One or more atoms are time integrated more than once

LAMMPS LAMMPS Mailing List Mirror
1

Hi all,

I have the following problem. I use the following commands in order to simulate my system using the Langevin thermostat and a barostat:

fix 1 all rigid molecule langevin 2.00 2.00 0.50 48279
fix 2 all nph iso 0.05 0.05 1000.0

For that I take the error:

One or more atoms are time integrated more than once (../modify.cpp:245)

Could you please explain me my fault about the commands?

Thank you so much!

Regards,
Nikoletta

2

The problem is pretty much in the error message: You are time integrating your atoms more than once. That is, both fix rigid and fix nph adjust the positions and velocities of the atoms in your box in a single time step, and this is not the way it should be.

To fix it, you can use the nph-variant of fix rigid. I am not sure if that supports a langevin-style thermostat directly or if you should add a fix langevin separately, see http://lammps.sandia.gov/doc/fix_rigid.html for more info.

3

Hi Nikoletta,

the documentation for fix rigid (http://lammps.sandia.gov/doc/fix_rigid.html) says:

"You should not update the atoms in rigid bodies via other time-integration fixes (e.g. fix nve, fix nvt, fix npt), or you will be integrating their motion more than once each timestep. When performing a hybrid simulation with some atoms in rigid bodies, and some not, a separate time integration fix like fix nve or fix nvt should be used for the non-rigid particles."

so I suppose, in fix 2 nph instead of all you should include all the rest of fix 1.

Best,
Michalis

4

Thank you both you!

I think I understand my fault. So, I changed the commands to:

fix 1 all rigid/nph molecule iso 0.05 0.05 1000.0
fix 2 all langevin 2.00 2.00 0.50 48279

Now, the results are good and the previous error is not exist any more.

Regards,
Nikoletta

5

Hi Nikoletta,

I am not sure whether you solved your problem. I am under the impression that if you visualize the trajectory, the non-molecule particles will be frozen and this is because you do not integrate the equation of their motion. I don't know if this is correct, but possibly you have to add an integrator for the rest as well:

fix 3 nomolecule nph iso 0.05 0.05 1000.0

I might be wrong though.

Best,
Michalis

6

Hi Nikoletta,

I am not sure whether you solved your problem. I am under the impression that if you visualize the trajectory, the non-molecule particles will be frozen and this is because you do not integrate the equation of their motion. I don’t know if this is correct, but possibly you have to add an integrator for the rest as well:

fix 3 nomolecule nph iso 0.05 0.05 1000.0

I might be wrong though.

You are.

Particularly, having two integrators that want to change the cell dimensions is a mistake.

Axel

7

Dear all,

Thank you for your help! I appreciate it.

Regards,
Nikoletta