Can someone gett washino/capillary/viscous modell to run?

Submitted by abrahamp on Wed, 10/14/2015 - 15:14

Im trying to use the newly added cohesion modell for capillary forces. I can not manage to get it to compile, I always get an MPI error.
I use the following in file to try it, which is just a modification of one of the tutorials. Could someone please confirm that it is working and that its just not a bug. If it is working then I would ofcourse also appreciate if some can see what Im doing wrong. Im doing my best to figure that out.

## Initialization

units si
atom_style sphere
boundary f f f
communicate single cutoff 5.0 vel yes
newton on

# Declear the Domain
region reg block -0.300 0.300 -0.03 0.03 -0.01 0.5 units box
create_box 1 reg

## setup

#Material and interaction properties

fix m1 all property/global youngsModulus peratomtype 5e7
fix m2 all property/global poissonsRatio peratomtype 0.32
fix m3 all property/global coefficientRestitution peratomtypepair 1 0.9
fix m4 all property/global coefficientFriction peratomtypepair 1 0.1
fix m5 all property/global minSeparationDistanceRatio scalar 1.01
fix m6 all property/global maxSeparationDistanceRatio scalar 1.1
fix m7 all property/global surfaceLiquidContentInitial scalar 0.00001
fix m8 all property/global surfaceTension scalar 0.07564
fix m9 all property/global fluidViscosity scalar 0.0013
fix m10 all property/global contactAngle peratomtype 12.0

# Physics
pair_style gran model hertz tangential history cohesion washino/capillary/viscous
pair_coeff * *

..... It continues, but running line by line it gives the MPI error at the pair_style line.
Best regards
Per

ckloss's picture

ckloss | Tue, 10/27/2015 - 14:37

Hi Per,

which error are you getting? We always run a series of tests (test harness) before every release, so on our end the model works fine

Christoph

abrahamp | Tue, 11/03/2015 - 13:47

Hi Christoph

I now managed to get it to work, and the issue was in installation of the new updated version while stil having the old version installed.
I was just looking for a conformation that there was no bug in the verry newly added model. Sorry for that mistake and Im happy that the model is added in LIGGGHTS now.

Thank you for the answear
/Per

giuraso's picture

giuraso | Tue, 11/03/2015 - 14:36

Hi all,
Running a script with washino or easo cohesion model was quite smooth. Great job!
However, from a first check, simulating a collision of two particles with Washino and Easo models the velocity after the impact is higher than the initial velocity... still looking at it.
Moreover, if you try to dump local properties at contact (force, pos, id) you get Segmentation fault.

I will try to post my test case. If someone has time to run a simulation of two colliding particles we can have a double check.

Regards,
Giuseppe

aaigner's picture

aaigner | Wed, 11/04/2015 - 10:02

Hi Giuseppe,

a testcase would be great. Than I can have a look on it.

Bests
Andreas