9 Replies Latest reply on May 27, 2008 2:02 PM by Larry Kutcher


      dear friends,
      i want to analysis one composite material.
      it is made up of glass and plastics.
      it is a assembly one.
      we can define the material prop directly.
      can i solve this problem using cosmos non linear analysis?
      otherwise please send me procedure or help file for compositeanalysis.

      please help me

          Vince Adams
          Tell me more about your material... If it is simply glass-filled plastic, an isotropic material is probably sufficient. Some long fiber filled materials can orient themselves so that a orthotropic material might be warranted but I usually advise against using an ortho material unless you've gone thru the exercise of determining the properties empirically.... AND... for injection molded parts, determined local part flow orientations using a MoldFlow type program. Otherwise... still best to use isotropic.
          Composite modeling methods are typically used for laminates. Is this your problem?
              Fabian Zander
              I am also very new to SolidWorks and Cosmos in particular. I am interested in analysing some composite laminates but have no idea how to start this in Cosmos. In other FEA packages I would define a ply and them stack them at various orientations in order to build up a laminate.
              I can't figure out how to go about this in Cosmos and haven't found anything on the topic at all other than this thread. Would you be able to refer me to further information or offer some starting pointers?
              Anthony Botting
              Hi Fabian: You can do that directly in the interface known as "GeoSTAR" - you may or may not have it, but it's also known as "COSMOS/M" or just "COSMOS". Currently, the ply stacking sequence is not available in the "COSMOSWorks " Windows-native interface. A workaround for the Windows interface is to use orthotropic material model. You will have to obtain some software to determine net EX, EY, EZ and NUxy, NUyz, and NUzx, and the Shear stiffnesses, then you can put that in the COSMOSWorks Windows-native interface. I have used the Orthotropic material model in the linear realm and it does behave as you might expect. I have not identified (nor have I searched for) a ply-stacking sequence EX, EY, EZ calculator. I know my professor developed one fairly easily and that leaves me to believe you might be able to find one on the internet for free or little cost. Let us know how you do! We love your comments. Tony
                  Fabian Zander
                  Hi Anthony,
                  Thanks for the information. This was the only way that I could find to do the analysis so far. I have the capability to estimate the parameters so I will continue using this method for now.
                  The only question I have with this is to do with the material input properties. The moduli and poisson's ratio's are fine but I'm wondering about the tensile, compressive and yield strengths that need to be inputted. The material property inputs only allow sigxt, sigxt and sigyld to be defined. Should these be the strengths in the principal axis? As I understand it this would be a significant simplification. I am currently looking at a pipe and using the radial co-ordinate system for the properties and am unsure as to how the strengths convert to this co-ordinate system.
                  I do still wonder about the capability of COSMOSWorks Advanced Professional(CWAP) as the data sheet states that the 'behaviour of composite materials' can be seen. Do you know what this capability is referring to? From what I can find on the Net COSMOS/M (Geostar) seems to have been incorporated into the CWAP however all the references I have found only talk about non-linear analysis but don't discuss any of the other features.
                  Thanks for taking the time to help me out with my queries!
                  Anthony Botting
                  Hi Fabian: Thanks for your response. You are on the right track. The issue is, if you have a product called "COSMOSWorks Advanced Professional" it is supposed to include the GeoSTAR interface (a.k.a. "COSMOS/M" or "COSMOS"). The developers are in the midst of transferring all of the functionality from that interface to COSMOSWorks interface, but the process is slow, and currently they really have not incorporated all of the composites functionality. The workaround at this time is to output a GeoSTAR file from COSMOSWorks and open the file into COSMOS/M GeoSTAR. In the GeoSTAR interface, you will find the controls you seek, but not without some exploring of the interface - a daunting task. So, as sales literature would have it, they are technically correct by stating they have the functionality. That being said, I am unsure of the strength parameters listed in the COSMOSWorks interface - they may simply not be present, and that is unfortunate. What to do? I assume a conservative approach could be adopted in COSMOSWorks, where you simply use the lowest strength value. In the past I have used GeoSTAR, and it is quite capable and I have witnessed a very stable interface for it. I did conduct some composite laminate analysis on tube designs in GeoSTAR, and it worked quite well - I believe it was version 1.65, using a 386 PC. That statement dates me! However we did not conduct strength analyses as we were only interested in stiffness (they were for spacecraft bus structure applications and needed to be really, really stiff). Well, I hope this information helps, if only a little. Regards, Tony.
                    Adam Latour
                    Hi Fabian,

                    In the custom materials menu you have the option to pull down a menu that lets you switch between isotropic and orthotropic linear elastic material properties. When you switch to orthotropic you get the features your looking for--a unique property for each orthogonal direction. One thing to be wary of is that you cannot change the reference plane from the 'Front Plane' (I believe there is an open SPR on this) and so you have to make sure your material properties are correct relative to the assembly reference axes. As well, if you're using that material in multiple orientations within that assembly, this static reference issue will result in incorrect results for components that are not aligned with the reference used to define the composite material. I suggest making a 'new' material for each orientation with the correct relative properties. If you're applying this material to curved geometries, you may not be able to produce a realistic analysis within the standard COSMOS interface. GEOSTAR may be necessary here, but I'm not familiar with that package as I don't have premium.

                    Hope this helps!

                      • ANALYSIS OF COMPOSITE MATERIAL
                        Fabian Zander
                        Hi Adam,
                        Thanks for your input. I have been able to use an orthotropic material and input the elastic properties for the materials. This seems to work ok. My big question with doing that is about the use of the strengths that you input. The material requires a tensile, compressive and yield strength but only gives one option for each. I was wondering whether these values should refer to the principal axis and then COSMOS evaluates the other orthogonal values using the elastic properties? My feeling is that this would not be the best approach but I can't see any other way of doing it.
                        In regards to the reference axis - I'm not sure if I simply have a newer version (I only just bought mine) but I was able to change the reference axes and also able to change it to a radial co-ordinate system. I have to admit I only found how to do this by chance and a lot of fiddling. I had to set up the material properties using the front plane. Then go out of the properties dialog, select the new reference axis and then go back into the relevant 'apply materials to all' dialog. If you do this with the new axis selected COSMOS will ask you if you want to change your material reference axis. I hope this may be of some help to you as well!
                        Thanks again for your reply, all this new information certainly helps,
                          • ANALYSIS OF COMPOSITE MATERIAL
                            Ariane Beix

                            I am also interested by the answers to all these questions, and there is still something unclear for me here, as Fabian noticed:
                            There is only one option for tensile strength, compressive and yield. Should we register the one in the principal axis? Does Cosmos evaluates the other values?

                            Thanks in advance for your help,
                        • ANALYSIS OF COMPOSITE MATERIAL
                          Larry Kutcher
                          Just linking in...
                          All ears on this thread...What comes of this may be of benefit here as well.