10 Replies Latest reply on Dec 11, 2013 1:43 PM by Kyu-Hyung Kim

    Flow simulation to improve heat sink design - Problem! No temperature gradient across fins!

    Jamshid Sadeghi

      Hello all,

       

      I am trying to do a study into LED heat sink design and performance in Solidworks with the Flow Simulation, but I am getting results which I think are wrong (although this is most likely to do with how I set up the study).

       

      Until now I had been using an external study, selecting Conduction in solids and Gravity from the wizard menus.

       

      I applied a volume heat source to the part that is the LED COB in the assembly.

       

      I then made a thermal joint between the bottom of the LED COB part and the surface of the heat sink which it is sitting on.

       

      Finally I selected all the exposed faces of the heat sink and set a contact resistance of ground aluminium and air.

       

      When I run the study and check the results the temperature throughout the whole heat sink is constant, i.e. it is, say, 100 degree C in ever part so there is no temperature gradient across any of the fins or in the sink.

       

      Of course this cannot be but I do not see where I have made my mistake. I have seen mention of a 2-Resistor component on some other disucssion but am not sure how to apply this.

       

      I am running the study using an assembly of parts, one of which is the heat sink and another one of which is the LED COB package. I have modelled this as a single part made of silicon, i.e. the entire package, not just the chip. Perhaps this is wrong?

       

      Really stuck here so any help would be really appreciated!

       

      Thanks in advance!

       

      Jamshid

        • Re: Flow simulation to improve heat sink design - Problem! No temperature gradient across fins!
          Jared Conway

          a picture is worth a thousand words here to understand what your model looks like.

           

          also, i'd suggest to start simpler. get rid of the thermal joint, get rid of the contact resistance. don't worry abotu 2R unless you have 2R data. modeling it as silicon is probably fine. i suspect the thermal resistance is your problem. not sure why you've selected that option for this type of analysis.

           

          i'd also recommend going through all of the tutorials if you haven't done so already. they built upon each other so don't cherry pick the ones that look like they apply.

           

          if you're still having trouble, please post the model or a report showing all the setup parameters.

            • Re: Flow simulation to improve heat sink design - Problem! No temperature gradient across fins!
              Jamshid Sadeghi

              Hi Jared,

               

              Thank you for your reply.

               

              I will put up a picture of the model shortly and hopefully it will help.

               

              Actually, I had started out with just putting together the assembly, applying the volume source and running the flow simulation, but still had the same problem. All the body of the heat sink just appeared as the exact same temperature (very high!) as everywhere else. I will try to put a picture of the result I get as well if it helps. It was at that point when I tried to compare the set up to the initial study that was done using the Simulation tab (a thermal analysis) where I had to specify the conduction between the chip and also the convection condition for the fins. So it was just a matter of hoping that trying to mimic this in the flow simulation study would fix the problem (sadly it didn't!)

               

              I have gone through most of the tutorials, but I will double check as you suggest and make sure I have gone through all the relevant ones.

               

              Also, another reason why I tried the contact resistance and 2R conditions was that I noticed in what I think is a promotional video from solidworks (that actually tackles exactly what I am trying to do) these items have been used. The link to the video I am referrring to is given below. The temperature distribution on the heat sink in this study shows a niec clear gradient, but on mine the whole thing is just equally heated to a unrealistic temperature!

               

              http://www.youtube.com/watch?v=PP0YgufCWBM

               

              As a simple starting point then, what would you suggest is the best way to set up the model for flow simulation? The simplest I can think of is what I started with which was a volume heat source and that was all (unfortunatley I know that this gave me a wrong result so perhaps this arrangement is too simple!)

               

              Thanks again for your help.

               

              Regards

               

              Jamshid

              • Re: Flow simulation to improve heat sink design - Problem! No temperature gradient across fins!
                Jamshid Sadeghi

                Hi Jared,

                 

                Please see attached images: the first one is just an image of the assembly, the second one is a picture showing the surface plot of temperature and the problem I am having.

                 

                lamp 1.JPG

                 

                lamp 2.JPG

                 

                 

                Hope this helps.

                 

                Regards

                 

                Jamshid