# non-orthogonal box

Dear all,

I need a fully-flexible non-orthogonal box. I think fully flexible box can be obtained by fix npt. Am I right?
Regarding to non-orthogonal box, I have created a non-orthogonal structure (monoclinic, with beta angle not equal to 90) by read_data command. But I see this in the log file: orthogonal box = (-4.32725 -0 -2.43251) to (25.7228 40.22 42.2275) So an orthogonal box encompasses a non-orthogonal structure which has made problems in the simulation.

Can anyone tell me why the box is considered as orthogonal when the atoms are created in non-orthogonal shape?
How can I resolve this problem?
I used the command change_box triclinic ” but I see 0 for all tilt factors by thermo_style.
A part of my data file is below:

LAMMPS data file. CGCMM style. atom_style full generated by VMD/TopoTools v1.1 on Wed Mar 07 21:39:55 CST 2012
4000 atoms
0 bonds
0 angles
0 dihedrals
0 impropers
2 atom types
0 bond types
0 angle types
0 dihedral types
0 improper types
-4.327247 25.722752 xlo xhi
-0.000000 40.220001 ylo yhi
-2.432508 42.227492 zlo zhi

# Pair Coeffs

Dear all,
I need a fully-flexible non-orthogonal box. I think fully flexible box can
be obtained by fix npt. Am I right?
Regarding to non-orthogonal box, I have created a non-orthogonal structure
(monoclinic, with beta angle not equal to 90) by read_data command. But I
see this in the log file: orthogonal box = (-4.32725 -0 -2.43251) to
(25.7228 40.22 42.2275) So an orthogonal box encompasses a non-orthogonal
structure which has made problems in the simulation.
Can anyone tell me why the box is considered as orthogonal when the atoms
are created in non-orthogonal shape?
How can I resolve this problem?
I used the command “ change_box triclinic ” but I see 0 for all
tilt factors by thermo_style.
A part of my data file is below:

the data file does *not* describe a non-orthogonal box.

from the first line of the data file, i can see that you have
used VMD and the topotools plugin to generate the data file.
if you want topotools to *write* non-orthogonal boxes, you
first have to teach it to support this (i.e. by adding the
corresponding Tcl code). i don't have the time and interest
at the moment.

axel.

You need a line in your data file like

1.0 0.0 1.0 xy xz yz

to tell LAMMPS you want a triclinic box.

Steve