14 Replies Latest reply on Nov 7, 2012 6:57 PM by Kevin Corr

    Bolt Connector in Thermal Study

    Peter Biggert
      I bolt two parts together using bolt connectors. If I do a thermal study the forces in the bolts go through the roof. Apparantly, because they are still at the stress free temperature while everything else has expanded with temperature. In real life the bolts would heat up too.

      Fine, I could model the bolts. But the old method to preload them was to specify a temperature difference between the bolts and the part. In a thermal study, I am not sure how to get the initial preload and still have the bolts reach a steady state temperature with their surroundings.

      Does anyone have suggestions on how I can perfom a static thermal study on parts bolted together?
        • Bolt Connector in Thermal Study
          Devi Prasad Samal
          Actually i caould not understand what you looking for....

          but this is a suggestion....

          if you want to go for thermal study. then you can define two parts with different temperature as required. and in option you can put your actual room temperature.

          you got the output as final temperature.

          after completion of this thermal study again took the study results to static study and do a stress analysis with preload condition of bolts. and you got your desire stress analysis.

          may be this helps....

          if this is what you thinking then reply if you have some doubts.

          otherwise please help me to understand your problem.....
            • Bolt Connector in Thermal Study
              Peter Biggert
              Let me try to explain my issue better.
              I perform a thermal study which calculates a temperature distribution across my parts. The no stress temperature I use is room temperature. My parts, in use, range between 175°F and 350°F. The temperature varies uniformlly from the high temperature to the low.
              I use the results of the thermal study in a static load study which also includes these bolt connectors. Because the initial preload is given at room temperature and the bolt connectors are not affected by the thermal study, the thermal expansion of the parts adds to the bolt load. In reality the bolts would heat up too.
              OK, I think to myself. If I actually include bolts in my model then the thermal study will change the temperature of the bolts. However, how do I put in the initial preload on my modeled bolts. If I specify a temperature for the bolts to get the preload then either this temperature gets overwritten by the thermal study or it doesn't and temperatue I specified to get the preload is a couple of hundred degrees less than the parts.
              Does anyone know how to preload bolts and still have a thermal study affect them?
            • Bolt Connector in Thermal Study
              Model in interference between the bolt head and part you are bolting together and use shrink fit contact set between them.
              • Bolt Connector in Thermal Study
                No, shrink fit contact sets only work with interference being modeled in.

                I did try a negative bolt preload thinking that you could calculate the reduction in preload to thermal expansion of the bolted connector, but did not accept a preload of less than zero.
                  • Bolt Connector in Thermal Study
                    Bill McEachern
                    I haven't been following this too closely so I might be missing the target with this one but you could try to model in the bolt as a solid but cut the shank of the bolt and then use a spring connector to ensure the preload is dealt with so that the change can be seen in the thermal stress analysis. It might require a space but it might not.
                      • Bolt Connector in Thermal Study
                        Peter Biggert
                        Good idea Bill!
                        I will have to experiment a little to see if this works. This may also allow me to model the bolts as beam elements too (If thermal results can be applied to manually entered beam elements). Still a lot of unanswered questions but I have something to try.
                          • Bolt Connector in Thermal Study
                            Hello Pete, let me have some questions to you:
                            1) have you progressed on the idea of bolts schematised with beams + springs?
                            2) are you sure (see a preceeding post of yours) that the bolt connector temperature can be affected by a thermal study? Perhaps you add the connectors in the mechanical study and afterwards you copy the mesh in the theraml study? Even so, I am not able to make it work
                            3) Have you an idea of the reason why we can give in input to SW the thermal expansion coefficient to the bolt connectors but not the connector temperature?
                            Thank for your attention
                            Leonardo
                              • Bolt Connector in Thermal Study
                                Peter Biggert
                                Leonardo,

                                To answer your questions...

                                1) I ended up using Wayne Matus' suggestion of "Model in interference between the bolt head and part you are bolting together and use shrink fit contact set between them."

                                I modeled the bolt as a solid. (no beam elements used and no springs). Just a single body shaped sort of like a dumbell.

                                2) The built in "bolt connector" was not being affected by a thermal study. That is why I started this thread back in March 09. I did a mechanical study first and copied the mesh to use in the thermal study and the built in "bolt connector" was not affected by the thermal study. So I am not surprised you could not get it to work either.

                                3) The authors of the simulation package have not yet gotten around to allowing thermal results to be applied to bolt connectors. It would be a welcome addition.
                                  • Bolt Connector in Thermal Study
                                    Thank you Pete for your prompt and valuable reply
                                    • Bolt Connector in Thermal Study
                                      Hi, Pete,

                                      Could you please give more details on how you modeled the bolts?

                                      I need to physically model the bolts (using solids) with preload of 11 lb-in torque. In my case, I am not modeling thermal.

                                      In my assembly, I also need to apply a load on one of the parts.

                                      Thanks!

                                      Pei
                                      • Re: Bolt Connector in Thermal Study
                                        Kevin Corr

                                        at the risk of beating a dead horse, this thread being ancient, I did want to add what I've learned over the times I used bolt connectors with thermal effects successfully predicted. It is a half truth that the connector behaves as a rigid bar in thermal strain since the beam element actually does expand or contract with temperature change. This applies when the static study uses a thermal study's resulting temperature. The solver does an average of the nodal temperatures of the head and nut contact surfaces and uses that single value to compute the variation in length of the bolt due to temperature. The truth is, the bolt head and nut are modeled as rigid bars connecting the beam element to the mesh being clamped by the bolt. So, the thermal effects of the head and nut are not well simulated by bolt connector since the bolt head is not rigid and rigid bars don't "give" the way bolt heads and bolt nuts do.  That is one possible reason we see extremes in stress and displacement near the bolt head - clamped partcontact region.