Matching compute temp/profile with compute temp/deform

LAMMPS LAMMPS General Discussion
1

Dear LAMMPS users,

I am attempting to perform an NEMD shear flow simulation of dumbbell particles using fix nvt and fix deform. The dumbbells are comprised of a linear spring and two spherical particles.

I have attached a minimal script with the appropriate molecule file. The lammps version used is lammps-2Aug2023 compiled in Linux.
in.nvt_with_hookean_DB_particle_lang_eq_minimal (3.1 KB)
molecule.Old_B_db (182 Bytes)

Does anyone have any advice how to get compute temp/profile and compute temp/deform to match closely? At all flow rates considered there seems to be a systematic difference between the temperature computed from compute temp/profile and compute temp/deform.

Thanks,
Luke Debono

2

Why do you think they should match?

3

I’m using this as a test to determine whether the particles are streaming according to the box deformation, as stated in the documentation for compute temp/deform.