[Patch for 1.5.3] Error message on pair/gran/hertz/history/stiffness

Submitted by willsmithumich on Mon, 06/04/2012 - 20:22

Hi All,

I am trying to use the "simple" version of gran/hertz/history (gran/hertz/history/stiffness) but I am unable to run on the current version on Liggghts (1p5p3). I have provided a simple example code below.

When I run using an older version (1p2p9) it appears to work well, however with 1p5p3 I receive this error:

ERROR: Could not locate a fix/property storing value(s) for gamman_abs as requested by (null)

The problem is that I am using hertz (not hooke) so I should not need gamman_abs. If I define gamman_abs and gammat_abs, I receive this error:

ERROR: Pair gran/hooke/history/simple: Must provide either 'gamman' and 'gammat' or 'gamman_abs' and 'gammat_abs'

This leads me to believe that somewhere in the code it chooses to use hooke instead of hertz.

For completeness, if I provide gamman/t_abs but not gamman/t, then I receive the following error:

ERROR: Could not locate a fix/property storing value(s) for gamman as requested by (null)

I am concerned that Liggghts may be choosing hooke instead of hertz even when not using the "simple" pair type.

I would greatly appreciate any help.

%%% Code %%%
boundary f f f
newton off
communicate single vel yes
units si
echo both
neighbor 0.01 bin
timestep 1e-5

atom_style granular
atom_modify map hash

lattice sc 0.01
region reg block 0.0 0.05 0.0 0.05 0.0 0.825 units box
create_box 1 reg
region insRegion block 0.01 0.04 0.01 0.04 0.01 0.04 units box

pair_style gran/hertz/history/stiffness 0 0
pair_coeff * *

fix gravi all gravity 9.81 vector 0.0 0.0 -1.0
fix xwalls all wall/gran/hertz/history/stiffness 0 0 xplane 0.0 0.05 1
fix ywalls all wall/gran/hertz/history/stiffness 0 0 yplane 0.0 0.05 1
fix zwalls all wall/gran/hertz/history/stiffness 0 0 zplane 0.0 0.825 1

fix m1 all property/global kn peratomtypepair 1 30000
fix m2 all property/global kt peratomtypepair 1 30000
fix m3 all property/global gamman peratomtypepair 1 0.5
fix m4 all property/global gammat peratomtypepair 1 0.5
fix m5 all property/global coefficientFriction peratomtypepair 1 0.5

fix integrateFix all nve/sphere

fix ins all pour/legacy 250 1 100 region insRegion dens 2600. 2600. diam 0.01 0.01

run 100

AttachmentSize
File pair_gran_hooke_history_simple.cpp7.18 KB

Forums:

ckloss's picture

ckloss | Mon, 06/04/2012 - 23:24

Hi will,

thanks for your report. I dared to change the original title of this thread - I hope the explanation below will clarify that

>>I am unable to run on the current version on Liggghts (1p5p3)
Yes, there is an issue which prevents the /stiffness styles to be executed in 1.5.3

>>I am concerned that Liggghts may be choosing hooke instead of hertz even when not using the "simple" pair type.
This is not the case. But the thing is that due to object orientation, if you choose hertz there is always some code from hooke being executed. That was the reason for a recent internal change where the error message identifies itself with the correct style. Unfortunately, a bug slipped in at this point which was the reason why the code did (a) not run although the script was correct and (b) did not identify itself correctly. That will be fixed in the next release (2.0) along with a clearer syntax for the pair styles

I attached a corrected version as a quickfix

Cheers, Christoph

willsmithumich | Tue, 06/05/2012 - 22:15

Hi Christoph,

Thank you for your quick reply and explanation!

oliverio | Wed, 06/13/2012 - 14:34

Hi Chris,
this issue is already fix in the 1p5p3 or we need make copy-past of the attachement file?
Another question, where can we download the 1p2p9 version?
thanks!
Oliver

ckloss's picture

ckloss | Sat, 06/16/2012 - 11:11

The issue can be fixed by copying/pasting the files from this thread into the /src dir of the 1.5.3 version.
Regarding your 2nd question: 1.2.9 is an old version and therefore not longer supported

