Fickian Diffusion Problem

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • pcundergrad
    Junior Member
    • Oct 2019
    • 10

    Fickian Diffusion Problem

    Hi,

    I am trying to recreate the analysis of the 'cylindrical disk of porous material' from the paper; Gerard A. Ateshian Michael B. Albro Steve Maas Jeffrey A. Weiss Finite Element Implementation of Mechanochemical Phenomena in Neutral Deformable Porous Media Under Finite Deformation.

    I have created the model and applied the boundary conditions as best I can, however, when I run the simulation, it seems to only be able to converge on one or two time steps, then the problem diverges and error terminates. I can't figure out where I am making the mistake with my model. Can anyone help?

    Thanks,
    Peter
    Attached Files
  • ateshian
    Developer
    • Dec 2007
    • 1821

    #2
    Hi Peter,

    Since the solid matrix does not deform in this problem, you should set dtol=0. Currently your model struggles with convergence on the displacement, even though the displacement residual is tiny. So setting dtol=0 overcomes this numerical roundoff error.

    Best,

    Gerard

    Comment

    • pcundergrad
      Junior Member
      • Oct 2019
      • 10

      #3
      Hi Gerard,

      Of course, that makes sense. Worked perfectly, thanks for your help!

      Regards,
      Peter

      Comment

      • jedwards13
        Junior Member
        • Nov 2020
        • 12

        #4
        Hi Gerard,

        I messaged Peter above but he no longer has access to FEBio so I was hoping you could assist me?

        I have a project where I need to create a model very similar to that of the porous cylindrical disk. I have tried to attach my model but it seems to think the file size is too big, I can email this to you if possible?

        The main element I'm unsure of is creating the boundary conditions necessary for the diffusion to take place. Mainly the displacement. For all the boundary conditions a large number of nodes have been selected rather than simply the bath or the tissue, why is this? And how do I know which nodes to select to lock displacement? It may be an issue on my model side of things.

        Hope to hear back from you soon

        Jordan

        Comment

        Working...
        X
        😀
        😂
        🥰
        😘
        🤢
        😎
        😞
        😡
        👍
        👎