Biphasic Unconfined Compression Relaxation

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • jogomez
    Junior Member
    • Jul 2011
    • 15

    #31
    Thank you very much. Now, linux version also works .

    José A.

    Comment

    • chiehhou
      Junior Member
      • Oct 2011
      • 17

      #32
      about must point

      Hi,

      I have a problem as I load the must point file (excel file save as *.mustpoint).

      should I have two column of numbers (as below) to describe my must points?
      0.1 1
      0.14 2
      0.19 3
      0.27 4
      0.37 5
      0.51 6

      Thanks.

      Chieh

      Comment

      • maas
        Lead Code Developer
        • Nov 2007
        • 3400

        #33
        Hi Chieh,

        We are getting a bit off-topic here. Can you please start a new thread?

        Thanks,

        Steve.
        Department of Bioengineering, University of Utah
        Scientific Computing and Imaging institute, University of Utah

        Comment

        • Farrell1
          Junior Member
          • Oct 2011
          • 25

          #34
          Hi,

          I am wondering which meshing configuration is most appropriate for the unconfined compression of a 1/4 cylindrical specimen? I believe in the tutorial you specify a butterfly mesh, can you tell me why this is appropriate?

          Regards

          Mark

          Comment

          • ateshian
            Developer
            • Dec 2007
            • 1824

            #35
            Hi Mark,

            If the analysis is axisymmetric (which is usually the case for unconfined compression), a wedged center would be preferable. This configuration preserves axisymmetry better.

            Best,

            Gerard

            Comment

            • Farrell1
              Junior Member
              • Oct 2011
              • 25

              #36
              Thank you for the reply Gerard,

              I have another question - I notice radial variation in both apparent Poisson's ratio (measured at the end of the ramp) and equilibrium Poisson's ratio (smaller values were obtained at nodes toward the centre of the disc, and the greatest values obtained toward the circumferential edge). I have modelled a Holmes-Mow solid phase coupled with Holmes-Mow permeability. Since Poisson's ratio is an intrinsic material property is this variation a related to the model conditions?
              Regards

              Mark

              Comment

              • ateshian
                Developer
                • Dec 2007
                • 1824

                #37
                Hi Mark,

                Please keep in mind that Poisson's ratio has a well-defined physical meaning (-lateral strain/axial strain) only when strains are infinitesimal, and only under equilibrium conditions (in the case of a biphasic analysis). FEBio performs full-fledged finite deformation analyses with nonlinear material responses, so this physical meaning is only recovered when using really small strains (e.g., 0.001 or smaller).

                Based on analytical solutions of biphasic unconfined compression for small strains, we know that a step compressive strain produces an isochoric response everywhere in a cylindrical disk, except right at the lateral boundary, where fluid is immediately squeezed out, leading to a reduction in pore volume. This means that, for an infinitesimal strain analysis with an isotropic solid matrix, the apparent Poisson's ratio is exactly 1/2 everywhere except on the boundary, where it is less than that. Over time, as fluid flows out of the material from regions inside, the pores transiently lose their volume in a non-uniform way, producing an apparent Poisson's ratio that varies radially. Eventually, when the fluid pressure subsides to zero everywhere, the loss of pore volume becomes uniform along the radial direction, producing the true Poisson ratio (the value prescribed by the user) everywhere.

                To reproduce these findings, you need to use very small compressive strains and make sure that the transient analysis has reached equilibrium. Please try it out and let me know if you still get unexpected outcomes.

                To evaluate the apparent Poisson's ratio, you need to use principal strains (and correctly identify the radial and axial values), not the Cartesian components of the strain tensor. That's because elements in a wedge geometry become larger in the circumferential direction as the radial position increases. This means that the Cartesian components of the strain (evaluated at the center of the element) deviate more from the principal strains at increasing radial positions. Your post did not indicate how you evaluated the apparent Poisson's ratio, but hopefully this clarification might also help you.

