2 Replies Latest reply on Jan 4, 2015 9:21 PM by Jared Conway

    Best Practice for Modeling Bolts Instead of Using Bolted Connector

    Bjorn Sorenson

      I have a model with a couple of plates tack welded together so that holes can be drilled and tapped through both plates and they line up, threads are continuous, etc.  There is another component with through holes that gets bolted to the threads in the two welded plates.

      I have tried running a static simulation on the assembly replacing all of the bolts with the bolted connection, and I'm getting a classic singularity on an edge of one of the welded plates in the vicinity of one of the bolt holes.  I've tried radiusing the corner of the plate and this just moves the singularity closer to the bolt hole.  Unfortunately, this region of the model is my area of interest, so I'm thinking about modeling the bolts themselves (in order to remove the high-stress error introduced at the bolt hole region by Simulation's connection method) and am wondering about the best way to apply the preload to physically-modeled bolts.

      I came across this:  SolidWorks Simulation: Bolted Joints in FEA - Sim3 Engineering Consulting , and the Split Bolt method looks promising.  Does anyone have any experience doing it this way or have their own preferred method?

       

      Thanks!

        • Re: Best Practice for Modeling Bolts Instead of Using Bolted Connector
          Seckin Uslu

          You can design a bolt like you referred in other software (hypermesh, lsprepost.)

           

          However in the solidworks simulation you can only define a bolt like parameter. you can not choose nodes or faces, only edges.

           

          My suggestion is , learn the result force on the bolts faces and manually calculate it (Stress=Force/ area)

          • Re: Best Practice for Modeling Bolts Instead of Using Bolted Connector
            Jared Conway

            split the bolt and apply forces like that article shows

            or you can split the bolt and apply a temperature difference to create the preload (essentially what the bolt connector is doing)

            i don't know if that is going to help your situation much, you're creating the same loading condition and your geometry/area of interest isn't changing and it sounds like it has elements that are getting torn, creating a singularity. but it is worth a shot. i have a feeling you're probably not going to get the resolution of the contact that you want which is going to create other issues for you and to minimize those effects would recommend going to nonlinear.