Kinetic energy oscillations with granular & fix deform

Dear all,

I’m studying simple shear flow of granular materials in a triperiodic cell by means of “fix deform” with lammps/11Aug17.
The grain particle sizes are uniformly distributed in the range 0.8-1.2. The cell is 20x20x20.
I’ve prepared different volume fractions as in Chialvo et al Phys. Rev. E 85, 021305 (2012).
Simulations run smoothly (on 10 cores) for solid fractions lower than 0.6. However, when phi=0.6 (~9080 particles) I’m not able to reach a steady state, because the center of mass velocity of the system displays large, low frequency, oscillations (see attached plot).

I tried different preparation methods for phi=0.6 ( lammps crystalline packings, random deposition, also “rainfall” with a contact dynamics code to reduce initial interpenetration), but oscillations are still present.

Does someone have an idea of what is happening? Here’s an excerpt of my simulation script :

atom_style sphere
boundary p p p
newton off
comm_modify mode single vel yes

region reg prism 0 20. 0 20. 0 20. 0 0 0 units box
create_box 1 reg

initial packing taken from external file generated externally

read_data file.dat add append

neighbor 0.2 bin
neigh_modify delay 0

pair_style gran/hooke/history 1000000 285714 159 NULL 0.5 1
pair_coeff * *
timestep 0.00002

fix 1 all nve/sphere
compute 1 all erotate/sphere
fix 2 all deform 1 yz erate 0.01 remap v
velocity all ramp vy -0.1 0.1 z 0 20
velocity all zero angular

run 1000000000

Thank you,




You’re right that this is weird - but I’m not that weirded out that its happening so close to jamming. Since there is not full input deck…

Is this kinetic energy only, i.e. 1/2 * total_mass * center_mass_velocity^2, in the center of mass (not going to go through the effort of estimating your particledensity) of the simulation? Are you getting shear banding? (this can happen due to the sliding boundary images) Have you tried looking at chunks (volume averaging)? Whats happening to the volume averaged or global granular temperature? Stress? deformation tensor?