← Back to team overview

yade-users team mailing list archive

Re: Strange behavior of periodic cell

 

> I'm trying to do some simulations using periodic cell. In one case, I prescribed some extension of the periodic cell, but the particles in the simulation behave really strange - the dispacement of particles si continuously increasing.

I had a look again. It looks as if kinetic energy is increasing, even
with damping=0.9.

I tried with Law2_Dem3DofGeom_FrictPhys_Basic and also
Law2_ScGeom_FrictPhys_Basic (instead of the Cpm law) and the result is
always the same: spheres bumping into each other going crazy after a
short while. What is happening??

Can someone have a look as well? Especially tracking stored elastic and
kinetic energy could be useful. (there is no plasticity involved) The
script is attached (run with -j1, it might be faster).

Weird thing.

Václav



from yade import pack, log, qt
import itertools

# create material
O.materials.append(CpmMat(young=1e7,epsCrackOnset=20,relDuctility=3,sigmaT=3e30,frictionAngle=.5,G_over_E=.2))

O.periodic=True
# create 3x3x3 particles
a=1.; r=a/6.
for i,j,k in itertools.product(range(3),repeat=3):
	O.bodies.append(utils.sphere([(2*i-1)*r,(2*j-1)*r,(2*k-1)*r],r))
O.cell.refSize=(a,a,a)

O.engines=[
	BoundDispatcher([Bo1_Sphere_Aabb()]),
	InsertionSortCollider(),
	InteractionDispatchers(
		[Ig2_Sphere_Sphere_ScGeom()],[Ip2_FrictMat_FrictMat_FrictPhys()],[Law2_ScGeom_FrictPhys_Basic()]),
		## the same with other consitutive laws:
		# [Ig2_Sphere_Sphere_Dem3DofGeom(distFactor=-1)],[Ip2_CpmMat_CpmMat_CpmPhys(cohesiveThresholdIter=-1)],[Law2_Dem3DofGeom_CpmPhys_Cpm()]),
		# [Ig2_Sphere_Sphere_Dem3DofGeom()],[Ip2_FrictMat_FrictMat_FrictPhys()],[Law2_Dem3DofGeom_FrictPhys_Basic()]),
	NewtonIntegrator(damping=0.8)
]

# forced compression of periodic cell along x
O.cell.trsf+=Matrix3(-.01,0,0, 0,0,0, 0,0,0)
O.dt=1e-9


Follow ups

References