General Mechanical

General Mechanical

Topics related to Mechanical Enterprise, Motion, Additive Print and more.

Thermal analyses, correct approaches?

    • icostin
      Subscriber

      A thermal analysis, specifically a temperature distribution, is requested for a situation involving conduction and convection. The scenario involves insulation of a pipe exposed to an interior temperature of 600 degrees Celsius and an ambient temperature of 20 degrees Celsius with a convection coefficient of 9 W/m^2°C. There are two approaches: 1. Introducing a point thermal conductivity value, namely 0.119 (for the interior temperature of 600 degrees Celsius), and case 2, where the conductivity is introduced linearly, as shown in the figure. Is there someone who can specify which of the two cases is the correct approach? 

      1. T=600 C(on inner line), and thermal conductivity at 600 C is 0.119 (for calcium silicate insulation)
      2. T=600 C(on inner line), and thermal conductivity is inserted according with fig2. (temperature dependent) (for calcium silicate insulation)

         Why do the results still differ, in the sense that the exterior temperature in case 1 is 131 degrees Celsius, while in case 2, it is 113 degrees Celsius?

          ANSYS interpolates thermal conductivity values based on temperature (during the analysis)???. 

         So conductivity values must always be introduced as distributions rather than point values?

         Stady state thermal approach is ok

       

       

       

    • icostin
      Subscriber
      1. T=600 C(on inner line), and thermal conductivity at 600 C is 0.119 (for calcium silicate insulation)
      2. T=600 C(on inner line), and thermal conductivity is inserted according with fig2. (temperature dependent) (for calcium silicate insulation)

         Why do the results still differ, in the sense that the exterior temperature in case 1 is 131 degrees Celsius, while in case 2, it is 113 degrees Celsius?

          ANSYS interpolates thermal conductivity values based on temperature (during the analysis)???. 

         So conductivity values must always be introduced as distributions rather than point values?

         Stady state thermal approach is ok?

       

    • peteroznewman
      Subscriber

      Approach 2 using Temperature Dependent material properties is the correct approach. With a high constant thermal conductivity, heat can more easily move from the inner surface to the outer surface, resulting in a higher temperature.  With the thermal conductivity dropping through the thickness with temperature, heat flows to the outer surface more slowly allowing the convection to cool the surface to a lower temperature. Yes, ANSYS wil interpolate for values between 400 C and 600 C.  If you have thermal conductivity data below 400 C, it would be better to add some lower temperatures into your material model, all the way down to 20 C if you have it because with no data below 400 C, it just uses the last value in the table for all temperatures below 400 C.

      Steady State Thermal is the correct analysis to run. 

      • icostin
        Subscriber

        @peteroznewman, Thank you very much for the message and especially for your promptness.

Viewing 2 reply threads
  • The topic ‘Thermal analyses, correct approaches?’ is closed to new replies.