5 Replies Latest reply on May 27, 2014 4:10 PM by Bill McEachern

    Improved Mesh Control Area Selection Tool

    Peter MacDonald

      An informal question for the experienced users.

       

      Has there ever been a suggestion put forwards for SW to improve the mesh control tool? I'm thinking it would be very handy to be able to 'dynamically' sketch control areas on the model (rather than creating split surfaces). These sketched areas would only exist in the assembly (or in that particular simulation).

       

      I imagine the above tool would be extremely hard to implement, but I'm curious to know if anything like this has been suggested.

        • Re: Improved Mesh Control Area Selection Tool
          Jared Conway

          have you checked the solidworks KB? all enhancements that users have submit are located there

          that sounds like a good one, theoretically you could mesh and then paint the mesh

          • Re: Improved Mesh Control Area Selection Tool
            Bill McEachern

            I have put forward that they have a complete decoupling option between CAD and analysis to reduce overhead and increase performance - never seemed to get any traction. They have the ability to add spit lines in the loads specification on the fly - can't say I have used it though. You could use this and then just add some insignigicant load and then use it for mesh controls.

            • Re: Improved Mesh Control Area Selection Tool
              Shaun Densberger

              Not a bad suggestion at all; Dassult needs to put more focus on meshing in SW. I'm still absolutely baffeled that SW Simulation, an h-method code, doesn't allow the user to create quad/brick elements.

                • Re: Improved Mesh Control Area Selection Tool
                  Bill McEachern

                  Well quads would be really easy to add and be very helpful in shell work.  Many people have asked that know what they are about in analysis. However, when they ask the typical user, those users don't see the need because they don't know they need them as there are so many new users to FEA via SWX Sim. This is a good thing on the one hand but one can quickly get to limiting functionality and to keep people on side they need a place to go.

                  On quads: the surface functionality of swx is not all that sturdy in my view -things can get wonky pretty quick and the behavoir can change significantly from release to release. So while adding quads would be good it might not get you were you want to go. 

                  Bricks might be a big help in NL contact problems but I am a long way from convinced of that. From what I can surmize by observation: they just don't (never really have) view this tool as being something worth investing in with respect to technology (when you also own ABAQUS why would you?) - they just do UI pieces which they are pretty good at really (setting aside the bugs that often occur in the UI  functionality). It would be better if they could write out good input decks for ABAQUS and load ABAQUS output files for the NL functionality and let the Cosmos stuff be a low cost linear code with a way to migrate to better technology. You could keep the UI and then just buy some ABAQUS analysis tokens and enjoy watching NL problems solve in a hurry and far more robustly. It all sort of becomes a bit moot with the advent of the 3DX platform though......