Read_restart and FENE Bond Too Long

Dear Dr. Axel Kohlmeyer,

Hello Dr. Kohlmeyer. Thank you for your advice.

I have attached a short LAMMPS script entitled, “Stress_Strain.in” that applies a tensile load to the polymer matrix at a strain rate of 0.01 for 10000 timesteps. After 10000 timesteps, the compressive load is applied to the polymer matrix thereby a cyclic stress strain simulation for a total of 20000 timesteps. This LAMMPS script also outputs restart file every 1000 timesteps.

The attached LAMMPS script, “Stress_Strain.in” runs fine. However, I imagined that the computer crashed during my simulation run, and I want to restart my LAMMPS script from 10000 timesteps (immediately after finishing the tensile loading) and continue the simulation until its final completion at 20000 timesteps thereby finishing the compressive loading.

I have attached a LAMMPS simulation entitled, “Restart_Stress_Strain.in” where the code reads the restart file entitled, “10000_Load_Unload_Restart_File.restart,” and the restart simulation starts at 10000 timesteps and ending at 20000 timesteps. (10000_Load_Unload_Restart_File.restart will be outputted when the “Stress_Strain.in” runs.)

However, when I run my restart LAMMPS code, I am obtaining “FENE bond too long.” when, in fact, the restart simulation from 10000 to 20000 timesteps should replicate the stress strain curve of the second portion of my original stress strain simulation by Stress_Strain.in.

I would greatly appreciate if you could inform me the method in resolving this issue. I have included all of the necessary fix commands in the restart file. Thank you.

Sincerely,

Masato Koizumi

0.0714_rho_Single_HET (221 KB)

Stress_Strain.in (1.58 KB)

Restart_Stress_Strain.in (1.79 KB)

10000_Load_Unload_Restart_File.restart (188 KB)

Dear Dr. Axel Kohlmeyer,

Hello Dr. Kohlmeyer. Thank you for your advice.

I have attached a short LAMMPS script entitled, "*Stress_Strain.in*" that
applies a tensile load to the polymer matrix at a strain rate of 0.01 for
10000 timesteps. After 10000 timesteps, the compressive load is applied to
the polymer matrix thereby a cyclic stress strain simulation for a total of
20000 timesteps. This LAMMPS script also outputs restart file every 1000
timesteps.

The attached LAMMPS script, "*Stress_Strain.in*" runs fine. However, I
imagined that the computer crashed during my simulation run, and I want to
restart my LAMMPS script from 10000 timesteps (immediately after
finishing the tensile loading) and continue the simulation until its final
completion at 20000 timesteps thereby finishing the compressive loading.

I have attached a LAMMPS simulation entitled, "*Restart_Stress_Strain.in*"
where the code reads the restart file entitled, "
*10000_Load_Unload_Restart_File.restart*," and the restart simulation
starts at 10000 timesteps and ending at 20000 timesteps. (
10000_Load_Unload_Restart_File.restart will be outputted when the "Stress_Strain.in"
runs.)

However, when I run my restart LAMMPS code, I am obtaining "FENE bond too
long." when, in fact, the restart simulation from 10000 to 20000 timesteps
should replicate the stress strain curve of the second portion of my
original stress strain simulation by *Stress_Strain.in*.

I would greatly appreciate if you could inform me the method in resolving
this issue. I have included all of the necessary fix commands in the
restart file. Thank you.

​sorry, but it is *your* job to debug *your* inputs. there is plenty of
helpful information in the LAMMPS manual. people here don't have the time
to do your work for you.
the strategy is straightforward:
- simplify your input until it works as ​expected
- then add back features piece by piece and compare the behavior with what
you should expect from what is said in the documentation.
- watch out when restarting in the middle of a gradual change. have a close
look at the documentation for the run command

axel.

Dear Dr. Axel Kohlmeyer,

Hello Dr. Kohlmeyer. Thank you for your e-mail.

I was reading the reset_timestep and the run command in the documentation. Is my logic correct in that when a restart file is captured at, say, 10000 timesteps, I am supposed to specify reset_timestep 10000 so that the simulation will start exactly from the timestep to which the restart file was created?

Regarding the run command, I am understanding that “run 100 start 10000 stop 20000” performs a run from 10000 to 20000 timesteps in an incremental intervals of 100. However, when I run my LAMMPS code with this command, I am obtaining simulation results for only 10000 and 10100 timesteps.

I would greatly appreciate if you could guide me if my understanding is correct. Thank you.

Sincerely,

Masato Koizumi

“is my understanding correct” is just a different way to have other people do your work for you.
you are a scientist, right? apply the scientific method and do figure it out by yourself.
you have more than fully exploited my willingness to help you on such trivial issues that you should be able to resolve yourself.

axel.