using morse + coul/long

Hi Lammps developers, I have a problem with compiled, what package are
needed for use pair_style hybrid/overlay morse coul/long rep/exp,
I try use the following potential:

V1= D1*[exp-2*alpha1(r-ro1)-2*exp-*alpha1(r-ro1)] + q1*q2/r ->morse+coul/long
V2= D2*exp-alpha2(r-ro2) + q2*q2/r -> rep/exp+coul/long
V3= D2*exp-alpha2(r-ro2) + q1*q1/r -> rep/exp+coul/long

pair_style rep/exp is type morse that modified.

In log.lammps file not write any type of error.

any idea
Thanks

Samuel

in.test (1.94 KB)

lammps.8.o20109 (3.02 KB)

(Attachment log.lammps is missing)

I'm not clear what your Q is. You added
a potential rep/exp to LAMMPS. You are
trying to use it with pair hybrid. You
say you have a "problem", but no error message.

How do you know there is a problem?

Steve

Based on the LAMMPS output file you provided, your simulation has so many problems that it’s not even clear where to begin:

WARNING: System is not charge neutral, net charge = -51.2
G vector = 0.308274
grid = 12 12 12
stencil order = 5
RMS precision = 0.000414596
brick FFT buffer size/proc = 2197 288 2028
ERROR: Cannot use FFTs with FFT_NONE set
ERROR on proc 0: Could not create 3d FFT plan
MPI: Global rank 0 is aborting with error code 1.
Process ID: 7974, Host: n009, Program: /home/onsager/aantone/samuelif/lammps-7Dec10/src/lmp_altix

So you’re running a system that doesn’t conserve charge neutrality, you’ve built LAMMPS without FFT support so you can’t run PPPM, and you’ve tried to use a new potential that doesn’t seem to work with the old ones. It’s little wonder that you’re running into problems!

You need to isolate the problems one by one:

  • Start by building LAMMPS with FFTW support.
  • Correct the input file so that you have a charge-neutral system.
  • Then worry about the new potential.

—AEI