[SOLVED] atom frozen

Hi,

I am trying to simulate a system with two type of particles, the firsts are rectanuglar, and the others are normal spherical particles. The problem is that when I put all particles in my system, only rectangular particles move.
Here is my script, i home someone could help

thank you in advance for your help

Attachment	Size
system2.in_.init_.txt	245 bytes
system2.in_.settings.txt	1.9 KB
system2.data_.txt	598.35 KB

Forums:

LIGGGHTS® - User Forum

kunzhang | Fri, 02/21/2014 - 17:53

From the second file "system2

From the second file "system2.in_.seetings", it seems that only type 2 particles are inserted. Probably, the following insert command format would be helpful (just change ID-group to your own):

fix pts1 droplet particletemplate/sphere 1 atom_type 2 density constant 1000 radius constant 0.004
fix pts2 coreparticle particletemplate/sphere 1 atom_type 1 density constant 2900 radius constant 0.02

fix pdd1 droplet particledistribution/discrete 11250 1 pts1 1.0
fix pdd2 coreparticle particledistribution/discrete 81250 1 pts2 1.0

fix ins coreparticle insert/pack seed 100001 distributiontemplate pdd2 vel constant 0. 0. -2 &
insert_every once overlapcheck no all_in yes region core particles_in_region 4000

fix ins1 droplet insert/stream seed 100001 distributiontemplate pdd1 maxattempt 100 mass 1.0 massrate 1.0 overlapcheck yes vel constant 0. 0. -1.0 insertion_face inface extrude_length 0.1

Also, could you please share your experience on the rectangular particles definition? How to define this shape of particles?

Hope it helps!

Kun

Riccardo Maione | Mon, 02/24/2014 - 08:46

for the rectangluar particle

for the rectangluar particle definition i used moltemplate to make the datafile. However i successfully insert particles, but they didn't move. i attached my datafile to the post.

Regards

PaulWinkler | Mon, 02/24/2014 - 10:45

Hi,

so far as I know, Lammps, Liggghts and most DEM software in general support basically just spherical particles, since the collision detection would else cost to much computational effort. Your data file contains rectangular particles in form of bonded (rigid) spheres. Actually I just can see three bonds between four spheres and all 8400 atoms are bonded to rectangles, no spheres left. Also you are defining two atom types, but type 2 is not in use. After reading the data in, you add some spheres via fix ins but don't integrate them in your simulation. Throw the line 'fix M acier nve/sphere' under the fix ins lines and they will be part of the game.
What should the line 'fix m7 all rigid molecule' do? Why are you 'unfix ins' before the main 'run'?
Thanks for the moltemplate advise, I should read more Lammps stuff actually...

Regards,
Paul

Riccardo Maione | Mon, 02/24/2014 - 10:57

I will basically explain what

I will basically explain what i'm trying to do: there are rectangular particles formed by spherical particles of type 1 in the data file (the bonds exist to implement after the breakage), the second type of particle will be inserted after in the input file. the fix rigid is necessary to calculate the other property for the single rigid body, if not LIGGGHTS would not. the unfix is to insert a determinate number of particles (1400), I unfix it after because the number is achieved.

Regards,
Riccardo

Riccardo Maione | Mon, 02/24/2014 - 10:59

However if you want the

However if you want the moltemplate input file i can send it to you by email

PaulWinkler | Mon, 02/24/2014 - 16:07

Hi Ricardo,

if you use insert_every for just one loop, you may take a look at the 'particles_in_region' key word.