31 Replies Latest reply on Sep 8, 2015 11:17 AM by Randy Miller

    Yield stress changed on its own.

    Chris Pellegrino

      When I specify a material in the Simulation tree, it makes a new folder for this component/assembly and lists the material there even if it's an already existing material.

       

      1) Is this so I can make a tweak to it and not change the original material?

       

      2) This is my real issue: It changed the ultimate and yield stress in that material data. I was thinking that maybe it converted it to MPa, but the values don't work out. It only changed ultimate and yield, not the modulus or anything else I entered. (This was a custom material.) Even after reselecting the material and rerunning, it still lists the yield incorrectly on the stress plot. It does this even if I go back to the component and select the material there and rerun it (this is a single component analysis.)

       

      3) I can't find where to change the target stress that shows up on the legend in the stress plot in 2015. Is this no longer available?

       

      Anyone know what's up? This was a simple static analysis.

       

      SolidWorks Premium SP1.1

      Windows 7 Pro SP1

        • Re: Yield stress changed on its own.
          Shaodun Lin

          1. If you want to make a tweak, you just make a copy and paste into your custom material library"

          2015-03-25_15-10-18.png

          3. You still able to change the target stress in the graphic window (SW2015SP2):

          2015-03-25_15-21-50.png

            • Re: Yield stress changed on its own.
              Chris Pellegrino

              Thanks - I know about making copies and creating my own materials - I have done that a lot.

               

              My real question is why did SolidWorks change the ultimate and yield stresses without my input? Again, in this analysis, I had selected the material for the component in the Simulation tree, not using the material as called out in the component. In cases like this, SolidWorks makes a material folder for this analysis that shows up in the material database under a Custom Materials folder. It then makes a copy of the material I selected for this simulation in this folder. It also changed the ultimate and yield stresses. Did it also change them or save them incorrectly in the original material definition? I don't know because right after creating the material, I selected it in the Simulation and ran the Simulation. Then I corrected the stresses in the material that was listed in the unique folder for this Simulation. Then I went back to the original material definition and that one was correct. I don't know if they are linked or not...

               

              My other problem is that I cannot find where to change the yield stress as reported in the stress plot. I know how to change the max and min stress on the plot, but I remember being able to also manually change the yield stress as it is reported on that plot, but cannot find it now. Maybe I never could and I am thinking of something else.

               

              Again, I tried going back and rerunning the analysis with the corrected ultimate and yield but it does not change the reported yield stress in the plot (still incorrect.)

               

              I also have no controls to make sure I know that SolidWorks is using the material as defined in the component as opposed to defined in the Simulation. You can define material in the Simulation tree, but you can't "remove" that definition so it reverts back to the material defined in the component. At least there is no action I can take so that I know what it is doing.

                • Re: Yield stress changed on its own.
                  Jared Conway

                  yield strength comes from the material, so there is nothing in the plot that you change, it comes from the material

                   

                  can you post the exact workflow/sequence and an example model?

                   

                  the only time a "new folder" should be created in custom materials if if you bring in a model with a material with the same name but different properties

                    • Re: Yield stress changed on its own.
                      Chris Pellegrino

                      Yes, it comes from the material, and it's very curious how it's not reporting the yield stress as shown in the material... It did match that first time when it created the folder for this analysis and put another copy of the material there with the wrong yield value. It changed the ultimate and yield for some reason and from that point on is reporting the incorrect yield.

                       

                      Screenshot showing the analysis folder with the material (with corrected ultimate and yield values.)

                      Original material data:

                      Workflow was as described in discussion above.

                       

                      1- made this material (second screenshot)

                      2- selected this material for the component in the analysis I am doing on the "7100-0102 mount FEA" assembly (right now only 3 components in the assembly but this configuration is only one of the components - the other two are suppresssed) I selected this material through the RMB menu on the component in the Simulation tree

                      3- ran analysis, saw Yield Strength 1.400e+.003 in the stress plot

                      4- right clicked on component in the Simulation tree and brought up the material - it brought up the material in the new folder for this analysis as seen in the first screenshot, but the ultimate and yield stresses were wrong. They were different from each other, but I don't recall what the ultimate was.

                      5- I corrected the ultimate and yield stresses in the material in this 7100-0102 mount FEA folder

                      6- Went back to the original material (as seen in the second screenshot) and the ultimate and yield are correct.

                      7- Tried replotting but still shows 1400 yield in the stress plot.

                      8- Reran analysis - still shows 1400 yield in the stress plot. Checked material in both locations and yield is correct.

                      9- Reselected material to use the original (from second screenshot - as I originally defined the material) - reran - still shows 1400 in stress plot.

                      10- Rebooted and reran - still shows 1400 in stress plot - still showing up correct in both material locations

                       

                      I don't know what else to tell you. The 1400 seems to be coming from somewhere else. I didn't do an exhaustive search, but I did search around and couldn't find that material anywhere else in the database.

