particles disappear from the packing

Submitted by prasantud on Mon, 10/24/2011 - 13:17

Hi everybody,
I am making a packing from the standard read_data file in LIGGGHTS example. The particles are falling under gravity.
I am facing problems similar to the particles disappear problem at node/456

I have also increased the spacing of wall to avoid touching with the wall. But the particles are disappearing faster with this.

Disappearance starts when there are many neighbors.

Any comments...

Best regards,
Samantaray
###################################
read_data file: Ordored_Liggghts_long.sys

LIGGGHTS Description

1687 atoms

1 atom types
0 bond types
0 angle types

-0.0500000 1.0500000 xlo xhi
-0.0500000 1.0500000 ylo yhi
-0.0500000 1.0400000 zlo zhi
#######################
file in.packing

units si

atom_style granular
atom_modify map array
boundary f p f
newton off

# Read data
read_data Ordored_Liggghts_long.sys

set group all diameter 0.10

# Some grouping (useful for fixing)
group A type 1

communicate single vel yes

neighbor 0.001 bin
neigh_modify delay 0

#Material properties required for new pair styles

fix m1 all property/global youngsModulus peratomtype 1.e5
fix m2 all property/global poissonsRatio peratomtype 0.3
fix m3 all property/global coefficientRestitution peratomtypepair 1 0.9
fix m4 all property/global coefficientFriction peratomtypepair 1 0.05
fix m5 all property/global characteristicVelocity scalar 2.
fix m6 all property/global cohesionEnergyDensity peratomtypepair 1 200

#New pair style
pair_style gran/hertz/history 1 1 #Hertzian with cohesion
pair_coeff * *

timestep 0.000002
variable dump_every equal 1000

fix gravi all gravity 9.81 vector 0.0 0.0 -1.0
fix xwall1 all wall/gran/hertz/history 1 0 xplane -0.05 -0.05 1
fix xwall2 all wall/gran/hertz/history 1 0 xplane 1.05 1.05 1
fix zwall1 all wall/gran/hertz/history 1 0 zplane -0.05 -0.05 1
fix zwall2 all wall/gran/hertz/history 1 0 zplane 1.04 1.04 1

fix integ all nve/sphere
compute str all ke
compute_modify thermo_temp dynamic yes
thermo_style custom step atoms ke
thermo ${dump_every}
thermo_modify lost ignore norm no
run 1
run 800000 upto

#####end of file######

output is as follows
The

149000 1687 3362.5506
150000 1685 3340.999
150000 1685 3340.999
INFO: more than 15 touching neighbor atoms found, growing contact history.
151000 1685 3317.0417
152000 1685 3291.7887
153000 1685 3265.5073
154000 1683 3239.073
155000 1683 3212.684
156000 1683 3185.5881
157000 1683 3156.75
158000 1683 3125.2994
159000 1683 3090.683
160000 1681 3052.8409
161000 1677 3012.8147

250000 732 426.10796

Forums:

msbentley's picture

msbentley | Mon, 10/24/2011 - 13:40

I'm not sure if this is your problem, but it looks like you've double defined the walls:

fix xwall1 all wall/gran/hertz/history 1 0 xplane -0.05 -0.05 1
fix xwall2 all wall/gran/hertz/history 1 0 xplane 1.05 1.05 1

The format of fix wall/gran allows you to create both "high" and "low" walls for a given axis in one command, e.g.

fix xwall all wall/gran/hertz/history 1 0 xplane -0.05 1.05 1

or you can specify "NULL" if you only want one wall, rather than a pair, e.g.

fix xwall1 all wall/gran/hertz/history 1 0 xplane -0.05 NULL 1

As I say, I don't know if this is the problem - but see if it helps!

Cheers, Mark

prasantud | Mon, 10/24/2011 - 14:05

Hi Mark,
I changed fix wall/gran. But particles continue to disappear.
Is it so hard to make a packing from read_data file or lattice file and let them fall through gravity?
I would have thought that the compression, pour, insert_pack are more difficult in making a simple packing where particles would be at rest(ke=0) for further study.

#new walls
fix xwall1 all wall/gran/hertz/history 1 0 xplane -0.05 1.05 1
fix zwall1 all wall/gran/hertz/history 1 0 zplane -0.05 1.04 1

the output is now.....

565000 824 13.748121

Any more suggestions.
regards,
Samantaray

ckloss's picture

ckloss | Mon, 10/24/2011 - 14:09

>>Is it so hard to make a packing from read_data file or lattice file and let them fall through gravity?
not really, pretty straight-forward

>> m1 all property/global youngsModulus peratomtype 1.e5
that's far too low...

generally, I would advise you to change the tutorials step by step and make a quick check of the result after every change you make. you can then narrow down the problem which will save both you and us time looking for the problem :-)

Christoph

prasantud | Mon, 10/24/2011 - 15:46

Hi Mark,
I have used the Ordored_Liggghts_long.sys file used for the LIGGGHTS.
I have kept the dia as 0.1 metre.

Is there a simple way to compute from the read_data file regarding the contacts or overlapping of the particles.
regards,
Samantaray