4 Replies Latest reply on Dec 9, 2010 10:44 AM by Ryan Werner

    SOLVER ERROR

    Apostolos Lilias

      HELLO

      I AM TRYING TO SOLVE MY MODEL AND I HAVE THE FOLLOWING ERROR MESSAGE:

      THE ITERATIVE SOLVER STOPPED.STATUS CODE 2:PRECONDITIONER FAILS TO FACTOR. MAY NOT HAVE SUFFICIENT RESTRAINTS.

      CURRENT TASK:PRECONDITIONING

      CAN YOU GIVE PLEAS AN ADVICE WHAT IS THIS ERROR AND WHAT TO SEARCH IN ORDER TO CORRECT IT.

        • Re: SOLVER ERROR
          Ryan Werner

          That error usually means that some part of your model is not adequately restrained or connected to the rest of the model so it ends up "floating."  Go to the properties of the study and turn on the Soft Springs stabilization option.  This will help artificially stabilize your model so that the solver can converge and you will be able to see the results showing which piece/s of your model are floating.  It will tell you there is large displacement occurring and ask you if you want to rerun with the Large Displacement flag turned on, hit No and it will finish the analysis.  Once you see what parts are moving you can add the appropriate restraints and/or contacts to hold everything in place.

           

          I did this quickly on your model and it worked, showing which pieces needed more constraining.

           

          Ryan W.

            • Re: SOLVER ERROR
              Apostolos Lilias

              HELLO RYAN

              THANK YOU FOR YOUR ANSWER. I FOLLOWED YOUR INSTRUCTIONS AND I HAD THE SAME RESULT. ALSO I TRIED TO CHANGE SOLVER THE SAME. I TRIED A SOLVE WITH LARGE DISPLACEMENT FLAG BUT THERE WAS NO CHANGE. CAN YOU TELL IF YOU MADE ANY MODIFICATIONS IN MY MODEL? IT IS PROBLEM FOR YOU TO SEND ME BACK THE MODEL THAT YOU MADE YOUR SOLVE TO TRIED MYSELF

              THANKS

                • Re: SOLVER ERROR
                  Anthony Botting

                  Make sure to choose "No" when you get the large displacement option/flag/notice. This way, it will solve using small displacement assumption and hopefully provide an answer. Then, plot displacement contours (URES) with the deformation flag turned off. You should be able to see the offending component colored red. Hope that helps. It happens quite frequently to our customers - the impulse is to check "Yes" when you see that large displacement message. -Tony

                  • Re: SOLVER ERROR
                    Ryan Werner

                    Soft Springs would only work using the Direct Sparse solver.  I originally did not make any changes to your model to get the results showing the pieces that were not restrained.  However, since you asked I went ahead and created a new study using all solid bodies and set it up so that it gives good results without any errors or warnings.  Hopefully, your computer has the resources to run this study.  Since it is a pretty small and simple assembly I am confident that you should be able to.  I did make one change to your model though to get the mesh to work properly.  Part SB1-10-8800L had a 80.5mm diameter hole that I changed to 80mm so that it was the same size as the hole in the other part.  Simulation did not like the very small shoulder that was being created there as a result of the difference in diameters.  I also had to add numerous contact sets to make sure everything was bonded.  The mirrored part for SB1-11-8800L is not touching the rest of the model.  There is an very small gap there and that is why you have been getting the errors.  I just added a contact set but you might want to look into why your plane and mirror feature are not working right.  Good luck.

                     

                    Ryan W.