Hi,
I have a granular silo and spherical grains(2d ,monodisperse) are poured in the silo upto a certain height and ample time is given for the particles to settle and consolidate under gravity in the silo(Ke is 10e-6 per particle. When i visualize the system in vmd ,I notice there is a slight oscillatory (up and down) movement of particles as a whole mass in the silo.Also there is a incremental decrease in areafraction(2d) of particles from bottom to top. what could be going wrong here? I am clueless. Is it something to do with the elasticity(stiffness)of particles. my Kn value is 250000 and Kt is 71428. unit is cgs. Am i missing any relaxations to be done in between the timesteps I record? The problem persists when the hopper orifices are opened and particles start to flow.
Thank you
Regards
Amit
Hi,
I have a granular silo and spherical grains(2d ,monodisperse) are
poured in the silo upto a certain height and ample time is given for the
particles to settle and consolidate under gravity in the silo(Ke is 10e-6
per particle. When i visualize the system in vmd ,I notice there is a slight
oscillatory (up and down) movement of particles as a whole mass in the
silo.Also there is a incremental decrease in areafraction(2d) of particles
from bottom to top. what could be going wrong here? I am clueless. Is it
something to do with the elasticity(stiffness)of particles. my Kn value is
250000 and Kt is 71428. unit is cgs. Am i missing any relaxations to be done
in between the timesteps I record? The problem persists when the hopper
orifices are opened and particles start to flow.
have you checked under your desk? perhaps there is a troll sitting and
shaking your screen.
but seriously, there is not a lot that people can do via a mailing
list with such a description.
if you are certain, that you know what you are doing and that LAMMPS
is not executing your input correctly, then you have to provide a
complete(!) set of input files that reproduce this behavior. the input
has to be structured in such a way, that the simulation can be run
quickly on a small desktop system. it doesn't have to be realistic,
just reproduce the suspected error.
if you are *not* certain that you know what you are doing, you should
to talk to somebody *local*, e.g. advisers, more experienced
colleagues, collaborators and discuss your understanding of the
physics of the problem. occasionally, the problem is easy to spot
(again, a simple input deck that reproduces it helps much more than a
vague description) and then people here can provide a response, but in
the vast majority of the cases, it is not possible to make up for lack
of training/advising via e-mails. it is far to time consuming and
distracting from solving *real* LAMMPS problems.
you can take this as a general rule: the more somebody else has to do
to figure out what your problem is, the less likely you are to get
help. and this relation usually goes exponential with the amount of
effort.
axel.
Something is wrong if the particles don’t come to rest after
you stop pouring and they just sit there. The granular potentials
are dissipative, which means the kinectic energy of the system
should go to 0.0. It’s possible, especially with cgs units, that
you have messed up some parameters (e.g. the damping term
is way too small or the timetsep too big), but what I’ve told you is the physics of
a granular model.
You should also verify this isn’t some viz artifact by checking a dump
file that your particles continue to move.
Steve