Yade

Soil-Tool Interaction

Asked by Selma on 2014-05-06

Hi,

I am trying to detect interaction forces when depositing an object of conical shape into the soil.
For that purpose I use gravel material for soil and steel for object which is made out of facets. These are the parameters I set up for soil and object, respectively:

gravel = O.materials.append(FrictMat(density=1602,young=120e6,poisson=.25,frictionAngle=.61))
stainless_steel = O.materials.append(FrictMat(density=7740,young=180e9,poisson=.305,frictionAngle=.79))
This is how I set up engines:

O.engines=[
   ForceResetter(),
   InsertionSortCollider([Bo1_Sphere_Aabb(),Bo1_Facet_Aabb()]),
   InteractionLoop(
      [Ig2_Sphere_Sphere_ScGeom(),Ig2_Facet_Sphere_ScGeom()], # handle sphere+sphere and facet+sphere collisions
      [Ip2_FrictMat_FrictMat_FrictPhys(frictAngle=None)],
      [Law2_L3Geom_FrictPhys_ElPerfPl()]
   ),
   NewtonIntegrator(gravity=(0,0,-9.81),damping=.3),
]
# set up time step
O.dt=.5*utils.PWaveTimeStep()

As for moving the object I use PyRunner and in every time step I change its position.

When I try to move the object while in soil (with the aim of finding trajectory with which minimum force would be exerted), the soil becomes unstable, and particles move outside of the bounding box. Everything is ok if the deposition is made as a straight line without any additional movements.
I would like to know why does this happen and how can it be fixed? Is it due to high density of the object, high young modulus? What particular effect has velocity of the object?
I would be very grateful for any further advices on this matter, since I am new to Yade and the matter of soil behaviour and soil models in general.

Thank you,

Selma

Question information

Language:: English Edit question

Status:: Answered

For:: Yade Edit question

Assignee:: No assignee Edit question

Last query:: 2014-05-06

Last reply:: 2014-05-06

Link existing bug

Revision history for this message

Jan Stránský (honzik) said on 2014-05-06:

Hi Selma,

Now we can only guess. To help you properly, we will need to know the way
you move the cone and other things, like your geometry, time step
definition etc.. Would it be possible to send us whole script (if it is not
too long)? In general, instabilities are mostly the consequence of too
large time step. Another reason could be, that what you see is correct, but
the set boundary and initial conditions are non-physical. Again, it would
be very good if we could have your script.

thanks
Jan

Revision history for this message

Jérôme Duriez (jduriez) said on 2014-05-06:

Hello,

> As for moving the object I use PyRunner and in every time step I change its position.

I advice you to not do that. (Tangential) Contact laws are based in Yade on bodies velocities. With such change in position, the contact forces will not reflect the relative displacements. Better set your object as non-dynamic (*), and assign it a velocity that will then be constant because of the non-dynamic feature. NewtonIntegrator engine will take care of changing the position of this body (according to this velocity).

Jérôme

(*) : https://www.yade-dem.org/doc/yade.wrapper.html#yade.wrapper.Body.dynamic