2 Replies Latest reply on Jul 13, 2010 1:45 PM by Scott Pratt

    Singular matrix error.

    Scott Pratt

      Hello All,

      I am not sure why I am getting the following message:

       

      "Singular matrix. Instable model or too large Inplane loads or displacements. Specify a shift, or reduce load/displacement."

       

      I am preforming a frequency analysis on a set of weldments. Thoughts? Thanks.

       

      -Scott

        • Re: Singular matrix error.
          Anthony Botting

          Hi Scott: Suppress loads and try again. If you get the same message, the SIngular matrix means there's a zero on the main diagonal and it cannot invert the stiffness matrix (analogous to dividing by zero). Physically it means there is Rigid Body Motion (RBM), which in turn means a single body can translate or rotate without deforming - i.e.,  something is loose. It happens frequently when bonded contacts are not enforced. You can investigate what is loose by creating a static study. Drag and drop the materials and fixtures to the static study. Turn on load "gravity". Make sure you set the solver to "Direct Sparse" and switch-on "Use soft spring to stabilize model" in Properties of the study. Then run. You will probably get a rather voluminous message stating the model has large displacements. Choose the "no" option because you want to solve using small displacement theory. Now, plot displacement contours - undeformed shape (do not allow a deformed plot). Blue colored bodies have URES =0 and red-colored bodies have URES = 1E8 (they moved as a rigid body). Those red ones are the loose bodies - you'll have to investigate the contacts to other bodies.

            • Re: Singular matrix error.
              Scott Pratt

              Hey Anthony,

              Thanks for the response. I tried to do what you suggested but I was not able to get it to work. I removed all but three components: one ring weldment that is treated as a beam and two vertical weldments that I treat as solids. I am thinking that it is the means of connecting these together that is causing my problem. I tried to use the joints on the ring beam and bond them to the bottom surface of the vertical weldments. Is there a certain way that you are suppose to connect solids and beams together? Anyhoo, I performed a static analysis as you suggested and it deformed as expected, see attached pic. Thoughts?

               

              -Scott