Hi Everyone,
I am wondering how EDEM handles the contact radius and if it is possible to achieve my goal with some work-arounds.
First of all, I understand that a contact pair is registered if particles have contact radius overlap: triggering the calculateforce() function. Now what I dont understand is what are the criterion for EDEM are to stop evaluating the calculateforce() function. Because in the JKRV1 + HM model, it is discouraged to enable a contact radius i.e. physical radius = contact radius. So, a contact is registered if physical radius of both particles touch. However, this model does claim that there are non-contact adhesive forces i.e. at zero overlap (gap between particles). So apparantly the calculateforce() function is still triggered if particles are outside of the contact radius. Can anyone explain?
Then the following: currently I have a shrinkage plugin that sets particle radius using the setScale() method. Now my understanding is that this will always automatically also change the Contact Radius together with the physical radius. Correct? Or does setScale only scale the phsyical radius and not the contact radius?
I am trying to create a contact model which simulates the melting of a particle. My idea is to realise the following:
- A particle is initialised at 12mm. Its contact radius and both physical radius are set to 12mm.
- During simulation, shrinkage is applied based on temperature of the particle. The idea is that only the physical radius should be changed. The contact radius should remain 12mm to represent the molten layer. The difference between the physical radius and the contact radius can then be seen as a molten layer surrounding the core of the particle. A contact should remain registered between particles (calculateforce() triggered) if molten layers interact. See image below.
Source: https://research.utwente.nl/en/publications/numerical-and-experimental-studies-of-laser-powder-bed-fusion-pro
Now, this would allow me to define a contact model that calculates forces when the molten layers of two particles start to interact i.e. when we start to have positive contact radius overlap. Then, once the physical unmolten core of the particles interact I will have additional hertz-mindlin elastic contact forces. Effectively, this allows me to create a cohesive model for partially molten particles.
The problem I am phasing is combining shrinkage via the setScale together with this representation of the molten layer. For this to work I need the solid core to keep shrinking but somehow keep the contact radius representing the molten layer to remain constant at 12mm. Because, I want the calculateforce() function to be triggered even if the physical particles no longer touch, as long as the molten layers interact. Also, new contacts should be registered if only the molten layers interact.
Can anyone explain to me if the contact radius for example can be set manually for example after the setScale() function to counter this behaviour?
On the page for HM+JKRV2 it says that the relative approaching distance is calculated independent fromt the contact radius. And the model allows a negative approaching distance (until the rupture distance). However, what happens when we have a particle size distribution? Will that not scale both contact radius and physical radius but disproportionately? And setting the rupture distance for the different particles is not automatically changed? The same for having HM+JKRV2 with shrinkage of particles. Will using setScale() to shrink the particle not also result in the risk of setting the contact radius to a smaller value than the rupture distance? I assume that the rupture distance is not automatically adjusted.
Also, I can nowhere find any information on how the contact radius behaves with particle size distributions. In the end I want my contact model to also work with particle size distributions. If I set in EDEM a contact radius of 12mm as wel as the physical radius to 12mm but we do have particle size distributions, will the contact radius correctly scale with the particle during creation? I assume yes but just to check.
I know I am trying to test EDEM here with some trickery but for my purpose I really need to have both shrinkage as well as "non contact" adhesive forces.
Thanks in advance!
Kind regards,
Jan-Thijn Wijnker
