Doubt in the correctness of fene.cpp

LAMMPS LAMMPS Mailing List Mirror
1

Hello,

I stumbled across the following lines in fene.cpp (March 31, 2017):

fbond = -k[type]/rlogarg; (line 96)

fforce = -k[type]/rlogarg; (line 265)

Shouldn’t there be a multiplier of sqrt(rsq) in the following form :

rsq = delxdelx + delydely + delz*delz;

fbond=-k[type]*sqrt(rsq)/rlogarg;

and vice versa

Thanks.

Sonu

2

Hello,

I stumbled across the following lines in fene.cpp (March 31, 2017):

    fbond = -k[type]/rlogarg; (line 96)

  fforce = -k[type]/rlogarg; (line 265)

Shouldn't there be a multiplier of sqrt(rsq) in the following form :

rsq = delx*delx + dely*dely + delz*delz;

fbond=-k[type]*sqrt(rsq)/rlogarg;

and vice versa

​what LAMMPS computes here is not F_ij, but F_ij/r_ij, so that it can be
easily converted into F_x, F_y, F_z by multiplication with delta_x,
delta_y, and delta_z, respectively.​

​axel.
​