force network for the granular packings

LAMMPS LAMMPS Mailing List Mirror
1

Dear users,

I am trying to find a way to make a force network image between the granular particles.
I have been reading the lammps manual, but still I was not able to figure this out.
I just use the following lammps code to make velocity (vx) snap shot image of the
particles.

dump 6 all image ${irunf} image.normaforce.*.png vx diameter zoom 1.5
dump_modify 6 backcolor white pad 9 amap min max cf 1 2 min red max blue

I need to take the every neighbor pair in the packing and represent the magnitude of the contact
forces (say, normal force) by the thickness of the force network lines. I know how to dump out
all the force details with the neighbor list into a dump file. But I still couldn’t figure out
a way to make an image for the force network.

Can anyone give me a clue ?

Thanks in advance.

Nira.

2

One of my former group members actually asked this long ago, to little response (http://lammps.sandia.gov/threads/msg06377.html).

Unfortunately, there isn’t any out of the box way of doing this (though I am admittedly clumsy on utilizing lammps commands to their full potential). What you need of course are the particle positions, center to center vectors for contacting particles, and interparticle force magnitudes. The first is straightforward to collect with the dump command. The last you can obtain via compute pair/local. However, the center to center vector (as opposed to distance) can’t be obtained in this way.

Nontheless, if you can read and manipulate the interparticle forces, locations of particles, and radii with the language of your choice - everything you need should be trivial to compute. Then you just have to manipulate the info into a form your chosen visualization program can read. Shouldn’t be too tough - and it’s a nice exercise if you are just starting out.

3

Dear Eric,

Thank you so much for your reply with lot of valuable information. I can try the way you explained. It makes more sense to me.

Thanks again,

Nira.

4

Also, forgot that you can get the center to center vectors, since the fx, fy, and fz components only contain the normal forces for the granular models. Just construct the unit vector for the normal forces and multiply by the distance.

5

Thanks again, Eric. Appreciated!

Nira.