8 Replies Latest reply on May 1, 2013 5:02 PM by Jared Conway

    Thermal Expansion Analysis

    MIKE SCHULTZ

      I am trying to run FEA on an adapter assembly but I am having issues with fixing the end conditions.  The Stress is high since it is trying to hold a pipe that needs to expand.  These end conditions have to be fixed/fixed guided in order to determine the correct results.  Anyone have an answer?

       

      Boundary conditions:

      Adapter material: 4140 steel

      Ambient Temperature: 68 degrees F

      Thermal load: 500 degrees F

      Fixtures analysis 1: Fixed, Guided sliding connection with x and z axle restrained.

      Pipe wall thickness: .38”

      Pipe ID: 1.25

      Results required: Displacement, safety factor and resultant forces.

      fea.jpgexample deflection.jpg

        • Re: Thermal Expansion Analysis
          Bill McEachern

          restrain them axially and let them expand radially. You may also need to tie up a single node to avoid free body motions.

            • Re: Thermal Expansion Analysis
              MIKE SCHULTZ

              Bill,

              Thanks for the reply.  Is there another way to preform this analysis without guessing everytime at the translations distance?

              Untitled.jpg

                • Re: Thermal Expansion Analysis
                  Bill McEachern

                  I am not follwoing that question? In your image you are restraininghte radial direction. In my expereince in piping you would leave that free as whatever it is attached to will expand that way as well.I would restain the axial driection onthe flat face and then restrain it from moving in the plane without invoking unreasonable stresses due to Poissons ratio issues.. Say Tie up the circufrential and a single node or something. split it down the middle and use a symmetry on the plane - might have to split it twice.....

                  • Re: Thermal Expansion Analysis
                    Jared Conway

                    Hi Mike, you're not guessing. If you put a value, you're telling the software to move it. If you enter zero, you're fixing that DOF. If you don't click it, it is free to move.

                     

                    Something that might help us is a FBD of what you want the restraints to look like.

                     

                    Also, in general, if your restraints are causing you to over constrain at a specific location, you need to add the next part in the analysis.

                      • Re: Thermal Expansion Analysis
                        MIKE SCHULTZ

                        When fixing a body with a 432degree F differential the thermal expansion fails (aka stress is to high) at the fixed restraints.  Our design intent is to flex the adapters to account for this thermal expansion.  That help??

                         

                        Untitled3.jpg

                          • Re: Thermal Expansion Analysis
                            Jared Conway

                            Hi Mike, how did you apply the thermal condition? is it applied to all of the bodies?

                             

                            When you say guided on the X and Z, is that fixed? Seems to me that guided would mean move but along a specific path. Does move mean free to move?

                              • Re: Thermal Expansion Analysis
                                MIKE SCHULTZ

                                Reference temperature is set to 68 degrees.  All bodies are set to 500 degrees.  Yes it is applied to all the bodies.  It is free to move on the Y axis but it is guided/sliding/rolling in the X and Z axis.

                                  • Re: Thermal Expansion Analysis
                                    Jared Conway

                                    What I would do is cut this down the middle to leverage symmetry. That will remove part of your issues.

                                     

                                    Then it sounds like you need to soften up the restraint on the fixed end. I would start with restraining the cylindrical edge in the axial direction and then like bill said, one point. That might be enough to hold things together. If it fails, what is the next part in the assembly and can you clamp that one down so that the part you're interested in can expand? This part is going to be trial and error to get a restraint that allows it to run. You could also try inertial relief but make sure to read the documentation on how that option works.