Hi helpful lammps users... I'm having a bit of a problem creating my lattice. Maybe some kind of strange bug? I am doing a cylindrical region, oriented down the y-axis. I'm using a strangely oriented coordinate system, so my axes are x:[1 1 -2], y:[1 1 1], z:[1 -1 0], thus I have adjusted the spacings as well, to get a proper bcc system. My spacing in the y direction is sqrt(3)/2=0.86602540378. When I put all this in however, lammps is creating a lattice that's not equally spaced in the y-direction... here are the y coords of successive planes: 0.0, .824254, 1.64851, 2.47276, 3.29702, 4.12127, 4.94552, 5.76978...
All of those planes have a distance of ~.82425 between them, except the 4th and 5th planes, which have .86942 between them. I looked further out, and the next 10 planes also have the .82425 spacing... but all the spacings SHOULD be .866025... I'm not sure what's going on, but it's making things very difficult, because the lattice won't relax properly. Below is my input, in case anyone wants to try it out, or can see a problem (I get the same thing will all the region commands):
units metal
boundary f p f
atom_style atomic
#this creates the configuration (cylindrical)
lattice bcc 2.855300000 orient x 1 1 -2 orient y 1 1 1 orient z 1 -1 0 spacing 2.44948974278317809 .8660254037844386467637 1.414213562373095049
#need small negatives to avoid roundoff effects cleaving atoms - hopefully it's still periodic enough
#region main cylinder y 0.0 0.0 30.0 -3.891143170298 3.89114317029 units box
#region main cylinder y 0.0 0.0 30.0 0.0 7.7942286340596 units box
region main cylinder y 0.0 0.0 5.0 -0.00000000001 5.0
create_box 1 main
create_atoms 1 region main
dump testspacing all atom 1 dump.testspacing
dump_modify testspacing scale no
pair_style eam/fs
pair_coeff * * Fe_p1.eam.fs Fe
run 0