7 Replies Latest reply on Jan 27, 2015 8:36 AM by Bernie Daraz

    Mate errors - seem random, can't get heims to mate with assembly

    Douglas Kimber

      I'm trying to model a late model race car. I built heims, control arms, strut rods, spindles, wheels and tires, a frame, and the rear end. Ultimately I want to use SW Motion to analyze the model. I was able to assemble the front suspension (with difficulty), but I cannot seem to get the rear end assembly into the main assembly. I added the rear end assembly and the trailing arms and third link, but it seems I cannot mate more than one heim to the rear end assembly. I made sure all the links and heims were flexible. Sometimes SW lets me add a mate without error, but does not move the part to the mate location. Other times mates seem to disappear. I had a partial assembly that was error free until I closed it and re-opened it - then it was full of mate errors. Right now I'm looking at an empty dialog box that says "View Mate Errors" (meaning there aren't any?) I got it by clicking where SW told me to click to see the mate errors, of which there are many in the tree. The entire rear end assembly has disappeared, except for a handful of bolts from the toolbox that started floating around the assembly while I was trying to perform some of the first mates. I don't know why the bolts are floating around - they were mated in the sub-assembly. I also have had parts that were locked to each other with a "lock" mate re-align themselves - despite the lock mate still being active. Ctrl-Q doesn't seem to help.

       

      I'm running SW 2015 x64 SP1.1 on Windows 7 with an Intel Core i7 CPU and an NVIDIA Quadro K2000 video card.

       

      So, my questions are:

       

      1) Is there a preferred mate for something like a heim? I tried concentric on the spherical surfaces first, then switched to coincident on the two pivot points. Both work on the heim itself and within the trailing arm sub-assembly.

      2) How should I mate the heims to the assembly? I was using a slot mate (because the frame is slotted) and a distance mate from the slot face to the heim ball face.

      3) Should I make all mates in one main assembly, or is it ok to build sub-assemblies and include them with the flexible option?

      4) Are there other tips that would help me? Sometimes I cannot get the parts to move - they seem locked, and other times they jump - they won't move just a little.

       

      Thanks in advance!

       

      Doug

        • Re: Mate errors - seem random, can't get heims to mate with assembly
          Alan Thomason

          Hi Doug...

           

          1. Either should work.  Gut feeling is that coincident would be easier to solve.  I use a lot of coincident mates between sketch points and they seem fine.

           

          2. Slot and path mates cannot be used dynamically, but that does not seem like what you want to do.  You need to fix the part along the path.  I think what you would want to do is fix the ball along along the path and then mate the heim body to that either concentrically or with coincident points.

           

          3.  For Motion (I don't know for Animation) using dynamic mates in the subassembly does not work.  You can use subassemblies if the parts are all fixed in the sub assembly.  I realize this seems frustrating at first, but I have not found a way around it.  When working with more than one person on a project I find myself recreating the entire assembly because the methodology is different, and often results in an unfortunate lack of restraint in the main assembly that is totally unacceptable to the other users. 

           

          4.  This sounds like me minus a year of cursing over Motion!  I would guess there are many Motion users who feel the same.  I find it extraordinarily difficult to use, but over much time and with help from Fisher Unitech, DASI soulutions, and this forum I have finally got some pretty useful results, so keep hope.  If you can post a pack-n-go of your project, there are forum members who will provide a surprising amount of help.  If you can't post your project exactly for confidentiality, it might be worth creating a simple non-confidential version.

           

          Good luck,

           

          Alan

            • Re: Mate errors - seem random, can't get heims to mate with assembly
              Douglas Kimber

              Alan-

               

              Thanks for the reply!

               

              If I understand correctly, you would mate the heim balls to the final assembly, then mate the body to the ball (using coincident spherical centers) - rather than trying to bring in a heim sub-assembly and then make mates in the final assembly.

               

              You are correct about the slot mates - I was using them to assemble, but planned to lock them down before running motion.

               

              When you say dynamic mates within a sub-assembly will not work in Motion I assume you mean everything within the sub-assembly must be constrained in all 6 degrees of freedom. This seems like a very big restriction - it would make it very difficult to use motion with anything designed by several teams of designers (e.g., a car engine and chassis would typically have different designers, but this restriction means the car would have to be assembled in one large assembly rather than combining the engine and chassis sub-assemblies). Fortunately (sort of...) I don't have this issue - its just me on this project.

