changing radius results in ERROR: Cannot use neighbor bins - box size << cutoff (../neighbor.cpp:1737)

Hi there,

I am running a 2 dimension problem, several solid particles in liquid. The “bug” is like that: if the atom radius is 3.0, the program runs fine and the result is good, if I change the radius from 3.0 to 4.0 or larger, the “ERROR” come out. Below is the input file and the ERROR message, is there anyone can give me some clue why does thit happen, or this is a bug in LAMMPS? Thank you.

dimension 2
units lj
atom_style sphere
boundary p f p
newton off
comm_modify vel yes
region box block -11 11 -21 25 -0.01 0.01
region space block -10 10 -20 20 -0.01 0.01
region temp block -10 10 -20 20 -0.01 0.01 side out
region wall intersect 2 temp box
create_box 4 box
create_atoms 1 random 10 880503 space
create_atoms 2 random 10 891025 space
lattice sq 2
create_atoms 3 region space
group al type 1
group HDPE type 2
group ethanol type 3
group granular union al HDPE

set type 1 diameter 3.0 #if diameter is 4.0, Error
set type 2 diameter 3.0 #if diameter is 4.0, Error
set type 3 diameter 1.0

set type 1 density 13.683143219264894
set type 2 density 4.765524822584579
set type 3 density 1.909859317102744
fix 82 all wall/gran 5000.0 NULL 1.0 NULL 1 0 yplane -20 21
fix 99 all enforce2d
pair_style hybrid gran/hooke/history 500.0 NULL 100.0 NULL 1 0 dpd 1.0 1.0 22717
pair_coeff * * dpd 18.75 4.5 1.0
pair_coeff 12 3 dpd 500 4.5 2.0
pair_coeff 1
2 1*2 gran/hooke/history
neighbor 0.3 bin
neigh_modify delay 5 every 1
timestep 0.002
dump 1 all custom 200 ovito.lammps id type xs ys zs radius
fix 404 all nve/limit 0.003
run 10000
fix 3 all nve
variable gravity equal -0.15
fix 2 granular gravity v_gravity vector 0 1 0
run 40000

ERROR message:
Neighbor list info …
6 neighbor list requests
update every 1 steps, delay 5 steps, check yes
max neighbors/atom: 2000, page size: 100000
master list distance cutoff = 4.3
ghost atom cutoff = 4.3
binsize = 2.15, bins = 11 22 1
Setting up Verlet run …
Unit style : lj
Current step: 0
Time step : 0.002

Hi there,

I am running a 2 dimension problem, several solid particles in liquid. The “bug” is like that: if the atom radius is 3.0, the program runs fine and the result is good, if I change the radius from 3.0 to 4.0 or larger, the “ERROR” come out. Below is the input file and the ERROR message, is there anyone can give me some clue why does thit happen, or this is a bug in LAMMPS? Thank you.

dimension 2
units lj
atom_style sphere
boundary p f p
newton off
comm_modify vel yes
region box block -11 11 -21 25 -0.01 0.01
region space block -10 10 -20 20 -0.01 0.01
region temp block -10 10 -20 20 -0.01 0.01 side out
region wall intersect 2 temp box
create_box 4 box
create_atoms 1 random 10 880503 space
create_atoms 2 random 10 891025 space
lattice sq 2
create_atoms 3 region space
group al type 1
group HDPE type 2
group ethanol type 3
group granular union al HDPE

set type 1 diameter 3.0 #if diameter is 4.0, Error
set type 2 diameter 3.0 #if diameter is 4.0, Error
set type 3 diameter 1.0

set type 1 density 13.683143219264894
set type 2 density 4.765524822584579
set type 3 density 1.909859317102744
fix 82 all wall/gran 5000.0 NULL 1.0 NULL 1 0 yplane -20 21
fix 99 all enforce2d
pair_style hybrid gran/hooke/history 500.0 NULL 100.0 NULL 1 0 dpd 1.0 1.0 22717
pair_coeff * * dpd 18.75 4.5 1.0
pair_coeff 12 3 dpd 500 4.5 2.0
pair_coeff 1
2 1*2 gran/hooke/history
neighbor 0.3 bin
neigh_modify delay 5 every 1
timestep 0.002
dump 1 all custom 200 ovito.lammps id type xs ys zs radius
fix 404 all nve/limit 0.003
run 10000
fix 3 all nve
variable gravity equal -0.15
fix 2 granular gravity v_gravity vector 0 1 0
run 40000

ERROR message:
Neighbor list info …
6 neighbor list requests
update every 1 steps, delay 5 steps, check yes
max neighbors/atom: 2000, page size: 100000
master list distance cutoff = 4.3
ghost atom cutoff = 4.3
binsize = 2.15, bins = 11 22 1
Setting up Verlet run …
Unit style : lj
Current step: 0
Time step : 0.002

Hi there,

I am running a 2 dimension problem, several solid particles in liquid. The
"bug" is like that: if the atom radius is 3.0, the program runs fine and the
result is good, if I change the radius from 3.0 to 4.0 or larger, the
"ERROR" come out. Below is the input file and the ERROR message, is there
anyone can give me some clue why does thit happen, or this is a bug in
LAMMPS? Thank you.

you have to either simulate a larger system, or use the "nsq" method
to generate the neighbor list instead of "bin".

please note that this input also warns about multiple time
integration, so likely one of your time integration fixes needs to be
unfixed or changed.

axel.