2 Replies Latest reply on Feb 1, 2012 1:43 AM by Unspecified Unspecified

    thermal stress simulation

    Unspecified Unspecified

      Hi,

        My company had just bought a premium version a month before.  As I am not so familar on using the simulation module, I would like to ask for some suggestions.

       

        Currently, I am doing a project on caculating the thermal stress of a bi-metallic strip under cooling.  From the training manual, I know roughly how to do it: just do a steady state thermal simulation first, and that attach a thermal load during the static part.

       

        But I have some questions that I make me don't know how to handle the case:

       

        1)  If the materials of my bi-metallic strip has a temperature dependent factor, how should I handle to run the solution?  Currently, the thermal expansion coefficient (CTE) of my material will be vary from time to time during the cooling process.  And I think this will affect the utimate thermal stress stored in the components.  So how can I make the settings/ what can I do so that the simulation program could calculate the effect?  (Note: I know how to set / import a temperature dependent curve, but I don't know how to set the software to handle the data)

       

        2)  In my case,  I think how fast I cooling would affect the magnitude of residual stress.  So, how can I use the simulation to handle the effect of cooling rate/ convection coefficient?

       

        Looking forwards for your recommendation,

       

        Best Regards,

       

      Aruthor

        • Re: thermal stress simulation
          Jerry Steiger

          Aruthor,

           

          Since nobody else has replied, I will give it a shot. I'm not at all familiar with the way Simulation handles this (I mostly use ANSYS) and I have no experience with thermal analyses, so I won't be much help.

           

          On your first question, are you changing the temperature so much that the CTE changes significantly? If so, you may have to pick an average value for the CTE. It seems to me that would probably be adequate, if you can't figure out how to tell the system to use a temperature dependant curve.

           

          On the second question, I don't believe that the cooling rate will affect the residual stress, unless you are cooling at really extreme rates. I think you would have to cool it so fast that the surface would shrink so much faster than the interior that you would yield the material just under the surface. I doubt that any normal cooling process would result in that situation.

           

          Jerry Steiger