               

              Thanks for the encouragement - I will keep plugging along. I figured it must be hard based on how hard it is to find info on the Internet. I did solve a couple issues: 1) closing and re-opening the assembly brought back the parts that had disappeared, and 2) a Ctrl-Q after every time I made a new mate seemed to force the parts to move to the new location (not sure why they would sometimes move without the Ctrl-Q and not others).

               

              I'll definitely use a pack-and-go when I have a more specific issue, but you are correct about the confidentiality - this is for a racing team, and info is carefully guarded!

               

              Doug

            • Re: Mate errors - seem random, can't get heims to mate with assembly
              Jeff Mirisola

              Without seeing your assembly, it's difficult to fully diagnose the issues, but I'll try.

               

              My general rule of thumb is to create the assembly as it would be created in real life. This would mean not using planes, sketch points or arbitrary distance mates (though this is more of a strong guideline versus rule). This is especially true when creating a mechanical assembly, as is the case here. Use logical sub-assemblies as applicable. Whenever I run into an issue of being able to mate one part successfully, but not its mirrored twin, it often has to do with something being off by a few thousands, especially if the parts are running at angles. In the case of the heims, do you have them modeled life-like, where the ends can move? If so, are they inserted as flexible?

              Something else to think about, while in a perfect world you'd have concentricity between the mating holes of the various suspension parts, anyone who's ever been underneath a car knows that doesn't happen. It might behoove you to use fasteners in your assemblies, creating tangential mates in areas where logical.

                • Re: Mate errors - seem random, can't get heims to mate with assembly
                  Douglas Kimber

                  Jeff-

                   

                  Thanks for the reply!  As for showing the full assembly, it is for a race car team, and info is held close...

                   

                  I think I generally have been creating the assembly as it would be in real life. For example, I built the heim first, mating the ball to the socket, then assembled two heims into a hex tube so the pivots were at the right distance (using a distance mate - not sure how to avoid this one) (this is the way it is done in the shop), then inserting the assembly into the chassis assembly (setting flexible for both heims). Is this what you mean? I attached the right rear trailing arm file as an example.

                   

                  I don't think I follow what you mean about using fasteners to avoid concentricity problems. I agree a real car won't have all holes lined up perfectly (another reason for hammers), but I think they should be in the SW models.

                   

                  Doug

                • Re: Mate errors - seem random, can't get heims to mate with assembly
                  Bernie Daraz

                  Personally I would skip the extra mates, for the arm and heim joints I would have a length between the two and point mates. A point mate will allow more movement in this regard, the heims float in all of the axis anyway so don't limit any motion by having extra mates. I would also mate to a theoretical center point mate on the mounting bolt that would be going through the heim. Adding a concentric mate there is worthless. I agree with Jeff on using sub-assemblies and the flexible option where necessary. Though in using point mates you may not need the flexible option for some of the parts, all depending on the design. Back in the mid 80's I invented the Vari-Torque Bar for modifieds so I have that racing blood too. I wish I had SolidWorks back then! Good luck!

                  • Re: Mate errors - seem random, can't get heims to mate with assembly
                    Douglas Kimber

                    Thanks to everyone for the help. I ended up using Bernie's suggestion of fixing the length of the arms and making coincident mates for the pivot points. Because of that I selected his answer as the correct one, despite all three answers being helpful. Bernie's suggestion was straight-forward and quick, and allowed me to get the rear end placed into the chassis correctly. I ran into a couple other mate issues when I tried to move the rear end around, but they seem to be separate issues and I'll start a new thread for them.

                     

                    Bernie - very cool that you invented the Vari-torque Bar! We've used Hoffman tube third links in the past, which effectively do 1/2 of what the Vari-torque bar does, and we use a "snubber" in the RR trailing arm to allow the RR wheel to move forward under acceleration to tighten the car on corner exit without making it tight on corner entry.

                     

                    Doug

                      • Re: Mate errors - seem random, can't get heims to mate with assembly
                        Bernie Daraz

                        Thanks Doug, I'm glad there was enough help to be found here. Seems my Vari-Torque had a short life compared to the other types of bars. Being involved before allowed me to see the issues you describe and originally design it. I was on the infield of Stafford Springs watching our car practice when the idea hit. The design of the cars had changed so much over the years, we had torsion bars in the front back then and no power steering! Our first car still races in NJ. I remember building the first car in a garage, we had to carry the roll cage tubing outside to spin it around while bending it!