I am trying to analyze the effect of packing style on the granular motion. A particular region needs to be filled by packed bed of specific style (bcc, fcc, hcp).
Is there any way or inbuilt command in LIGGGHTS to handle this task? On the other way, I think it is possible to replace the particle coordinates in dump file with the coordinates (that suits the specific style of packing)
generated by third-party package. But still, if there is a simple way to do the same?
richti83 | Fri, 10/18/2013 - 21:33
I've no clear idea what you
I've no clear idea what you are talking about, but google said you are looking for something like this:
After that I googled "lammps body-centered-qubic" and found:
http://lammps.sandia.gov/doc/lattice.html
well, I think this feature is build-in allready / inherited :-)
region bc block -0.08 0.05 -0.05 0.05 0.1 0.15 units box
#or for complex regions create a tetmesh and use
#region bc mesh/tet file volume_mesh_fine.vtk scale 0.001 move 0. 0. 0. rotate 0. 0. 0. units box
lattice sc 0.005 #style = none or sc or bcc or fcc or hcp or diamond or sq or sq2 or hex or custom
create_atoms 1 region bc
Hope that helps !
rahulsoni | Mon, 10/21/2013 - 14:06
Problem in detail
I have a drum mixer with its axis in horizontal plane. The particles which will fill the drum by 75% needs to be arranged in the styles such as random, bcc, fcc, hcp. After filling the particles are distinguished in two halves and are colored red and green. Then it starts rotating and the mixing pattern is the subject of inspection.
Please see the video to explain it better:
https://www.youtube.com/watch?v=AaTSbjtV5dQ
The problem is I can fill it in random mode but I have no idea how to fill it with specific style of packing.
rahulsoni | Tue, 10/22/2013 - 15:33
create_box & lattice
Hello Richter
I have understood what you suggested. I have gone through the command description and it is really amazing to find this inherent feature. I am attempting to fill my drum with specified packing. I am planning to use the region as cylinder and then use it for lattice creation. So that particles can be created on those lattice points. The problem is that it will have the lattice throughout the drum. However, I want particles up to certain height inside the cylinder.
In short I am asking that is there any way to customize/cutoff the cylindrical region so that only that can be filled with particles.
One more thing in lattice or create_atoms commands I have not seen any specifications for the particle diameter. I know it can be visualized in paraview by entering the diameter. But, then what bothers me is that how the simulation will take care the interaction of particles (force calculation between particles; cutoff/skin distance etc.)
rahulsoni | Tue, 10/22/2013 - 19:17
Cylindrical region
I think I have got the answer for partial filling for cylindrical region. I have first filled the full cylindrical region and then created new block style region in the volume where I do not need the particles. Thereafter, the delete_atoms command for new region has deleted the unwanted particles.
But yet one query remains answered that how to incorporate particle diameter in simulation so that the forces between particles can be taken care.
richti83 | Tue, 10/22/2013 - 19:21
for first question: read the
for first question: read the region command html help file (I would suggest to intersect a cylinder with one or more boxes)
for 2nd: you can set the diameter for every atom by the set command. I would make a loop over all particles and set lets say every atom-ID modulo 9 to another radius to infect the bcc-atom (or every id mod 5 for the fcc atom). its a little bit trial and error but I think it is possible ...
rahulsoni | Wed, 10/23/2013 - 05:56
Particle diameter
Thanks for your reply Richter
What happens if I do not use the set command to set diameters. What are the default diameters. Are they the maximum possible diameter by which the maximum possible volume of the unit cell can be occupied. For example, fcc has an standard packing fraction of 0.74 so if I am making unit cells of dimension to 1 cm then to obtain 0.74 packing fraction the diameter of the particle should be 1/sqrt(2) cm.
Please correct if I am wrong.
PaulWinkler | Mon, 11/04/2013 - 16:29
Hi,
Hi,
some time ago I had done some tests with the lattice command to get an ideal dense packing for monodisperse spheres. I got my stuff done, but you have to think in a crystallography way. Since the lattice command comes from LAMMPS you have to leave the granular thinking of LIGGGHTS. But with lattice and create_atoms you are able to build perfect packings with different atom types. You have a lot to try...
Regards
Paul
rahulsoni | Tue, 11/05/2013 - 07:16
hcp packing
Paul its working for with the same lattice and create_atoms command. I am trying to play with fcc, bcc, hcp and radom style packing. All are working fine except hcp. For all other packing the unit cell is cubic and the unit vectors in all cases possess same direction. However, it is different in case of hcp.
As mentioned in manual, the unit cell vectors for hcp are 1 0 0, 0 sqrt(3) 0, 0 0 sqrt(8/3). This appears peculiar to me as it says that the vectors are all perpendicular to each other which should not be that case of hcp. However, over internet I have seen something like sqrt(2) sqrt(2/3) 0, sqrt(2) -sqrt(2/3) 0 and 0 0 sqrt(8/3) as the cell vectors.
Please clear the facts. Actually, I want to create hcp lattice with atom radius 0.005 m. Can anyone tell me what should be the appropriate scale value in the command lattice hcp scale. The values that I have tried either created the lattice with overlapping spheres or sphere too far to each other resulting in very less packing fraction.