7 Replies Latest reply on Apr 5, 2017 12:13 PM by Bill McEachern

    Need help reading results

    Alex Dring

      Our engineer left our company and I been tasked on taking over his duties in modeling.  He was able to teach me how to run a few simulations such as frequency and static for vertical loads.  What he wasn't able to show me is how to determine stiffness of the assembly.  Can anyone walk me thru that?  He did mention to determine stiffness you remove the gravity and change the load to 1 newton.  I am not sure if that is only one newton per foot or one newton for the entire load.  Anyway I run it again and I am not sure what to do from there.  Shown below is URES: Resultant Displacement. (mm).

       

      As you probably already guessed I am no engineer and new to Solidworks as well.  We use inventor for our design work and he got Solidworks just for the simulation.  Any help would be greatly appreciated

       

        • Re: Need help reading results
          Jim Steinmeyer

          Alex,

          Not to demean or discourage you but this is the very reason I do not like FEA packages like Simulation being as "user friendly" as they are. It is very easy to place constraints and loads on a component and get some very pretty and convincing pictures that have no more value than the paper they are printed on but because they are run on a "sophisticated" program and have very pretty pictures and graphs people will take the results as proof that a design will work. Then the part breaks the first week of use, or worse yet the bridge collapses on a buss load of crippled lawyer's children.

               In running an analysis there are several key things to be aware of. Is your model set to have the proper material with the proper material properties? Does it have the correct thickness? Do you know how it is going to be mounted? Is it sitting on a flat surface with a force directly towards the surface? Does the mount allow it to rotate or translate in a direction? Is it a pinned or a fixed mounting, can you tell the difference? Do you have all the degrees of freedom accounted for? what about the load? Will it be centered of offset to a side? Will it be straight of at an angle, constant or cyclic? Where does the one Newton come from? Did a fig fall on it? Or is that the weight of the concrete that the block will be supporting? Why would you remove the effect of gravity? Is it standing on it's side with the desired deflection measurement to the right or is it filled with helium?

               I understand you are doing your job and attempting to please the boss but does he really understand what he is asking of you? Even as engineers most of us do not receive the proper training to really understand FEA correctly. It is a specialized subject that only a few take classes in. That said, with enough training and guidance lost of people, not just engineers, are capable of learning how to run this program and with enough experience can produce accurate results. It is clear you have not had the benefit of that guidance.

           

               Sorry for the soap box rant but your boss is placing quite a bit of responsibility on your shoulders that you have not been properly prepared for and I would hate to see you fail much like I would fail if dropped in a fourgin  country asked to spell correctly.

            • Re: Need help reading results
              Bill McEachern

              Hi Jim,

              Respectfully, I would argue that ease of use has nothing to do with who should or shouldn't be using an FEA package. I would also say that risk management is a corporate responsibility. If your the only guy in the company well you own the whole thing but if others are involved they need to have the grey beards check the stuff or execute a risk reduction plan the minimizes the risk or eliminates it. You can't learn how to ski powder unless you get in it. Same goes for FEA. I don't think that Corbets couloir with a meter of fresh should be your first ride either but one can work up to it. Alex, no disrespect intended but you need to take some serious lessons. Don't let some body push you to do something or endorse something you have no confidence in as it can lead to serious consequences for some one if things go bad as they are likely to. If they push I would go find another job.

                • Re: Need help reading results
                  Alex Dring

                  Well I have to try at least.  Is it ideal?  Of course not.  I am not going to tell my boss no or leave the company because I am asked to take on more responsibility.  Am I going to get training?  Of course I am, but I need to start learning it right now.  Its why I came to this forum, to get help.  I thought that's what forums were for, not to be told to give up or find another job.  All I asked was how to read the results to determine stiffness.

                    • Re: Need help reading results
                      Jim Steinmeyer

                      I understand your position. What I am saying is I think you have been given very incomplete guidance from the engineer that left. And your question indicates that the guidance given was 1. in error and 2. will get you in a lot of trouble. If all you needed to do to determine stiffness was to place one Newton of force (either total or per foot) don't you think SolidWorks would simply have a button indicating "Check Stiffness", you select that button and the stiffness is determined? In reality stiffness is a very relative to the situation and a number of factors. The metal staircase you walk up on the way to work is very stiff if you place 1 Newton of force on it. If you place say 50N over the side beam it may still be stiff, but if you place the 50N in the middle of the step it will not be anywhere near as stiff and will deflect. If you place 1000N anywhere on the stairs it will in all likely hood not only not be stiff but will collapse from the load.  Another example would be if you took a steel bar 1/2" thick and 5" wide and placed it between two tables. If the bar is only 8" long it will be very stiff and easily resist deflection with a load of one Newton. However make the same bar 20 feet long and place the same load in center of the bar and the bar of the same thickness and width will certainly deflect indicating less stiffness. Does that make sense?

                           There is no way we can tell you to place one Newton total load or one Newton per foot because we have no idea what the use of the item you showed us nor the amount of load that it will be seeing. However your asking this question tells me that you need to gain a much better understanding of what deflection or stiffness is and what causes it before tackling this project. I am not saying only you are incapable of doing this job, but you are with your current understanding of what defines or determines stiffness and you need more training than just having some idiot on the net saying "1 Newton per foot".

                      • Re: Need help reading results
                        Jim Steinmeyer

                        Alex Dring wrote:

                         

                        I am not going to tell my boss no or leave the company because I am asked to take on more responsibility. Am I going to get training? Of course I am, but I need to start learning it right now.

                        I admire your willingness to take on more responsibility, but asking you to do this project at the moment is akin to having someone tell you "Pull this stick to make the plane fly up and push it in to fly down, Now go fly that jet airplane".

                         

                        Maybe there is a reason the engineer is no longer with your company.

                        • Re: Need help reading results
                          Bill McEachern

                          Hi Alex,

                          sorry did not mean to be discouraging, like Jim I too applaud your efforts to make progress. However, I personally don't have the time to go back so far into the basics to bring you along. it would just be too onerous to do so over this sort of facility. Others maybe willing to provide more fundamental guidance. If can provide short snipets of help I will. You need a basic mechanics of material book at the very least - try Mechanics of Materials by Popov or something similar. You need to understand the basics of elastic theory and this book can help. Good luck.

                    • Re: Need help reading results
                      Jim Steinmeyer

                      Here is a link that should help you understand what is involved with determining the stiffness. I know this isn't a quick, get it done today answer, but if you really want to be able to do this and have accurate results this will be a good start.

                      http://www.colincaprani.com/files/notes/SAIV/4%20-%20Matrix%20Stiffness%20Method.pdf

                       

                      Here is another one. These deal with the underlying idea and what you are attempting to determine which is important to understand so you can understand how to set up the model correctly. Maybe you don't need to know how to do the math but understanding what the math is doing helps you understand what information is needed in the model.

                       

                      http://academic.csuohio.edu/duffy_s/CVE_512_7b.pdf