I am trying to use mfix to build a ‘rice pile’ to replicate some experiments on self organised criticality I want to do for my phd thesis. The overall structure of this set up is that there is an inlet to the system where grains are added from above, then a pile of grains begins to build, but an outlet is also present in the system where grains can leave during avalanche events.
I have built the container and have set the input rate of grains. Now I want to make the grains which are inputted into the system behave as rice would without changing the shape of the particles; I do not want them to bounce or roll when added to the system.
However, my issue is that the grains inputted to the system cause others to roll along the container and straight out the outlet, preventing the building of a rice pile.
I have changed the coefficient of restitution and friction coefficient to those found for rice. Is there anything else I can do to prevent these grains rolling and allow a pile to build up?
I think having a high friction coefficient should address your issue. Did you try increasing the values further? You should be adjusting both particle-particle and particle-wall friction coefficients.
Hello,
Yes, my supervisor suggested altering the friction coefficient and the coefficient of restitution.
I have currently got my numbers set at the highest possible, which I believe to be:
Friction coefficient: 1.0 (for both particle-particle and particle-wall)
CoR: 1.0 (for both the solid and the wall)
This has reduced the rolling I am getting, but after a certain build up the grains still roll straight out of the box.
Box1forum.mfx (26.0 KB)
CoR = 1.0 makes the collisions more elastic. You should try reducing it to form the pile. You can also try increasing the spring constant.
Ok that makes sense. I have tried keeping my friction coefficient at 1, my CoR close to 0, and then my spring constant near 5000 N/m and the issue of the spheres rolling and not building a pile like rice would is still happening.
Is it possible to recommend a range of numbers for the parameters in question which may help achieve this?
Your tangential collision factor is also too high. I would suggest following the DEM 3D Hopper as a baseline to set the parameters in your simulation.
Do you have any suggestion for a range in which the tangential collision factor should be in this circumstance?
I have used the DEM 3D hopper as a baseline as you suggested, however I am still unable to get the spheres to build up into a pile, where the grains are completely static and do not roll unless avalanched by inputting grains.
Box1forum.mfx (26.1 KB)
MFiX-DEM does not have a model for rolling friction. You might have to include an additional term to the effective torque during your calculations to account for such effects. This could be done through a UDF.
I understand what you are saying and I have found the editor pane within MFIX where i can create UDF’s. However is there a specific .f file i need to edit in order to do this?
Take a look at calc_force_dem.f and calc_collision_wall_mod.f, those should give you an idea.
Can you please help me to add the rolling resistance Coefficient?