penetrationDepth calculation

Hello Andrea,
shift2 is used for periodic contact detection, have a look at [1]. See [2] for the meaning of Interaction::cellDist.
cheers
Jan

[1] https://github.com/yade/trunk/blob/master/pkg/common/InteractionLoop.cpp#L113
[2] https://yade-dem.org/doc/formulation.html#collision-detection-in-periodic-cell

Hi,
It does correspond to x1-x2-2*r.
Cheers
Bruno

Thanks Jan and Bruno. From Jan'answer I understand that the variable shift2 should not be a problem in my example (I have no periodicity, it is just a collision between two spheres). From Bruno's answer I learn that there is something wrong in my script where I measure both the quantity overlap=2*r-(x2-x1) (I have x2>x1) and the quantity interaction.geom.penetrationDepth and they do not coincide.

My script is the following. The graph of the two quantity penetrationDepth vs. overlap can be found at this shared link
https://drive.google.com/file/d/1r6YWwbg56jIN3EGB4RzoBRGHgHyHAhir/view?usp=sharing

#!/usr/bin/python
# -*- coding: utf-8 -*-

from yade import pack,ymport,export,geom,bodiesHandling,qt,plot
import math

rad = 0.00774

f=open("forces.dat" , "w")
def forces():
 x1 = O.bodies[0].state.pos[0]
 v1 = O.bodies[0].state.vel[0]
 v2 = O.bodies[1].state.vel[0]
 x2 = O.bodies[1].state.pos[0]
        ov = 2*rad-(x2-x1)
        intrs = O.bodies[0].intrs()
        if (len(intrs)):
         nfx = intrs[0].phys.normalForce[0]
         nfy = intrs[0].phys.normalForce[1]
  nfz = intrs[0].phys.normalForce[2]
  sfx = intrs[0].phys.shearForce[0]
                sfy = intrs[0].phys.shearForce[1]
  sfz = intrs[0].phys.normalForce[2]
                dep = intrs[0].geom.penetrationDepth

        else:
                nfx = nfy = nfz = sfx = sfy = sfz = dep = 0
 f.write ("%6d %12.8f %12.8f %12.8f %12.8f %9.6f %9.6f %9.6f %9.6f %9.6f %9.6f %12.8f\n" % (O.iter,O.time,ov,v1,v2,nfx,nfy,nfz,sfx,sfy,sfz,dep))
        f.flush()

O.materials.append(FrictMat(young=0.1e9,poisson=.22,frictionAngle=0,density=2300,label='glass'))
Sph_1 = O.bodies.append([sphere([0,0,0], radius=rad, material = "glass")])
Sph_2 = O.bodies.append([sphere([0.1,0,0], radius=rad, material = "glass")])
O.bodies[1].state.vel = Vector3(-0.3,0,0)
O.bodies[0].state.vel = Vector3(0.3,0,0)

O.engines=[
 ForceResetter(),
 InsertionSortCollider([Bo1_Sphere_Aabb(),
        ],label='collider'),
 InteractionLoop(
  [Ig2_Sphere_Sphere_ScGeom()],
  [Ip2_FrictMat_FrictMat_MindlinPhys()],
  [Law2_ScGeom_MindlinPhys_Mindlin()],
 ),
 NewtonIntegrator(damping=0.0,exactAsphericalRot=True,gravity=(0,0,0)),
 PyRunner(command='forces()', iterPeriod=1),
]

O.dt=PWaveTimeStep()
try:
 from yade import qt
 qt.Controller()
 qt.View()
except ImportError: pass
O.saveTmp()
O.timingEnabled=True

Ahhhh, I've understood my error! The overlap contained in penetrationDepth is the one computed for the force, which corresponds to positions at the previous step! Sorry to make you lose your precious time and thanks for help anyway!