Meta-Particle

Mrage

Hi,

I am trying to model a particle which is surrounded by a soft shell. So I was wondering if it is possible to create a meta-particle with one sphere inside another (see attachment). Would that work?

 

Thanks and best regards

Leandro

Stephen Cole

Hi Leandro,

 

Does the outer shell need to be a flexible material (meta-particle) or a rigid surface with low Youngs Modulus?

 

Either aproach should be possible, you just need to create the particle shape in the EDEM Creator.  The shape could be the surrounding sphere or single spheres but placed in a position so they create a surround. I created a rectangle shape like this with a 3-sphere particle placed inside. You can set the positions in a spreadsheet and copy and paste them into the table. 

 

We have a meta-particle tutorial here:

https://community.altair.com/community?id=kb_article_view&sysparm_article=KB0037194

 

I also created a simple case (attached) where the rigid particle is created inside the block. I created it as a rectangle just as a quick example, it was easier to specify the particle positions to create the surround as a square however if you set the appropriate positions you can create any shape.

 

Mrage

Stephen Cole_21117 said:

Hi Leandro,

 

Does the outer shell need to be a flexible material (meta-particle) or a rigid surface with low Youngs Modulus?

 

Either aproach should be possible, you just need to create the particle shape in the EDEM Creator.  The shape could be the surrounding sphere or single spheres but placed in a position so they create a surround. I created a rectangle shape like this with a 3-sphere particle placed inside. You can set the positions in a spreadsheet and copy and paste them into the table. 

 

We have a meta-particle tutorial here:

https://community.altair.com/community?id=kb_article_view&sysparm_article=KB0037194

 

I also created a simple case (attached) where the rigid particle is created inside the block. I created it as a rectangle just as a quick example, it was easier to specify the particle positions to create the surround as a square however if you set the appropriate positions you can create any shape.

 

Hi Stephen,

thanks four all the usefull information.

I am trying to simulate the aggregates with the soft shell in an uniaxial compression test. So the shell needs to be felxible and the aggregate itself can be riged. To keep the simulation-time low i prefer the method where i only need to spheres per particle (inner spher and outer sphere). The setup is displayed in the following picture:

I have to questions about this:

 

- Is there a way to define the inner sphere as rigid other then giving it a very high young's modulus? (since this increases the simulation-time)

- When i try to simulate the particles as discribed the inner sphere shoots out of the outer sphere whenver PP-Interactions or PG-Interactions occur. I guess this is because of the huge overlap and the resulting normal force. Is that correct? And how can I lock them in place?

 

If this is not working, i will definitly try to set it up with multiple spheres surrounding as you suggested.

 

Best regards,

Leandro

Stephen Cole

Mrage said:

Hi Stephen,

thanks four all the usefull information.

I am trying to simulate the aggregates with the soft shell in an uniaxial compression test. So the shell needs to be felxible and the aggregate itself can be riged. To keep the simulation-time low i prefer the method where i only need to spheres per particle (inner spher and outer sphere). The setup is displayed in the following picture:

I have to questions about this:

 

- Is there a way to define the inner sphere as rigid other then giving it a very high young's modulus? (since this increases the simulation-time)

- When i try to simulate the particles as discribed the inner sphere shoots out of the outer sphere whenver PP-Interactions or PG-Interactions occur. I guess this is because of the huge overlap and the resulting normal force. Is that correct? And how can I lock them in place?

 

If this is not working, i will definitly try to set it up with multiple spheres surrounding as you suggested.

 

Best regards,

Leandro

Hi Leandro,

 

Unfortunately as per the picture this wouldn't work in EDEM. The spheres are shown this way graphically however in the solver they are a point and a radius, the distance from the centre points of each element and the radii determine the overlap and the subsequent force.  In a way you could say they are both solid volumes so it is not possible to place one inside another in this way.

Maybe with the API and a custom physics model the forces could be reversed to always force one type of particle towards the centre of another however I don't believe this would be correct either, my understanding is you need the inner particle to be able to contact the outer surface which is why creating the outer surface from groups of spheres maybe the best approach.

 

Something else to consider is having the outer sphere as a geometry in which case you could create the inner sphere inside this.  I assume you wold want the outer sphere to move in which case the 'geometry' could be controlled by MotionSolve and running EDEM-MotionSolve co-simulation or similar? This approach may work but the number of geometry spheres maybe limited compared to running these as particles.

 

Regards

Stephen