10 Replies Latest reply on Nov 18, 2013 11:48 AM by Jared Conway

    Problem with gravity related study

    Sjoerd Post

      Hello everybody,

       

      Currently i am having problems with the results of a study in solidworks which is using gravity.

      The results are not as i would suspect.

       

      The study i am trying to solve is from a air-duct installed on a deck of a ship.

      The study is not included in the attached zip file.

       

      The problem:

      When i solve the study without the gravity (Only the reaction forces of the air duct on the middle bulb-profile) I get the result which is shown on the attached picture.

      This is the result i would suspect. The deformation is symetrical and is the biggest in the middle.

       

      When i include the gravity and resolve the study ( so the reaction forces of the air duct and the gravity are included) i get the results which are showing on the other picture.

      This results are everything except symetrical.

       

      I also attached the assembly.

      Does anybody know a sollution or any explanation for these results?

       

      Already a big thanks for helping!!

        • Re: Problem with gravity related study
          Chris Michalski

          I don't see how it could be a cause, but your fixed supports do not appear symmetric either.  When I look at your images the indicators of fixed geometry are biased (an extra set farther in on the left where there is less deflection).  I wouldn't expect such a drastic difference, but you are also talking about fractions of a mm.

            • Re: Problem with gravity related study
              Sjoerd Post

              I checked the fixtures and they are not in any way different from the one on the other side.

              I have this problem more often, sometimes (depends on what view i am looking to the assembly) fixtures are not shown even tough they are there.

               

              I ran the study with only the gravity load. And the displacement plot is just the same as the one with forces+gravity. The only differences are the values of the displacement.

                • Re: Problem with gravity related study
                  Jared Conway

                  1. could you give a better description of what you think is wrong with the results?

                  2. how long does this thing take to mesh and run? i have a pretty good computer and gave up after an hour of attempting to mesh it. were there any errors trying to run it?

                    • Re: Problem with gravity related study
                      Sjoerd Post

                      1.      If i look at the results which are shown in the picture 'Simulation with gravity' , I would suspect that the displacement which is on the right side should appear on the left side.  So the deformed result would be symetrical.

                       

                      2.     Indeed the meshing is a bit complicated because the dimensions in the assembly are pretty big. The deck is taking the longest time for me to mesh. There are no erros when i run the study. Running the study is done by a couple minutes.

                       

                      I am thinking that I am looking to much on the deformed results, but not keeping in mind that the differences are only a fraction of a millimeter.

                • Re: Problem with gravity related study
                  Sjoerd Post

                  Okay guys, thanks for al the help over here!

                  I replaced the stiffeners on the deck for normal plates.

                  I noticed there is not a problem in the way i want to perform the study.

                   

                  The results i get is due to the stiffener which are al placed with the bulb to the right. Thats why the displacement is more on the right side than on the left side.

                   

                  Again, thanks for al the help!

                  • Re: Problem with gravity related study
                    Sjoerd Post

                    Thanks for the tip. I suppose i need to learn more about the posibillties of meshing. There is a lot more possible in Solidworks

                    Currently i am using a solids with a standards mesh. i checked  the box for draft quality mesh. the global size is set to 20mm

                    With this settings i think meshing only takes a minute and its done.

                      • Re: Problem with gravity related study
                        Jared Conway

                        make sure you review the assumptions and limitations of draft quality

                        but yes, that is a good way to check your setup quickly

                          • Re: Problem with gravity related study
                            Sjoerd Post

                            Just to be sure, the draft quality mesh only has a node at each corner of the element?

                            If i uncheck this box, does solidworks automatically make a quality mesh, with a 10 nodes tetrahedron? So a node on each corner and a node on every middle of a line in the tetrahedron.

                             

                            *EDIT

                             

                            If i use the curvature based mesh, the meshing time is reduced enormously. Does the curvature based mesh has any effects on the quality of the mesh or the results?

                            I am still meshing with solid mesh.

                              • Re: Problem with gravity related study
                                Jared Conway

                                curvature based mesher is multithreaded = faster

                                quality of the results will be similar but note that it may concentrate mesh in areas differently than standard so you may have to go back and add mesh in important areas that are different than standard

                                and you have to remesh if you change the mesh settings, it doesn't automatiycally remesh

                                and your definition of draft vs high quality seems reasonable, check the help for visuals