Strange behavior when using dynamic rotating region

Hello lammps-users,
I’m currently trying to use a rotated region to remove a rod of atoms in a simulation box with a rotated lattice containing a dislocation with the glide plane perpendicular to y.
But for me there seems to be stange behavior in the rotation of regions. For describing my problem I created a minimum working example:

clear
units metal
dimension 3
boundary s s s
atom_style atomic
timestep 0.001

lattice fcc 4.05
region box block 0 200 0 200 0 200 units box
create_box 1 box

variable p_ang equal $(0)

axis || y

region prm block 90 110 90 110 0 200 units box rotate v_p_ang 100 100 100 0 1 0

variable idx equal $(0)

label lpa
create_atoms 1 region prm

write_dump all custom dump.{idx}.cfg id type mass xs ys zs delete_atoms region prm variable p_ang equal (v_p_ang + PI/50.0)
variable idx equal (v_idx + 1) if "{idx} < 200" then “jump mwe.in lpa” else “quit”

(I removed the superfluous variables and used fixed values)

This should rotate a rod (simplified; orignial is monoclinic) around the y-axis.
One basic thing I rember from rotations is, that a rotation around the y-axis cannot affect the y-values of the region.
But if you look at the individual snapshots, you see not only a rotation around y, but also some kind of movement in y direction.

So my question is: What am I missing? Or is this a bug?
Because normally the origin of the rotation – in case I chose the wrong value – can also not affect the y location of the rod.

Thanks and bye,
Dennis Weber

Hello lammps-users,
I’m currently trying to use a rotated region to remove a rod of atoms in a simulation box with a rotated lattice containing a dislocation with the glide plane perpendicular to y.
But for me there seems to be stange behavior in the rotation of regions. For describing my problem I created a minimum working example:

Are you using the very latest patchlevel of lammps? There was a recent bugfix for rotated regions.

Axel