Cheers, Christoph

oliverio | Mon, 06/18/2012 - 17:04

Hi Cristoph,
I made copy/paste from file and I recompile 1p5p3, it was ok the compiling.
But when I run my script I have same error

#insert the first particles so that dump is not empty
run 1
ERROR: Pair gran/hooke/history/simple: Must provide either 'gamman' and 'gammat' or 'gamman_abs' and 'gammat_abs'

my script:

#TUBO U

units si

atom_style granular
atom_modify map array
boundary m f f
newton off
echo both

#Read Data

read_data initialTube_long_1440.sys
#read_restart restart.equil
communicate single vel yes

region reg block -0.1 0.1 -0.10 0.10 -0.50 0.5 units box
#create_box 2 reg

neighbor 0.1 bin
#neigh_modify delay 0
neigh_modify every 1

#Material properties required for new pair styles

fix m1 all property/global kn peratomtypepair 2 1000 200 200 200

fix m2 all property/global kt peratomtypepair 2 40 40 40 40

fix m3 all property/global gamman_abs peratomtypepair 2 100 50 50 50

fix m4 all property/global gammat_abs peratomtypepair 2 100 100 100 100

fix m5 all property/global gamman peratomtypepair 2 100 50 50 50

fix m6 all property/global gammat peratomtypepair 2 100 100 100 100

fix m7 all property/global coefficientFriction peratomtypepair 2 1.8 1.8 1.8 1.8

#New pair style
pair_style gran/hooke/history/stiffness 1 0 #Hooke without cohesion
#pair_style gran/hertz/history/stiffness 1 0 #Hooke without cohesion
pair_coeff * *

timestep 0.00001

fix gravi all gravity 9.81 vector 0.0 0.0 -1.0

#granular walls

#fix tubo_u all mesh/gran rectangular_big_closed_L-long.stl 1 1. 0.0 0. 0. 0. 0. 0.

#region
group nve_group region reg
region belowU block -INF INF -INF INF -INF -1 units box

#delete grains
delete_atoms region belowU

#apply nve integration to all particles that are inserted as single particles
fix integr nve_group nve/sphere

#output settings, include total thermal energy
thermo_style custom step atoms ke
thermo 5000
thermo_modify lost ignore norm no
#compute_modify thermo_temp dynamic yes

# make dir
fix tubo_u_open all mesh/gran rectangular_big_open_L-long.stl 1 1. 0.0 0. 0. 0. 0. 0.

label cycle

variable r index 1210000 2220000 3230000 4240000 5250000 6260000 7270000 8280000 9290000 10300000 11310000 12320000 13330000 14340000 15350000 16360000 17370000 18380000 19390000 20400000 21410000 22420000 23430000 24440000 25450000 26460000 27470000 28480000 29490000 30500000 31510000 32520000 33530000 34540000 35550000 36560000 37570000 38580000 39590000 40600000 41610000 42620000 43630000 44640000

variable l index 1220000 2230000 3240000 4250000 5260000 6270000 7280000 8290000 9300000 10310000 11320000 12330000 13340000 14350000 15360000 16370000 17380000 18390000 19400000 20410000 21420000 22430000 23440000 24450000 25460000 26470000 27480000 28490000 29500000 30510000 31520000 32530000 33540000 34550000 35560000 36570000 37580000 38590000 39600000 40610000 41620000 42630000 43640000 44650000

