13 Replies Latest reply on Feb 6, 2014 12:22 AM by Jared Conway

    How to read simulation results??

    Brent Trotter

      Hi All,

       

      I am doing my first exploration of the simulation function for SW.

       

      I was hoping someone could tell me what the result mean of this test.

       

      Any help would be greatly appreciated.

        • Re: How to read simulation results??
          J. Mather

          I am curious - what do you think it means?

            • Re: How to read simulation results??
              Brent Trotter

              Well, as an esitmate guess I would say that this support is deflectin approximately 2.3mm in the centre of the span.

               

              But as I am not entirely sure I thought it would be best to ask.

               

              As I am selftaught at SW I need to rely on the help of more experience users.

               

              Am I correct?

                • Re: How to read simulation results??
                  J. Mather

                  That is what that image indicates (based on how you set up the analysis right-or-wrong).

                  Note that SolidWorks exaggerates the displacement to make it clear to see - you can set it to no deflection or actual.

                   

                  Right click on the analysis folder and select Factor of Safety.

                  What does it report as the minimu FOS?

                   

                  Attach your *.sldprt file here for more information.

                    • Re: How to read simulation results??
                      Mark Greenwell

                      I look at the picture and assume if he has it fixed as shown and has the applied load as shown, the structure would deflect 2.3mm at center.

                       

                      What do you think it shows.

                       

                      I don't think his question is about if every think is set correctly or what the FoS is he just wants to know what is happening in the picture.

                       

                      Regards

                       

                      Mark

                        • Re: How to read simulation results??
                          J. Mather

                          Mark Greenwell wrote:

                           

                           

                          What do you think it shows.

                           

                          I don't think his question is about if every think is set correctly or what the FoS is he just wants to know what is happening in the picture.

                           

                          Regards

                           

                          Mark

                          What did I write?

                          Hint:  Read the first 6 words of my response.

                            • Re: How to read simulation results??
                              Mark Greenwell

                              J.Mather wrote:

                               

                              'I am curious - what do you think it means?'

                               

                              This was your first response so why answer a question with a question?

                               

                              If he's any thing like me when he's stuck he will post a question hoping to get a quick reply.

                               

                              Not sure what you do for a living but if he's like me (I work for a large British Structural Steel Fabricator) he cannot spend too much time going backwards and forwards with posts when a straight forward answer is required.

                               

                              He knew the answer but just wanted conformation, also why would you need his *sldprt file to give him an answer?

                               

                              Regards

                               

                              Mark

                                • Re: How to read simulation results??
                                  J. Mather

                                  Mark Greenwell wrote:

                                   

                                   

                                  .... also why would you need his *sldprt file to give him an answer?

                                   

                                  If someone asks a question confirming FEA results then there is a possibility that the problem is not set up correctly. (OP states - this is first simulation attempt.)

                                  I would never make a definitive statement on only an FEA "pretty picture" screen capture, especially when someone asks the question that was asked.

                                  Thus,  "(based on how you set up the analysis right-or-wrong)", I would need the file. 

                                   

                                  I have to wonder, isn't this obvious?  Did you read Mike's response?  FEA is a complex topic to give any meaningful answer based on a single pretty picture.

                                   

                                  Can you tell from the single pretty picture how long the part is?

                                  A 2mm deflection might be significant or insignificant, depending on the overall length of the part.

                                  More information is needed.

                                  If a picture is worth a thousand words, the actual file might be worth a thousand pictures.

                                  • Re: How to read simulation results??
                                    J. Mather

                                    Mark Greenwell wrote:

                                     

                                    ... cannot spend too much time going backwards and forwards with posts when a straight forward answer is required.

                                     

                                    This is a discussion forum.  Discussion implies back and forth.   The best "answer" might be a journey that is the result of  the  input of many users.  I suspect anyone looking for an "answer" to something as complex as a FEA question would benefit from that discussion.

                                    I wonder how many engineers would affix their professional seal, based only on the information given?

                        • Re: How to read simulation results??
                          Mike Pogue

                          I think it means that the part will deflect the indicated amount (observe that your picture is not to scale) under these assumptions, which others can add to.

                           

                          The indicated stress is below yield

                          The material the part is made from is perfectly elastic, homogeneous, flawless, and has the Young’s modulus you assigned to it

                          Your part has been fabricated to the exact size and geometry of your model

                          Welds and other bonds are identical in properties to the surrounding material

                          Your model is convergent, and has in fact converged to the correct answer

                          You have perfectly rigidly bonded the surfaces indicated to perfectly a rigid fixture

                          You have found  way to apply a perfectly uniform pressure that is exactly the pressure you have applied to the model and which either

                          1. Acts downward regardless of the angle of the surface or
                          2. Acts normal to the surface and the deformed angle between the surface and horizontal is very very small

                          All forces and boundary conditions are static with respect to time

                           

                          It further means that you have to evaluate the extent to which each of these assumptions is true, and, where they are not true, the extent to which they affect your results.

                          • Re: How to read simulation results??
                            Shaun Densberger

                            Both Mike and J.Mather hit the nail on the head, and I have to respectfully disagree with Mark. I completely understand the time constraints associated with working (especially in the private sector), but a quick (and direct) answer completely bypasses the bigger picture. Depending on the circumstances, this could have drastic consequences.

                             

                            Brent, it's great that you're taking the time to learn a new tool and I think it is absolutely worth your time to do. However, keep in mind that the software is (mostly) only as good as the person using it. Mike brought up a lot of great points that directly effect the validity of your simulation, and it's important understand what effect each has on the results you obtain. Perfectly homogenous material? Probably OK. Perfectly rigid fixture? Maybe; depends on how the physical system is set up. Has your model has converged? This is a BIG one that is discussed very rarely, but is arguable one of (if not the) most important points.

                             

                            There are two old adages I like to keep in mind when doing FEA: "Garbage In = Garbage Out" and "FEA makes a good engineer great, and a bad engineer dangerous."

                            • Re: How to read simulation results??
                              Brent Trotter

                              Hi Guys,

                               

                              Thanks for the info.

                               

                              I can see where everyone is coming from on their posts.

                               

                              I do love quick answers but if it take me another 24 hours to fully understand the topic I would say it is worth the wait.

                               

                              As a novice SW user it is good to have the ability to rely on the expereince of kind strangers and the knowledge they are willing to share.

                               

                              As for the similation and the part a 2.3mm deflection will work fine for the part.

                               

                              Thanks everyone!

                                • Re: How to read simulation results??
                                  Jerry Steiger

                                  Brent,

                                   

                                  It looks like you are assuming that the ends of your part are fixed. Does that really make sense? In real life it's hard to find truly rigid connections. Any flexibility in the mounting at the ends will have a big impact on the deflection. It's the assumptions you put into your analysis that usually cause the answers to be less than useful or even dangerous.

                                   

                                  Jerry S.