4 Replies Latest reply on Feb 17, 2012 12:32 PM by Daen Hendrickson

    Simple Simulation... I think, but apparently not for me!

    Daen Hendrickson

      I am using SW2011 SP3.

       

      I have never used simulation / Cosmos and finally had need to run - what I think to be - a quick static test. I have seen the videos: just click here, tell it this and that, pick one of these and click run. Instant results. Not happening for me. So with hat in hand, is anyone willing to take a look at this?

       

      I need to lift a server cabinet. The top of the cabinet has 3/8-16 threaded holes intended for hoisting with a load rating (per the cabinet mfg) of 500 lbs each hole. The holes are at each corner.

       

      The attached model is of my tricked out lifting fixture - a piece of 2x2 square tubing (mild steel) that bolts to the cabinet rigging holes and overhangs each side so I can use a fork lift. In my attempt to get the simulation to work, I added some temporary bodies to represent the forks, and then the cabinet below.

       

      I have no idea how much my cabinet weighs, but lets assume worst case and put 500 lbs load at each bolt hole. Does the lifting fixture fail?

       

      Thanks much. Daen

        • Re: Simple Simulation... I think, but apparently not for me!
          Dean White

          Hi Daen,

           

          I ran it in 2012 and had a few problems with the model. I did some tweaks and got a result. However I had to suppress the bolt connections!!!

           

          • I had to define the slot material in the model. right click Slots>Material>Edit Material
          • I excluded the cabinet part from the analysis and applied the force directly to the bolt holes
          • Another contact set between the forks and beam (bonded beam type again)
          • I added some mesh control to the beam to get better results (beam type again)

           

          I don't know if this helps as I mainly do thermal analysis, but hopefully it might give you some pointers.

           

          Dean

          • Re: Simple Simulation... I think, but apparently not for me!
            Daen Hendrickson

            I fought with this a while longer and finally stumbled across a setting that let the simulation run. SolidWorks assigned my model the designation of a "beam". Being a beam apparently restricts several things including being able to apply a point load (representing the bolt) to a beam face. Once I turned my model into a solid the simulation ran. I also just fixed the ends of the tube instead of successfully simulating the interaction of fork lift forks.

             

            I doubt my simulation was very representative of real-world conditions but at least I finally got it to run. I might even have time to revisit this and learn how to properly use the tool... someday.

             

            Thanks for the suggestions Dean.

              • Re: Simple Simulation... I think, but apparently not for me!
                Samuel Haworth

                The section you chose would handle 1000lbs of bending load without sweating, but the wall thickness will at least yield around the bolt heads, if not fail entirely at 500lbs per bolt.

                 

                You need to look at the square tube by itself, as you said, as a solid. Exclude the other bodies and change the global contact setting to allow penetration. As a general guideline, don't use no penetration contact unless absolutely necessary, since it will dramatically increase run time and the likelihood that you won't get any usable results.

                 

                Replace the fork blocks with split lines on the bottom face of the tube that represent the inner edges of the forks. The width between the forks will be a significant parameter to determine the bending stress in the tube. Apply a fixed restraint to one of the split lines (the line/edge, not a face), and apply a restraint only in the y direction to the other line. This will represent the worst case support, where the beam bends and is only supported on the edges of the forks. Make sure to only fix one end, because as the beam bends the split lines would tend to get closer together. Fixing both would prevent this from happening and make the beam act as though it were stiffer than it really is.

                 

                Finally, split a circular face the diameter of the bolt head or washer around each of the holes, and apply the expected load to the newly created faces. 500lbs per bolt, downward.

                 

                Mesh it fine since the tube's wall thickness is small relative to its other dimensions and you don't want to distort the elements more than you have to.

                 

                Run the study.

                 

                The Von Mises plot should show you two cone shaped deformed areas around the bolt holes, with peak stresses well above yield for mild steel. This doesn't mean it won't work, however. Static linear analysis can only accurately predict behaviour up to the yield point. What it does mean is that you will have some level of permanent deformation (up to and possibly including the bolts ripping out) in the tube after the lift, if the load is truly 500lbs per bolt.

                 

                If the bolt holes in the tube are actually slots this situation gets worse. I'd put some big, thick washers on those bolts if I were you.

                 

                -Sam