2 Replies Latest reply on Mar 31, 2014 3:18 PM by Shaun Densberger

    Problem in Drop test simulation

    Saggam Ravinder

      Capture.PNGSolidworks showing above mensioned problem when i try to solve the drop test.

      I am using Multibodies part and bonded perfectly but still i am facing this problem.. please help me..

        • Re: Problem in Drop test simulation
          Jared Conway

          have you looked up this warning in the solidworks kb at customerportal.solidworks.com? if i remember correctly this is just a warning and that it should have limited effects on the accuracy of your solution.

           

          that being said, it generally indicates that some work can be done to simplify the model to eliminate bad elements.

          • Re: Problem in Drop test simulation
            Shaun Densberger

            I haven't encountered this specific error before, nor am I too familiar with SW drop test tool, but as Jared said I think this issue is due to poor quality elements.

             

            In SW (and most FEA codes) a drop test is done with explicit dynamics (i.e. explicit integration) instead of implicit dynamics (i.e. implicit integration). One of the key differences between the two is that the latter is unconditionally stable (the former is only conditionally stable). The condition for stability is the size of the time step used; having too large of a time step (and even too small of a time step) can cause numerical problems that will either cause the analysis to fail or give very inaccurate values. The value that the time step must be equal to or smaller than is called the Critical Time Step.

             

            For a finite element model (undamped), the critical time step is defined by the Courant Stability Limit. This limit is generally defined by the highest eigenvalue of the system, which is usually the dilatational mode of one of the elements in the mesh. The relation between the critical time step and the highest eigenvalue is inverse, so the higher your highest egienvalue is, the smaller your critical time step needs to be. In essence, your critical time step is tied to the stiffest and/or smallest element in your mesh. This is where SW starts to get into trouble.

             

            It's well documented that tet and tri elements are more sensitive to the aspect ratio then brick and quad elements. Since SW can only create a tet or tri mesh, you need to make sure you have a good aspect ratio across all of your elements. It'd be worth your time to inspect the mesh and look for any high aspect ratio elements (you can make a plot of this to assist you).