#insert the first particles so that dump is not empty
run 1
fix wall all wall/gran/hooke/history/stiffness 1 0 mesh/gran 1 tubo_u_open
dump dmp all custom 10000 post/dump.tuboU id type type x y z ix iy iz vx vy vz fx fy fz omegax omegay omegaz radius
dump_modify dmp append yes
run $r upto
next r
undump dmp
unfix wall
#apply vibration
print $l
fix wall all wall/gran/hooke/history/stiffness 1 0 mesh/gran 1 tubo_u_open
fix vibration all move/mesh/gran wiggle 0 0 0.00323 0.01 tubo_u_open 0.01
dump dmp all custom 10000 post/dump.tuboU id type type x y z ix iy iz vx vy vz fx fy fz omegax omegay omegaz radius
dump_modify dmp append yes
run $l upto
next l
unfix wall
unfix vibration
undump dmp
jump in.tuboU cycle

thanks!
oliver

dianwa | Thu, 07/19/2012 - 23:20

Hi, you said that "But when I run my script I have same error".

Now I have changed the code and recomplied them, while I still can not run the script.

Could you tell me how you did that? Thank you!!

The following is my code:

dimension 3
atom_style granular

# Set the attributes of the boundary
boundary fm fm fm

# No pairwise or bonded interaction
newton off

# The ghost atom for computation
communicate single vel yes

print a
read_data data.test

# Define the parameters in collision
pair_style gran/hooke/history/stiffness 1 0
pair_coeff * *

timestep 1e-7

fix 1 all nve/sphere
fix 2 all gravity 9.81 spherical 0.0 -180.0
fix_modify 2 energy yes

fix kn all property/global kn peratomtypepair 1 280000
fix kt all property/global kt peratomtypepair 1 280000

fix gn all property/global gamman peratomtypepair 1 1
fix gt all property/global gammat peratomtypepair 1 1

fix gnabs all property/global gamman_abs peratomtypepair 1 1
fix gtabs all property/global gammat_abs peratomtypepair 1 1

fix rof all property/global coefficientRollingFriction peratomtypepair 1 0.2

fix xwalls all wall/gran/hooke/history/stiffness 0 0 xplane 0 2 1
fix ywalls all wall/gran/hooke/history/stiffness 0 0 yplane 0 2 1
fix zlower all wall/gran/hooke/history/stiffness 0 0 zplane 0 10 1

compute 1 all erotate/sphere

#thermo_style custom step atoms ke c_1 vol
thermo 100000
thermo_style custom step pe ke etotal vol

# dump name-ID type_of_output time_step name several parameters
dump reverse all custom 500000 stiffres/dump.bottom1.0 mass x y z vz
dump id all xyz 100000 stiffres/wiggle1.0.xyz

label begin
variable b loop 1000000
print This_$b_begining
run 100000

variable V equal etotal

if $V < 5e-23 then "jump in.stiff static"
next b
jump in.stiff begin

label static

dump static all xyz 1 stiffres/wiggle1.0.*.xyz
run 0
dump_modify static every 1000000000

label outloop

variable i loop 1

dump_modify id every 5000
dump_modify reverse every 5000

fix zlower all wall/gran/hooke/history/stiffness 0 0 zplane 0 10 1 wiggle z 0.02 0.1111111

run 555555

fix zlower all wall/gran/hooke/history/stiffness 0 0 zplane 0 10 1

label fview
variable vi loop 200
run 5000
next vi
jump in.stiff fview

print ok
label inloop
variable a loop 100000

print This_$a_loop
dump_modify id every 100000
dump_modify reverse every 100000

run 100000

variable V equal etotal
# print $V
if $V < 5e-23 then "jump in.stiff inbreak"
next a
jump in.stiff inloop

label inbreak

dump final all xyz 1 stiffres/wiggle1.0_*.xyz
run 0
dump_modify final every 10000000000

next i
jump in.stiff outloop

I am a student, who is major in visualization

ckloss's picture

ckloss | Fri, 07/20/2012 - 11:50

Could you please re-run the case with the version 2.0.4?
Bug fixes are made for the latest and greatest version only

Cheers, Christoph

dianwa | Sat, 07/21/2012 - 00:57

So quickly! the new version and your appreciated reply

Thank you very much!

I will try that!

I am a student, who is major in visualization