                       

                      Another thing - if I select a material manually for a component in the Simulation tree, I would like to have the means to "remove" the material so that I know it is coming from the material as defined in the component itself. Is there a straightforward way to do this? I assume right clicking on the component in the Simulation tree will lead to me where in the database it is getting it from, and if it is not the folder created for this analysis, then it is probably getting it from the material as defined in the component. Is that correct?

                • Re: Yield stress changed on its own.
                  Randy Miller

                  Perhaps it is the same issue I've seen with some of our materials here.  On my end I had a material that was set correctly in the material library but after running a simulation it seemed to pull a yield strength out of the air.  I was using Aluminum 6061-T6 with yield at 35 ksi and the simulation would use 16 ksi instead.  No bueno.  This was happening for another material as well so I wanted to find out root cause before declaring victory.  Making a long story short I opened the material property in the material library and selected the "Tables & Curves" tab.  I then went to Type and selected "Yield Strength vs Temperature" from the drop down menu.  The value at about 350 F was 16 ksi.  I deleted that row, saved the material, applied it to my part and ran the analysis again.  Here is my screen shot showing that the yield strength in the stress plot shifts from the previous 16 ksi to 30.5 ksi.

                  New Picture.png

                  I next went to the properties in simulation and checked that the default temperature is 77 F.

                  New Picture (1).png

                  So, I have not idea why simulation is pulling from the temperature dependent table but to fix this issue I've deleted the material and recreated it without any temperature dependent data.  For some reason I can't just delete the temperature dependent table and I have to have at least 2 data points.

                   

                  Hopefully this helps you with your issue.  Check your temperature dependent values for ultimate and yield to see if it tracks to what simulation is using.  If anyone knows of a good way to remove temperature dependent tables completely (without having to recreate the material) please let me know.

                   

                  Good luck.

                  • Re: Yield stress changed on its own.
                    Eric Christison

                    Hi Chris,

                     

                    I must admit a wry smile crossed my face when I read your post. I've been saying for years that SW materials is not fit for purpose because you continually encounter these sorts of bugs. Whenever I do though, I am assured that it works just fine and it's just me (and the rest of the engineers that work in my office) ;-).

                     

                    My advice would be the same as the first respondent to your post. Create a new material, at least then you can keep track of what SW changes or doesn't change without your knowledge.

                      • Re: Yield stress changed on its own.
                        Chris Pellegrino

                        I am also very leery of SolidWorks analysis - it is definitely not just you. Case in point: they give you fairly good orthotropic property capabilities, but don't provide adequate failure criteria to make use of those materials. This renders the orthotropic capabilities pretty much useless for analysis of orthotropic materials.

                         

                        I have already created a new material; it is that material that SolidWorks changed the yield and ultimate in without me telling it to do so.

                         

                        Thanks.

                          • Re: Yield stress changed on its own.
                            Jared Conway

                            what failure criteria are you looking for with respect to orthotropic materials?

                             

                            but in general, i'd agree with Eric that you need to be careful about material definitions.

                              • Re: Yield stress changed on its own.
                                Chris Pellegrino

                                As I understand it, Tsai-Wu is a common failure criteria for anisotropic materials. There is another I can't remember that I found when I was doing some orthotropic analysis a couple of years back. SolidWorks, at least back in 2013, and I believe 2014, did not have adequate failure criteria for properties like this.

                                 

                                I don't know how more careful with material definitions I can be; I don't know what you mean. Give me an example of not being careful.

                                  • Re: Yield stress changed on its own.
                                    Jared Conway

                                    tsai-wu is part of the composites module in solidworks simulation premium.

                                     

                                    from a material perspective, I don't think you've posted a model that I can take a look at yet but what i'm suggesting is that a couple things that I have seen be helpful are:

                                    1. unique material names

                                    2. separating solidworks and simulation materials

                                    3. having fully defined materials

                                     

                                    that being said, these aren't always practical and the software should be better able to handle them but without an exact workflow that causes the issues, the software can't be improved or fixed.

                                      • Re: Yield stress changed on its own.
                                        Eric Christison

                                        Perhaps I misunderstand you Jared but I'd have thought that there is enough information in this post to investigate several bugs in this part of the package.

                                        • Re: Yield stress changed on its own.
                                          Chris Pellegrino

                                          My names are pretty unique; are you referring to something else?

                                           

                                          I don't know what you mean by separating SolidWorks and Simulation materials. There's a database, and I use it.

                                           

                                          The materials are as defined as I need them at the moment. For whatever reason, I don't have compressive strength in there, even though I'm sure MMPDS has it. Even so, SolidWorks shouldn't be changing the values I put in there, I'm sorry.

                                           

                                          Tsai-Wu may be available in the composites module, but it's not in the "standard" module where they have orthoptropic properties and therefore need that criteria also.

                                           

                                          Again, sorry about not being able to post the model. If I had the time to make a featureless one that had the same issue I would. I was really hoping someone else has seen this and could offer some help.