Hello everybody,
after a bit of struggle I was able to build a model for stick-slip in granular materials (-> file in.shear, altogether with restart.equil).
The attached script pulls a rigid plate (blue) through a yellow ball which moves at constant velocity along positive x-axis. Below there is a granular bulk (green) which lies over a blocked granular bed (red).
Goal would be to obtain stick-slip dynamics for the plate.
I was studying the force between plate and bulk (obtained plotting the fourth column of xcm.dat) vs. the pulling force (obtained as 200*($3-$2) of xcm.dat, which is k*(x_yellow_ball - x_plate)).
Unfortunately, the two forces appear to be different in orders of magnitude. Pretty weird.
output: xcm.dat -> columns: || timestep | x_plate | x_yellow_ball | f_x | f_y | f_z | ??? ||
f_x is the x component of the force b/w plate and bulk
dump output: *.jpg, they will show you what happens
I DO suggest running as lammps-daily -screen none -log none
I will be grateful for any hint - acknowledgements in the paper will be guaranteed 
and I am clearly available for any further explanation
I apologise if the question is too ‘physical’ and does not pertain to LAMMPS mailing list
Best,
Mario
restart.equil (74.1 KB)
in.shear (2.9 KB)