[lammps-users] ecoul elong jumps during deformation

LAMMPS LAMMPS Mailing List Mirror
1

Hello,

I am using fix deform to run tensile testing. I am using pair_style lj/cut/coul/long/opt. I have broken the simulation into 10 fragments to reset the kspace. When plotting the ecoul or elong, the energy jumps at the run breaks. The overall PE does not seem to be affected. Any ideas on how to prevent the energy jump?

I’ve tried to use the run start/stop options and that did not help.

I suspect this has something to do with the kspace grid changing.

Thanks for the help,
Sam

image.png
image.png752×452 5.25 KB

2

Hello,

I am using fix deform to run tensile testing. I am using pair_style lj/cut/coul/long/opt. I have broken the simulation into 10 fragments to reset the kspace. When plotting the ecoul or elong, the energy jumps at the run breaks. The overall PE does not seem to be affected. Any ideas on how to prevent the energy jump?

the electrostatic contribution is the sum of ecoul and elong, so why would it be a problem if the partitioning changes?

I’ve tried to use the run start/stop options and that did not help.

I suspect this has something to do with the kspace grid changing.

you can easily check this by comparing log files.

Ultimately, you have to make a choice:

  • accept that the various ewald parameters are reinitialized but the overall accuracy is (mostly) constant.
  • explicitly specify the ewald parameters with kspace_modify and then have a varying accuracy.
1 Like