Fluids

Fluids

Particle pairs are overlapping in MPM model(macroscopic particle model)-ANSYS Fluent?

    • za36kuzo
      Subscriber


      Hi everyone,

      I am modelling the blockage of Particle in an air-filled pipe using Macroscopic particle model MPM. By means of particle-based numerical simulations of gravity-driven flows in vertical pipes and test the material transport by adding a helical texture to the inner-wall of the pipe, however, in the first case without helix structure, the blockage has been achieved due to the collision, but some of particle pairs tend to overlap, and I reserved Warning: 1 particle pairs are overlapping!

      In the second case, by adding the helix, the most of my particle are lost out of the domain in short time after running the simulation. Please check the images to get a better image of the problem.

      Initial time step size 1e-04

      I tried to reduce the coefficient of restitution , in order to induce the loss energy for MPM particle once it hits the wall and can ensure that the MPM does not lose contact with the wall ( please correct me if its wrong), and set a high value for minimum drag force for particle detachment., but it does work.

      Any help would be greatly appreciated.

      Regards!

    • Rob
      Ansys Employee
      Try decreasing the time step. If the step is too big the particles can overlap before that's noticed by the solver and the amount of energy this stores then causes the particles to separate VERY quickly. n
    • za36kuzo
      Subscriber
      Thanks for replaying, I will try it!n
    • za36kuzo
      Subscriber
      I reduce the time step size to 10^-6,but it doesn?t resolve this overlapping!n
    • Rob
      Ansys Employee
      How fast are the particles moving and how big are the cells? n
    • za36kuzo
      Subscriber
      For MPM the mesh element size must be smaller than the particle diameter size, so in my case particle size = 0,0012 m and element size was 0.0005 m. nDue to the helix structure the particle moves very very fast, so they lost faster in the domain, compared with the pipe without helix. I noticed that in both case the particle pairs tend to overlap, so far they are close to wall with high velocity( 2-5) m/ s, specially when particles are jamming in the narrow pipe.n
Viewing 5 reply threads
  • You must be logged in to reply to this topic.