Qustion about return value in contact model

Hi Raheem,

The torques in EDEM are calculated per particle and the forces are calculated on the contact and applied equal and opposite on each particle, just to confirm you are applying Torque1 and Torque2?  You should also apply additionalTorque not additionalusTorque, at least unless you also have damping in the rotational component.

For some additional background on the difference between  contactResults.additionalTorque (1 or 2) and contactResults.usAdditionalTorque any of the forces and torques that have the us is related to the energy loss/damping.  The API header say:

Hi Raheem,

The torques in EDEM are calculated per particle and the forces are calculated on the contact and applied equal and opposite on each particle, just to confirm you are applying Torque1 and Torque2?  You should also apply additionalTorque not additionalusTorque, at least unless you also have damping in the rotational component.

For some additional background on the difference between  contactResults.additionalTorque (1 or 2) and contactResults.usAdditionalTorque any of the forces and torques that have the us is related to the energy loss/damping.  The API header say:

NApiHelpersV3_4_0::CSimple3DVector usNormalForce

The unsymmetrical normal forces on element 1.

This represent the portion of the calculated normal force that causes energy loss / gain (eg damping).

 

This us (unsymmetrical force/torque doesn’t impact the simulation results such as particle positions but does change how the energy is calculated and how the force on the geometry or particle is calculated.

When you query ‘Force’ in the EDEM Analyst this is the undamped force (F_n), when calculating the particle motion the damped force is used (F_n + F_nd).  What you see in the Hertz-Mindlin model is:

       CSimple3DVector newF_n = F_n + F_nd;

       calculatedUnsymNormalForceX = F_nd.dx();

       calculatedUnsymNormalForceY = F_nd.dy();

       calculatedUnsymNormalForceZ = F_nd.dz();

 

       calculatedNormalForceX = newF_n.dx();

       calculatedNormalForceY = newF_n.dy();

       calculatedNormalForceZ = newF_n.dz();

 

calculatedNormalForce is used to determine the motion of the material (F_n + F_nd). When looking at the EDEM Analyst and reporting the forces on the geometry or forces on the particle EDEM reports ‘calcualtedNormalForce – calculatedUnsymNormalForce’.

The only force that is physical (real) at the contact is the contact spring force, the damping force is there to recognise that during a contact damping does occur but it’s very complex and simply represented in DEM by a dashpot. The real force between two elements in contact is the spring force only (+ cohesive forces).  This is why we return both the damped and undamped force to EDEM for use in particle motion calculation and returned forces on elements.

 

Regards

Stephen

Hi Stephen

Thank you for your patient answer，as you said, the torques in EDEM are calculated per particle, but why the API header say additionalTorque1 act at the contact point, 

Hi Stephen

Thank you for your patient answer，as you said, the torques in EDEM are calculated per particle, but why the API header say additionalTorque1 act at the contact point, 

NApiHelpersV3_4_0::CSimple3DVector additionalTorque1

Any additional torque on element 1 not accounted for by the above forces (which are deemed to act at the contact point

 

What I am concered is that the value assgined to the  contactResults.additionalTorque(1 or 2) is act on the center of mass of particle(1 or 2) or the contact.contactPoint.

 

Reagrds!

Raheem

 

Hi Raheem,

The torque is applied at the centre of the particle (centre of mass for multi-sphere).  The forces applied are applied at the contact point so for Normal force that won't create any additional torque but the Tangential force will.

Regards

Stephen