Help with Multi-step analysis

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • anastasia
    Junior Member
    • Jun 2011
    • 16

    Help with Multi-step analysis

    Hello!

    I have two questions with performing multi-step analysis and I would really appreciate it if someone could help me with it...

    1) I want to apply a pressure and after the pressure loading is completed I need to apply some nodal forces but without removing the pressure...
    Can this be done by appointing the pressure loading to step 1 and then the nodal forces to step 2?

    2)If I do this will the pressure stop being exerted on my elements?
    I mean after each step is completed do its forces continue to be applied on the next step or not? (I have tried FeBio manual but couldn't clarify this...)

    Many thanks in advance for the help!

    Anastasia
  • ateshian
    Developer
    • Dec 2007
    • 1824

    #2
    Hi Anastasia,

    Nodal loads will be superposed over the pressure, so it is possible to apply both types of boundary conditions simultaneously.

    You don't need to perform a multi-step analysis for this problem. You can prescribe the pressure and nodal forces within the same step. Set the load curve of the nodal force to zero over the time span when you initially increase the pressure. Once you have reached the desired pressure, you can ramp up the nodal loads.

    Best,

    Gerard

    Comment

    • anastasia
      Junior Member
      • Jun 2011
      • 16

      #3
      Hi Gerard!

      The reason I need to perform a multi step analysis is because I need to change the boundary conditions from i.e. fix"z" to fix "xyz" in some nodes during the application of the nodal forces. ( Btw can I do this by applying the constant boundary i.e. fix "z" as general boundary and then add the extra x or xy fixed foundary as prescribed displacement within the second step?- Itried this with a smeller file and it seems to be working...)

      Thank you very much for the help!

      Best,

      Anastasia

      Comment

      • ateshian
        Developer
        • Dec 2007
        • 1824

        #4
        Hi Anastasia,

        If you need to change the fixed boundary conditions then a multi-step analysis is indeed the way to go. You can prescribe the pressure in the initial step (so it stays on during the entire analysis). You can prescribe the fixed bc's in the first step and a different set in the second step. You can prescribe the nodal force in the second step. That should do the trick.

        Best,

        Gerard

        Comment

        • anastasia
          Junior Member
          • Jun 2011
          • 16

          #5
          Hi Gerard,

          So to make sure I understood right I should apply pressure under the general boundary conditions (before I define any steps) and then change my fixed b.c. in each step. The problem is that if I do that, I will have to define timesteps under the general <control> element, although from the manual I understand that this is not applicable in multistep analysis... Besides it is also written in the manual that only prescribed displacements can be defined in steps and not fixed b.c...

          Could you please help me if I have misunderstood your suggestions or the FeBio manual instructions?

          Many thanks in advance !

          Best,

          Anastasia

          Comment

          • ateshian
            Developer
            • Dec 2007
            • 1824

            #6
            Hi Anastasia,

            1) If you prescribe a pressure boundary condition in the Initial step, the load curve associated with that boundary condition will apply to all subsequent steps. For example, if Step 1 goes from t=0 to t=1 and Step 2 is from t=1 to t=2, then the load curve for the pressure bc should span the time from 0 to 2.

            2) You can use prescribed displacements instead of fixed displacements in a multi-step analysis. Simply prescribe the values to zero if you want the same behavior as a fixed bc.

            Best,

            Gerard

            Comment

            • ntjacobs
              Member
              • Nov 2009
              • 79

              #7
              Thanks for the interesting thread. I have a related question, I am trying to model a multi-step analysis of confined compression with several compression ramps. The simulation starts with a swelling (step 1) under a very small tare load (0.01 N) followed by 5 analysis steps, each consisting of a compression ramp using a prescribed displacement boundary condition.

              The prescribed displacement in each step appears to be a displacement to an absolute value instead of a relative displacement from where the sample finished at the end of the previous step. For example, Because I have an initial swelling under a small load, at the start of the first compression the sample is taller than at the start of step 1. Let's say that if the sample starts with a height of 1, it may swell to 1.04. When I apply the compressive ramp, I would like it to compress .15 mm from 1.04, to a height of 0.89, instead, it compresses 0.15 mm from the initial state of 1.to a new height of 0.85.

              Is there any way to switch to "relative" displacement boundary conditions? Because I will be simulating several different tissue specimens, each with their own permeability, it becomes cumbersome to run separate analyses for the swelling and compression steps so I was hoping to capture the entire sequence within one simulation.

              Thanks,

              Nathan

              Comment

              • ateshian
                Developer
                • Dec 2007
                • 1824

                #8
                Hi Nathan,

                As of now it is not possible to prescribe "relative" displacement boundary conditions in FEBio. As you state, all prescribed boundary conditions represent absolute values (or, more accurately, values relative to the stress-free reference configuration).

                The problem you describe also arises in other types of problems, e.g., performing a contact analysis under load control (where you first need to initiate contact under displacement control, before switching to load control in a multi-step analysis). In the contact case you would want a "relative" load boundary condition, to prescribe a contact load over and above the value achieved under displacement control.

                My initial thought is that it would be hard to implement a sufficiently general method of prescribing relative boundary conditions (i.e., a method that works for any number of steps, not just two; and for a mix of load and displacement boundary conditions). Maybe the swelling case can be treated as a special category of problems, it's worth looking into it.

                Best,

                Gerard

                Comment

                • ntjacobs
                  Member
                  • Nov 2009
                  • 79

                  #9
                  Gerard,

                  Thanks for the quick reply. I understand this may not be possible to implement. If you do devise a way to allow for an initial swelling followed by compression, let me know, otherwise I'll split the problem up into two separate analyses and try to arrive at the same end result.

                  Thanks again,

                  Nathan

                  Comment

                  • anastasia
                    Junior Member
                    • Jun 2011
                    • 16

                    #10
                    Many thanks Gerard for the help !!!! I will do as you advised...

                    Nathan, good luck !

                    Anastasia

                    Comment

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