8 Replies Latest reply on Jan 30, 2015 7:52 PM by Jared Conway

    Transient thermal problem setup

    Malcom Smith

      Hi there,


      I'm new to Solidworks Simulation and would appreciate some help understanding what kind of study to run for a transient thermal problem.


      To help me learn I made a simple model of two metal blocks side by side with a small air gap between. One block starts the simulation with an initial temp of 4 deg C. the other at an initial temp of 40 deg C.

      My expectation was that the cold block would heat up to ambient and temporarily cool down the hot block next to it across the air gap. I set convection on all faces to be 50W and ambient to 40 deg C (so that the hot block is the same as ambient).

      But all that happens is the cold block warms up to ambient without affecting the hot block...


      I read the solvingengineeringproblems and tutorial docs but cant see what I'm doing wrong.

      Any help greatly appreciated.

        • Re: Transient thermal problem setup
          Mike Pogue

          Hi Malcom,


          From your description, it sound's like the problem is that the blocks aren't touching. You have to apply a thermal resistance contact condition between the two blocks. You have to look up the thermal resistance of the air gap, which I believe is L/K, Where L is the distance between, and K is the thermal conductivity of air. You should get units of K*m^2/W. (K = Kelvin). SolidWorks can't see or account for the air. It will assume parts that are not touching have no interaction.

            • Re: Transient thermal problem setup
              Malcom Smith

              Thanks for the tips guys!


              The application is for a closed volume oven where cold blocks are introduced over time. The blocks do not touch each other but in reality newly added cold blocks pull down the temperature of blocks already in the oven at a higher temperature. The distance between blocks changes depending on placement and I'm not sure how I can fake that for each face of each block.


              Is a thermal simulation the right tool of this? Perhaps a flow simulation would be better in this case?

            • Re: Transient thermal problem setup
              Jared Conway

              solvingengproblems and tutorials are for flow simulation, you're using simulation professional


              like mike said, you have to "fake" the air in sim pro with either convection coefficients or a thermal resistance. otherwise any face you do not apply a BC to is considered insulated.

                • Re: Transient thermal problem setup
                  Malcom Smith

                  Hi Jared,


                  When you say fake the air is that in a global way, i.e. I can enclose a volume around the blocks and give that a thermal resistance or do you mean define a thermal resistance between all faces of all blocks? Adding convection coefficients to each blocks faces allows energy to leave the block but not enter nearby blocks.

                    • Re: Transient thermal problem setup
                      Mike Pogue



                      I think the effects between the blocks will be more or less negligible, since the oven has the ability to maintain its temperature.


                      The convection coefficients won't affect energy transfer between the blocks. Energy transfer between the blocks will only come from a contact condition. The convection coefficients will only simulate energy transfer between a thermal reservoir (the oven) and the blocks. You'll also have to calculate the convection coefficient by hand; and it is not a constant, but is a function of temperature. It's different on the sides than it is on the top than it is on the bottom. You will also probably find that the convection coefficient you calculate will be so inaccurate in air that it will invalidate your analysis. If you just wanted energy transfer through a thin air gap, you'd be on pretty solid footing here. But if you want to simulate convection in an oven with two different blocks, you'll probably need flow works


                      If you want to push ahead, though:

                      Calculate the convection coefficient curve by hand for the sides and the top and the bottom.

                      Apply the convection boundary condition to each surface of the block

                      Calculate the conductive resistance between the blocks, assuming some distance

                      Apply the thermal resistance contact condition between the blocks.

                      Let it rip.


                      I would expect the errors in this analysis to be large. But it could produce some useful qualitative information depending on what you need it to show.

                      • Re: Transient thermal problem setup
                        Jared Conway

                        fake = not solve the cfd/convection problem, apply a convection yourself


                        a solid around it might be another approach, but not as good as cfd in this case

                    • Re: Transient thermal problem setup
                      Attilio Colangelo

                      Essentially you have a convection model of a 4C solid in a 40C fluid.  The hot block is not participating since it is already in equilibrium with ambient (40C) and no way to transmit heat to the cold block.  Reading between the lines it looks like the 40C hot block creates the 40C ambient (outside of the simulation). 


                      Turn on radiation and things will get a little more interesting (even at these temperatures) between the blocks.   Another effect that could be included is the decay of the convection coefficient as the cold block heats up.  Beyond that you will be entering CFD domain.