7 Replies Latest reply on Jul 3, 2018 5:37 PM by Ryan Dark

    FEA Iterative Solver Stopped

    Afshin Zahraee

      Hi,

       

      I am a graduate student who hasn't worked too much with SolidWorks and am attempting an assembly to perform a FEA. When I used SAP2000, my advising professor wasn't too happy with the graphics despite some semblance of results.

       

      With SolidWorks, I have created a nice assembly and fixed all the meshing errors but have a failure in fixtures as the study failed to run. Eventually I need to put more complicated loading on, but I put simple loading just to make sure my geometry was ok. Unforutnately, it is not.

       

      I have been stuck on this problem now for a week and figured it would be a good idea to ask the community for help. It would be so appreciated as I realize this could be a mess of a file but I have no one here to help me for a project I have to take on in the next month or so on my own. I have attached all files.

        • Re: FEA Iterative Solver Stopped
          Ryan Dark

          Hi Afshin,

          Your model was a little more difficult than unstable studies tend to be.  In the end, in order to find a stable setup I had to start with the base, and start including attaching pieces one at a time.  Each time I added a piece I made a specific Bonded Contact Set to attach the pieces onto the last one then re-run the study to test it to see if it would solve.  Eventually I was able to get the entire model built up to include all bodies but I did need to 1) put a Bonded Contact Set between all interacting bodies and 2) turn on 'Use soft springs to stabilize model' under the study 'Properties'.  The end result was this:

           

           

          • Re: FEA Iterative Solver Stopped
            Afshin Zahraee

            Hi Ryan,

             

            This is perfect and exactly what I was looking for results-wise. In terms of attaching the pieces in the mating process, did you just do a series of coincident and concentric (when applied) mates and do the bonded contact sets at that moment? I need to look more into contact sets, which I'll do tonight before going at this again. Then, I'll try to re-create what you did at some point this weekend and may swing by another question or two by you if I run into trouble if that's ok.

             

            You don't know how much those tips are of help to me. I've been pulling my hair out all week! Thank you so much once again!

             

            Best Regards,

             

            Afshin

              • Re: FEA Iterative Solver Stopped
                Ryan Dark

                Afshin,

                I did not add any mates as the models appeared to be positioned well enough as they were.  Also, beyond positioning, mates do not directly assist in bonding bodies together.  I did notice that your study was setup with a Global Contact No Penetration definition which I changed over to a Global Contact Bonded but the study still did not stabilize.  You will want to look at putting Bonded Contact Sets (SOLIDWORKS Help - Simulation: Contact Sets) to ensure your bodies connect together for the analysis.

                  • Re: FEA Iterative Solver Stopped
                    Afshin Zahraee

                    Hi Ryan,

                     

                    Sorry to bother with this again. I have tried to go through and do the contacts as you mentioned and continue to have errors. I tried to troubleshoot myself over the past few weeks here and there and still cannot get to the final results you got to. I know you are likely busy, but any chance you can give me a step by step on how to use contacts so that I do not continue to get interference errors. It's killing me how quick you were able to come to the analysis I'm looking for and I'm sweating it over a deadline.

                     

                    I keep getting the errors between the middle tower and top tower along with errors between the rotor and each blade. I have gotten so close a few times to solving it but after I fix a couple interference, I get stuck no two or three that will not allow the program to work and cause it to crash. Any more help on this would be appreciated.

                     

                    Thanks,

                     

                    Afshin

                      • Re: FEA Iterative Solver Stopped
                        Ryan Dark

                        Afshin,

                        The process I used was to build up a stable model one component at a time.  This meant starting with all the bodies in the analysis excluded except for the single part that was attached to the fixture at the bottom.  I then ran the study with just a gravity load just to make sure it was stable before then including the next adjacent body in the analysis.  The process is detailed starting at the end of page 5 on this document (GoEngineer - Simulation: Unstable Study Diagnostics).  This video shows a method for making contact sets in general (SolidWorks Simulation: Applying Contact Sets - YouTube); I can't vouch for the music though as I didn't make that one.

                          • Re: FEA Iterative Solver Stopped
                            Afshin Zahraee

                            Hi Ryan,

                             

                            I was able to stabilize the tower and have been running simulations that are taking forever, which makes me think something is going wrong. I still cannot get blades to stabilize into the rotor for some odd reason. It's absolutely killing me that you were able to fix this in hours and, albeit as a bit of side work until the past couple weeks, have been going at it for months.

                             

                            Would it be possible for you to attach the analysis you ran above? Maybe I can see exactly what you did especially for the top part and it'll help me in the future better understand connections or create a more simplified model seeing what happened here (as it is the most complex model I have put together). I would then need to update the loading to apply to my situation, but it would really be much appreciated to see exactly what you did with contacts on the top.

                             

                            I have missed two deadlines and am pulling my hair out over it, but your help has been very much appreciated as I would be miles behind had I not sorted the tower.

                             

                            Hope you have a wonderful 4th!


                            Best Regards,

                             

                             

                            Afshin