Shock introduced using velocity command due to high strain rates

LAMMPS LAMMPS Mailing List Mirror
1

Dear All,

I have a doubt… If i want to apply shear on the upper region of my simulation cell by velocity command, and without RAMPING,to simulate processes like Dislocation motion, how do i do this…? I think we should allow the shock to settle in the whole domain by time integrating the atoms.

My domain contains three regions, upper, mobile and lower.

I am attaching a part of my script for the same. But it doesnt seem to work correctly. Can i get the example script of someone who has simulated edge dislocation motion.

2

The velocity command changes the velocity of atoms all at once, to
whatever values you set it. This happens before (or between) a simulation
run. If you want to change velocities gradually during a run while time
integrating the rest of the system, then you want a command like
fix move. It allows for the velocity to be set on an atom-by-atom basis (if
you choose) using a formula you define, so you can do whatever you like.

Steve

3

Thanks for the reply Steve…

But the documentation clearly says that “”"“IMPORTANT NOTE: The atoms affected by this fix should not normally be time integrated by other fixes (e.g.
fix nve, fix nvt), since that will change their positions and velocities twice.”""" . Then how to achieve the same… i am a beginner, so please help…!!!

Sagar

4

Thanks for the reply steve…

Please ignore my previous question regarding fix_move…

i will apply the fix_move command in upper region atoms, and time integrate the mobile region ones… i am planning to use fix_nvt for that… Am i correct…? Or should i use nve and apply temp. rescaling consecutively…??

Thanks for suggestions…!!!

5

The atoms that you specify for fix move should not
be otherwise integrated. Fix move does that. For
the other mobile atoms its up to you how you thermostat
them.

Steve

6

Thanks for the reply Steve. Much appreciated Sir.

Sagar