cyclic boundary conditions for particles
 

Hello,

I'm using MPPICFoam (v2.3.0). A simple test case in a column has been set up. Particles are sitting on the bed initially. I want to set the left and right patch to be cyclic for the particles, so that particles that escape from the left patch will enter through the right patch again. Therefore, equilibrium status can be achieved using less computational resources. I tried

but 'coincidentBaffleInteraction' is not recognised. I guess I have to dig out the code for coincidentBaffleInteraction and compile it. However I'm not sure whether this can help with the periodic boundary condition for particles.

Then I simply tried out using 'escape' for the left patch and right patch, both of which are of the patchType cyclic and correspond to each other. But as expected, particles just escape and cannot enter into the domain again. Just wondering how can I specify periodic boundary conditions for particles?

thanks,
Sophie

Hello Sophie,

I am currently working on developing my own code for particle interaction and i need to employ periodic BC on left and right faces. I am however working on a structured grid solver. In Control volume approach you usually generate ghost cells around the original grid wherein you can apply the BC's. So for example particles exit from right most cell of the original grid to the ghost cell, you then specify that the particle coordinates are changed so that it 'appears' in the leftmost grid cell at the corresponding coordinate. I am not an expert on unstructured grids but i am sure the implementation would be similar in principle.
I am curious if OF offers such a BC for particles. Please share if you have found a solution.

Thanks & Regards.

Hello,

Sorry to restart this thread but I am trying to tackle a similar problem using DPMFoam and require cyclic boundary conditions for the particle phase. Did either of you manage to find a solution to this problem?

Thanks,
Charlie

Hello all,

I am also trying to figure out how to implement cyclic BC's for the particles. Have you already a solution or any idea?.

Thank you.

Hi jarpve,

I havn't validated the model yet but it is possible to apply cyclic BCs if you change the 'patchInteractionModel' to 'standardWallInteraction' and just specify the interaction coefficients for the wall. Although there is no fvOptions term in the velocity equation in DPMFoam so you can't specify a bulk flow velocity. To stop the flow slowing down over time you have to add fvOptions to DPMFoam. You can look at the standard pimpleFoam solver to see how this is done.

Hope that helps!

Hello Charlie,

Thank you for your answer. I am going to do it and I will tell you how it was.

Particles are inserted in DPMFoam cyclic BCs
 

Hello everyone,


I have to add fvOptions to DPMFoam eventhough no particle has been inserted to DPMFoam when I applied cyclic BCs.

Thank you.

particle stress model
 

Hello dears,

Can anybody explain the parameters "Ps" and "beta" in Harris-Crighton model for particle stress?? I couldn't find any refrences.

what are the values for different particles??

Thank you

Hi,

Has anyone found a solution to this?

I have a test case where the left and right patch have different pressures so there is flow.
My question is: how can I keep this condition but also let those patches be cyclic for the particles?

Did you make the coincidentBaffleInteraction work or did you find another way to assign cyclic boundary conditions for a particle simulation? Unfortunately I also haven't made it work for me yet although spending a lot of time on it.

For people who run into this problem,

In order to get the cyclic for the particle follows:

1. give cyclic boundary condition in for the flow (i.e., cyclic in blockMesh and flow solver)

2. DO NOT specify the boundary condition in the "localInteractionCoeffs"

Thanks,
Rdf

To complement this, you can specify "type none;" to the patch in "localInteractionCoeffs" if OF complains you haven't specified an interaction for a patch.