3 Replies Latest reply on Jun 6, 2013 12:10 AM by Jared Conway

    Weld size calculation for beams to solid, beam to beam

    Ben Moore

      Hi,

       

      I'm struggling to find a reliable method of getting Simulation to calculate weld size for beams (structural members) welded to other beams and also beams to solids. I understand that Simulation makes heavy use of shell mesh to calculate weld size but that doesn't suit geometry such as an I-Beam or RHS members. I try to make as much use of beam mesh's as possible for quick processing times and it is extremely useful when designing structural frames which use a lot of I Beam type profiles. When using RHS or SHS the corner treatment is often 'mitre'd' and 'joints' in simulation go a little crazy when I try to convert them to shells. I understand that the software needs to get reaction forces at an edge rather than a surface and it uses something like the throat shear method to calculate the weld size, but surely it can use the reaction forces at the end of a beam/structural member to do the same calculation?

       

      Is there a method that can be used to get Simulation to calculate the weld size in the above scenarios? Do I have to somehow convert the beams to shells then somehow connect them?

       

      It seems a big let down that the edge weld connector can only handle extremely simple restricted scenarios like a piece of round tube (converted to a shell mesh) welded to a base plate (solid mesh) as I've seen in the SolidWorks Simulation book I got from the course.

       

      Surely the boffin's at SolidWorks saw this question coming when introducing the edge weld connector?

       

      Thanks

      Ben

        • Re: Weld size calculation for beams to solid, beam to beam
          Jared Conway

          in a beam, the beam element only knows the section properties to calculate the stiffness. so when you connect other things to it, it doesn't actually know the actual geometry that would be involved in an weld connector.

           

          the edge weld connector is a pretty great feature that you would normally have to do a lot of manual modeling/calculations to set them up and get the same outputs. but like you said, it does come with some limited capabilities as documented in the help topic: http://help.solidworks.com/2013/English/SolidWorks/cworks/HIDD_EDGE_WELD_CONNECTOR.htm?id=92d7a123af53418ca0dc7883f53ee244#Pg0 But that doesn't mean you can't submit an enhancement request and see what the developers can come up with in the next version. They are constantly adding features and are pretty creative people. If there is a solution, they will find it. But keep in mind, they get hundreds, thousands of requests and need to prioritize based on need.

           

          To get through your current issue, you'll need to go with shells/sheetmetal/solid as required for your connection type. Depending on your situation you could go with a hybrid model where your edge weld connector is defined on a shell section and then between those section you have a beam. you just have to validate the connection between them works as you expect. (remember, the beam doesn't know the exact geometry of the beam)

          • Re: Weld size calculation for beams to solid, beam to beam
            Ben Moore

            Hi Jared,

             

            Thanks for the reply.

             

            "To get through your current issue, you'll need to go with shells/sheetmetal/solid as required for your connection type. Depending on your situation you could go with a hybrid model where your edge weld connector is defined on a shell section and then between those section you have a beam. you just have to validate the connection between them works as you expect. (remember, the beam doesn't know the exact geometry of the beam)"

            -> does this mean I break each beam end up by using the split feature, convert the ends into shells and then somehow attach the shell to the beam? Then use the edge weld connector to calculate weld size to some other part?

            How would the software treat the split? Would it transfer the correct stiffness, reaction etc results from the "beam" side of the beam to the "shell" side of the beam?

             

            Thanks

            Ben