                Best,

                Gerard

                Comment

                • Farrell1
                  Junior Member
                  • Oct 2011
                  • 25

                  #38
                  Hi Gerard,

                  I have attached a screen shot of my model where I have plotted the XY displacement of three nodes. As you can see, there is radial variation with the greatest lateral expansion occuring toward the circumferential edge of the model (n12 - outer edge node, n126 - inner node).

                  If these XY displacements are used to calculate radial strain and subsequently Poisson's ratio, then the same variation will exist. Do you think this is the result of element geometry?

                  Regards
                  Mark
                  Attached Files

                  Comment

                  • ateshian
                    Developer
                    • Dec 2007
                    • 1824

                    #39
                    Hi Mark,

                    You are plotting nodal displacements, not strains. At steady state, the displacements will increase linearly along the radial direction, starting from zero at the center, so the variation you observe is correct. If you plot the radial normal strain instead (X-Lagrange strain in this case), you will find that these nodes equilibrate to the same value. If you take the ratio -(Z-Lagrange strain)/(X-Lagrange strain) at steady state, you should recover a value very close to the prescribed Poisson's ratio.

                    Best,

                    Gerard

                    Comment

                    • ateshian
                      Developer
                      • Dec 2007
                      • 1824

                      #40
                      I have added a biphasic unconfined compression problem in the test suite for FEBio, called bp05.feb. This example uses a neo-Hookean solid matrix and a constant isotropic permeability. Axisymmetry is modeled using a wedge geometry with a symmetry plane as described here. The mesh is biased to produce thinner elements at the radial edge, where the solution exhibits a boundary layer in the fluid pressure at early time points. A step compressive strain is prescribed, producing stress-relaxation.

                      Gerard

                      Comment

                      • maas
                        Lead Code Developer
                        • Nov 2007
                        • 3400

                        #41
                        This tutorial has been updated to use PreView 1.12 and FEBio 1.6.

                        Steve
                        Department of Bioengineering, University of Utah
                        Scientific Computing and Imaging institute, University of Utah

                        Comment

                        • maas
                          Lead Code Developer
                          • Nov 2007
                          • 3400

                          #42
                          This tutorial has been updated to use PreView 1.12 and FEBio 1.6.

                          Steve
                          Department of Bioengineering, University of Utah
                          Scientific Computing and Imaging institute, University of Utah

                          Comment

                          • acalik
                            Junior Member
                            • Jan 2015
                            • 5

                            #43
                            Hi,
                            I am trying to run the unconfined compression example using manuel.prv preview file. It seems like the zero pressure boundary condition on the outer face of the quarter cylinder model is not enforced. The pressure there is nonzero, comparable to the inside pressure in magnitude. Also, the pressure does not decay fast enough in the tissue. (After 4000 seconds, it is still far from being very small.) Since this is a tutorial example, I understand what is going one. I have not intervened with the input file.
                            Any help will be appreciated. Thanks.

                            Ahmet

                            Comment

                            • ateshian
                              Developer
                              • Dec 2007
                              • 1824

                              #44
                              Hi Ahmet,

                              Please don't use manuel.prv, this is not a file that is being updated as part of the distribution of PreView. If you have downloaded PreView from febio.org, look for the folder called Examples. It contains the tutorials that normally should be updated with every release. The file 'tutorial5.prv' contains the biphasic unconfined compression example.

                              I use PreView on the Mac. To locate the Examples folder, you need to right-click on the PreView application icon and select Show Package Contents. Then navigate through Contents/Resources/Examples. For other operating systems I am not sure where the Examples folder is located.

                              Best,

                              Gerard

                              Comment

                              • acalik
                                Junior Member
                                • Jan 2015
                                • 5

                                #45
                                Thank you for answering so quickly.

                                What does degrees of freedom (DOF) in the tet10?

                                A. (10 node displacement=30 DOF,
                                4 Node pressure=4 DOF
                                Total=34 DOF)

                                or

                                B. (10 node displacement=30 DOF,
                                10 Node pressure=10 DOF
                                Total 40 DOF)

                                Which is used in FEBio? (A?, B?)

                                Sincerely,
                                Ahmet

                                Comment

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