7 Replies Latest reply on Mar 3, 2017 9:04 AM by Amit Katz

    Solidworks is reliable?

    Nestor Hernández Domínguez

      Hello, it's my second post in this fantastic community.

      I have a doubt, or rather a healthy discussion, I study Ing in Instrumentation and Control, I am using solidworks to simulate the behavior of a hot air flow in an incubator, the simulator tells me that I will have problems in evenly distributing hot air In the whole incubation chamber, this is known as temperature homogenization.

      The problem is that the industrial process teacher does not trust the results of solidworks. I look for official information on how solidworks works, but the teacher still thinks the same. The teacher is accustomed to using Matlab, what should I do to show you that solidworks is reliable?

        • Re:  Solidworks is reliable?
          Jim Sculley

          You can view the technical reference material from inside SOLIDWORKS:

           

          • Re:  Solidworks is reliable?
            Wojciech Paterski

            maybe do some 'live' experiment and then model it in SW and compare results - to show it they are the same (if they are)

            • Re:  Solidworks is reliable?
              Jim Steinmeyer

              Nestor,

              I would suspect that the instructor is not questioning Solid Works as much as he is questioning the process. What I mean is that the software output is only as accurate as the information put into the system. The software is proven to work if the conditions are input accurately and if the correct assumptions are used. This can be very difficult to do when you are first learning the software.

                   I would suspect that the only way you will convince him of the accuracy is to run several simulations with known results in both Matlab and SolidWorks and compare the results. When the results become consistent between the two programs he will begin to trust SW and you will have learned the flow simulation package well enough to be able to trust it and your use of it as well.

               

                   Then comes the question, how accurate are the Matlab results? Again the only way to prove accuracy is to run several simulations against a verifiable test in the lab and compare the results.

                • Re:  Solidworks is reliable?
                  Nestor Hernández Domínguez

                  Thanks for your answer Jim, maybe you are right, I personally asked the teacher why I had this criterion on solidworks, or what was your criteria for not trusting the software, if you did not know to give me a clear answer, I just mention that it seemed to you The software solved the process in a linear fashion. He asked me how it was possible for solidworks to simulate the process if I did not have the mathematical model of the system yet. I'm using the Flow Simulation plug-in in Solidworks 2016, explain that the simulator gives you your boundary conditions and according to the data you provide to the software, you simulate the process.

                  • Re:  Solidworks is reliable?
                    Amit Katz

                    Well said Jim! This software is a complicated tool, it's up to the user to put it to good use and verify his/her results.

                  • Re:  Solidworks is reliable?
                    Gian Flavio Violi

                    Hola Nestor, Cómo estás?

                     

                    I had this problem all the time while at Engineering School. Teachers at my university were all about ANSYS and didn't trust SolidWorks at all. What I did was showing them the technical reference and the calibration results against NAFEMS (for FEA), and made sure to really nail the explanation regarding my approach to the problem. I did end up using what Jim suggested, 'validate' SolidWorks results against their software of preference sometimes and then do my own thing, but it can be time consuming.

                     

                    But as from an academic stand point, you can show him the technical reference, really put a lot of work into showing why you simulated the problem that way and most importantly, interpret the results correctly (does it make sense and why?). That would be my initial approach.

                     

                    Regards,

                    GF