5 Replies Latest reply on Jul 30, 2013 3:32 PM by Zvi Vikinski

    displacement at a roller slider support of a beam

    Zvi Vikinski
    Zvi Vikinski Jul 27, 2013 1:41 AM

      Hi

      i'm simulating a simple beam on two end supprts.

      one support is defined as immovable, the other is  a fixture with a reference geometry that will allow axial translation  (roller slider).

      a force is operating in zx plane at midspan on anywhere along the beam.

      i would expect to see a longitudinal displacement at the roller slider but that is not shown on results.

      what is the reason for that ,is there anything  wrong with the fixtures definition ?

      thanks for answering me

      Zvi

      • 2188 Views
      • Tags:
        • Re: displacement at a roller slider support of a beam
          Jared Conway
          Jared Conway Jul 29, 2013 2:35 AM (in response to Zvi Vikinski)

          Haven't looked at the files. Have you tried fixed instead of immovable? Many conversations about beam analyses similar to this one where the issue is the interpretation of dof for beam problems. Sounds to me like you're probably getting results that don't make sense because your system is improperly restrained. I'd suggest getting things to work with solids and shells before you move on to beams if you're just staring with fea and solidworks simulation.

            • Re: displacement at a roller slider support of a beam
              Zvi Vikinski
              Zvi Vikinski Jul 29, 2013 11:20 AM (in response to Jared Conway)

              Hi Jared,

              thanks for replying.

              First let me say that the beam should have a hinged and not fixed support at one end and that i believe hould be simulated by an immovable fixture.

              I ran this model on some primitive FEA software and got also zero axial displacement at the roller slider fixture.

              this made me think that the question is not actually  a SW simulation forum question  but rather a basic statics issue  which should be discussed on some enginnering forum.

              nevertheless, if somebody at  this forum can explain it to me i'll thank him,

              Zvi

                • Re: displacement at a roller slider support of a beam
                  Jared Conway
                  Jared Conway Jul 29, 2013 3:55 PM (in response to Zvi Vikinski)

                  "First let me say that the beam should have a hinged and not fixed support at one end and that i believe hould be simulated by an immovable fixture."

                   

                  remember that there are 3DOF that at translation and 3 that are rotation. so when you fix, you're fixing 6. this would be done for troubleshooting but you can still create a pin restraint by using something other than immovable.

                   

                  i took a quick look at your model, i don't see anything wrong with the results. take a look at it with shells and solids like i originally suggested and leave one side fixed. i think you'll learn a lot about what is going on. but it looks like to me that with your loading, there just isn't that much axial displacement. what gives you the impression that there should be significant axial displacement? another thing you could try is contact if you're concerned that the restraints are causing the "incorrect" results.

                  • Re: displacement at a roller slider support of a beam
                    Justin Strempke
                    Justin Strempke Jul 30, 2013 5:36 PM (in response to Zvi Vikinski)

                    Zvi,

                     

                    I've been curious of this to, have tried all different sorts of end conditions, and after some reading (my best guess) is that it's a formula "issue".  Most beam models/formulas assume the loading to be perpendicular to the neutral axis and just don't have provisions for axial displacement.  They compute information at the cross-section area CG using the areas properties, with the cross-section assumed perpendicular to the neutral axis at all times.  When the load is perpendicular, no axial load transfers.  If you apply the load normal to the end, however, you'll see a compressive stress and displacement.  Why the formula models don't accound for the element length along with angular rotation to compute an overall axial displacement is beyond me, but they just don't see to.  This is more obvious when you try a cantilever load, where it should be much more obvious, but you still get zero.

                     

                    If you're interested in this value, as Jared suggested, start with solids/shells as they'll solve for all those variables.  Even if it is a 'beam', 'beam elements' may not be the best option to use.

                      • Re: displacement at a roller slider support of a beam
                        Zvi Vikinski
                        Zvi Vikinski Jul 30, 2013 3:32 PM (in response to Justin Strempke)

                        Hi Justin

                        thats right, if you look at the free end of a catilevered beam under load, it seems that it will deflect in both y and in x direction.

                        all formulas i saw  in textbooks, give only the y deflection value.

                        if we assume no length change of the beam (no axial force components) than from simple geometry its obvious that  the end will displace in  x direction as well.

                        i remember that some years ago i saw a reference to this displacement in some book but i can't find it now.

                        so i guess i'll have to search the literatre of statics to find an answer.

                        thanks,

                        Zvi