you want to add energy to the system, rescale>the velocities with the velocity command.>>Steve)
``````
``````
At the same time, should I turn off the NVE integration and turn it on after the rescaling? Or I should ask: Does fix NVE(1st fix) mean that the total energy of the system will be always the same as the initial energy? If I add energy to the system, will the total energy of fix NVE be the new one?

Thanks

>2008/10/22 Xiaopeng Huang

``````

The velocity command is a one-time operation. It doesn't matter
what fixes are set. Fix NVE will generally conserve energy, but
there are various ways to change that.

Steve

2008/10/23 Xiaopeng Huang <[email protected]...>: