2 Replies Latest reply on Jul 12, 2013 11:53 AM by Jared Conway

    Transient Thermal Study Failing to Complete When Using Temperature Dependent Cooling Curve

    Scott Hopkins

      Hi SolidWorks Users

       

      I have a transient thermal simulation problem when using a temperature dependent cooling curve & was interested to see if anyone has had the same problem or if someone can suggest how to over come the problem.

       

      In my transient study, I have applied temperature dependent materials to all parts (the study is of an assembly) & have applied various heat power & heat flux loads to these parts.  I have also set the initial temperature of all the parts to 300K & set the total study time to 28,800 secs (8 hours) with the set to time increment at 14,400 secs.  All parts in the assembly meshes fine, but when I run the study, I get the following error message:

       

       

       

      I have since tried a few other total study times & found that it will run at 21,600 secs (6 hours), but anything longer than this, I get the same error message.  The range on the cooling curve I use is 410W @ 300K to W @ K.  The study does not run to conclusion as I know the temperature the coldest item is approx 20K & I know it would take probably 36-48 hours to reach this temperature.  I have triple checked all my materials data & heat loads & they are all fine.  I was using SW2011 & have now upgraded to SW2013 SP3 & the problem still exists.

       

      Has anyone experienced the same problem or can anyone suggest what the problem might be?

       

      Thanks in advance.

       

      Scott

        • Re: Transient Thermal Study Failing to Complete When Using Temperature Dependent Cooling Curve
          Scott Hopkins

          Sorry I forgot to add the error message I am getting.  It is:

           

          The Iterative Solver stopped.

          Numerical difficulties.(Negative diagonal term)

          Possible large variation in thermal material properties.

          No results saved.

           

          In addition, the correct range on the cooling curve is 410W @ 300K to 4W @ 12K.

           

          Scott

          • Re: Transient Thermal Study Failing to Complete When Using Temperature Dependent Cooling Curve
            Jared Conway

            so if i understand you correctly:

            1. the results for the shorter duration don't match your expectations

            2. you can't get it to run the full length that you want

             

            i'd focus on #1 first. how do you do this? bring the problem down to a more simple problem to start. get rid of the temp dependent materials, get rid of transient, get rid of the heat input curves. does a basic problem solve?

             

            then you can start adding complexity. sounds like to me that you've got an input problem because i've run similar studies with those conditions with success before. but potentially your combination is affected by a software issue.

             

            something to consider. rather than waiting a long time for something to have an issue. you can consider breaking your thermal analysis until multiple studies. if there is a limit on your study, you could do 6hrs in one and 2hrs in the other.