7 Replies Latest reply on Jul 27, 2018 9:56 AM by James Riddell

    Static Simulation Bolt Preload Yielding

    Matthew Zola

      I am using a static simulation to determine whether I am using a sufficient number of fasteners for my load case. I am currently using 18-8 stainless steel socket head screws into a 6061 aluminum frame. The screws are 4-40 in size and are called out to be torqued to 5.2 in-lbs. When I make a simple static simulation as shown below, the FOS on the screw is already 0.95. If I then put 50 G's of gravity on the assembly, the FOS goes up to around 0.96. I am using the same methods on my larger assembly and am getting similar results. What I am wondering is how can I tell if these screws are sufficient given that the preload is already so close to yield and thus a low FOS?

      I have read through this thread from 2011 (including the theory behind a bolted joint), and what I took from it was that I need to create two simulations - one with just the preload and one with the applied load. If the axial force on the fastener after running the simulation is greater in the applied load case, then the bolted joint has yielded. Again, I am not sure whether this logic is correct (and if it is correct, how much of a deviation from the pre-load only case is indicative of a joint failure?), or if I should be determining this in a different way.  I have also attached the simple simulation and parts that I am using to help explain what I am doing.

       

      Thanks in advance for the help! Let me know if you need clarification on anything, I understand this post can be a bit hard to understand.

        • Re: Static Simulation Bolt Preload Yielding
          James Riddell

          First, what are you doing putting stainless steel in contact with aluminum?

          Second, this really should be a hand calculation unless you are researching/validating bolt strength.

          Third, does your FEA take into account thread root diameter/stress riser?

            • Re: Static Simulation Bolt Preload Yielding
              Matthew Zola

              Hello! The parts I am building are actually for a small satellite and it is standard for stainless steel fasteners to be used in to aluminum structures. Due to the space application and small amount of steel compared to aluminum, galvanic corrosion is not a top concern.

               

              In regards to the hand calculations, I see where you are coming from, but I am unsure as to how to get the expected loads on each of the screw joints. I have included a (very) simplified version of my full structure below to help explain what I am doing. My end goal is to ensure that the bolt quantity and placement will be sufficient to survive up to 50 G's from the launch environment. I would like to test the structure in multiple configurations, as I do not yet know how the structure will be oriented during launch. Consequently the loads on the bolts could be compression, tension or shear depending on its orientation. How should I model the screw joints in order to get the axial and shear forces on each of the joints in order to determine if the fasteners are sufficient via hand calculations?

               

              Thanks for the quick reply!

               

                • Re: Static Simulation Bolt Preload Yielding
                  James Riddell

                  Ah, there is not as much moisture/air in space.  (There is no air in space but there's an Air & Space museum! ~ H. Simpson)

                  Next question - what do you expect the loading through the screws to be?  Since this appears to be some sort of lid could something inside come loose and impact it?  Is the 'box' supposed to be fixed vertical (w.r.t. the lid) during the 50 G loading?  Are you interested in the shearing of the screws (easy to check)? 

                  I'd recommend imagining the worst types of loading and combine the stresses.  It is fairly straightforward to get the tension, bending, shear, and torque loading, then combine those stresses and see if you end up with a suitable SF/UC/MS value for the given material.

                    • Re: Static Simulation Bolt Preload Yielding
                      Matthew Zola

                      There should never be a case where something comes lose or impacts the structure - it will be purely g-loading from the acceleration of launch. The entire structure is supported by a frame that comes into contact with all of the side edges, as pictured in green below. The front/back are supported by feet that extend to the blue squares also pictured below. As a result of this support structure, there is not much movement allowed, but I still need to make sure that the screws on the lid (as well as a bunch of smaller screws that I am not including in my simplification) can survive the 50G load.

                       

                      As you described, I plan on orientating the 50G load in what I would consider the "worst case scenarios" to get the highest expected loads in tension, bending and shear (yes, I am interested in shear). How should I model the screw joints (in terms of the concentric holes, as well as surface bonding/penetration) to get these values?

                       

                        • Re: Static Simulation Bolt Preload Yielding
                          James Riddell

                          If the box rests on those blue areas during the 50 G loading then it's simply a beam (looking from the side).  Worst case shear is ~50x the total weight spread over the screws.  Don't forget to utilize your friction force from the screw pre-load if you are close to shearing the 4-40s off.  Conservatively, you have three screws on each green side holding the load (the 'shelves' can flex).

                            • Re: Static Simulation Bolt Preload Yielding
                              Matthew Zola

                              Thanks for the help! I will work on determining the highest load case for shear, compression and tension and go from there. For the hand calculations, is it sufficient to go with a worst-case scenario and assume the screws diameter is equal to the minor diameter of the threads (for 4-40 it would be .0805"), or do I need to do take into account the stress propagation from the "fillets" on the threads?

                                • Re: Static Simulation Bolt Preload Yielding
                                  James Riddell

                                  If I'm imagining the approximate size of your construction (given #4-40 HSCS) then, IMHO, you don't need to worry about the stress riser from the thread form (unless you get some huge stress numbers).  That said, for a properly designed bolted joint you would NEVER place the threads in shear, the hole would  be counter bored similar to the location you have for the screw head and you'd either use a longer screw or a shoulder bolt.  There are plenty of sources for bolted joint design.  Good luck and let us know if your design takes off!