Dear All,
I received this warning msg when trying to do a cg style minimization on a toy molecule of 4 bonded atoms in a periodic unit cell using lammps-22July2016 built. This toy molecule is part of a bigger molecule.
I have pasted extracts of my input files below. Note that all my atoms are within the unit cell and positioned within one half of this unit cell. The above warning only appears when I include the dihedrals. There is no warning when just used with the bonds and angles.
I probably do not have a good understanding on how to calculate the dihedral extent. Can anyone help?
Thanks,
william
data file:
LAMMPS description
4 atoms
3 bonds
2 angles
1 dihedrals
4 atom types
3 bond types
2 angle types
1 dihedral types
0 8 xlo xhi
0 16 ylo yhi
0 16 zlo zhi
Masses
1 12.01
2 1.008
3 1.008
4 14.01
Atoms
1 1 1 0.771 5.687628 6.227462 2.917555
2 1 4 -1.1 7.052829 6.228613 3.490690
3 1 2 0.023 5.765226 6.226435 1.821131
4 1 3 0.54 7.058474 6.229674 4.532278
Bonds
1 1 1 3
2 2 2 4
3 3 1 2
Angles
1 1 4 2 1
2 2 2 1 3
Dihedrals
1 1 3 1 2 4
Dear All,
I received this warning msg when trying to do a cg style minimization on a
toy molecule of 4 bonded atoms in a periodic unit cell using
lammps-22July2016 built. This toy molecule is part of a bigger molecule.
I have pasted extracts of my input files below. Note that all my atoms are
within the unit cell and positioned within one half of this unit cell. The
above warning only appears when I include the dihedrals. There is no
warning when just used with the bonds and angles.
I probably do not have a good understanding on how to calculate the dihedral
extent. Can anyone help?
you have to look at the source code for that:
LAMMPS will first look for the longest bond bmax in your system and
then compute the max extent as:
1.1*bmax, if you have only bonds, 2.2*bmax, if you have angles, and
3.3*bmax, if you have dihedrals in your system.
in your case, bmax = 1.4806 => for dihedrals the estimated max extent
is 4.886, which is longer than half your box length in x dimension
(0.5*8=4.0). hence the warning.
in this case, it is harmless, and can be easily silenced, by
increasing your box in x direction (e.g. to 0.0 10.0)
axel.
Hi Axel,
Thanks for your clarification!
wee
Dear All,
I received this warning msg when trying to do a cg style minimization on a
toy molecule of 4 bonded atoms in a periodic unit cell using
lammps-22July2016 built. This toy molecule is part of a bigger molecule.
I have pasted extracts of my input files below. Note that all my atoms are
within the unit cell and positioned within one half of this unit cell. The
above warning only appears when I include the dihedrals. There is no
warning when just used with the bonds and angles.
I probably do not have a good understanding on how to calculate the dihedral
extent. Can anyone help?
you have to look at the source code for that:
LAMMPS will first look for the longest bond bmax in your system and
then compute the max extent as:
1.1bmax, if you have only bonds, 2.2bmax, if you have angles, and
3.3*bmax, if you have dihedrals in your system.
in your case, bmax = 1.4806 => for dihedrals the estimated max extent
is 4.886, which is longer than half your box length in x dimension
(0.5*8=4.0). hence the warning.
in this case, it is harmless, and can be easily silenced, by
increasing your box in x direction (e.g. to 0.0 10.0)
axel.