5 Replies Latest reply on Dec 11, 2013 8:56 PM by Alexander Lam

    Modelling post simulation - centre of rotation

    Alexander Lam

      Hi all,

       

      First time poster here. I'm currently running a simulation on the sprinting prosthetic legs, and I need to find the centre of rotation. In order to do this, I need to find the most equidistant point to the entire prosthesis which in the most part requires me to be able to model after it has been deformed from my simulation. Would you guys happen to know how to do this?

       

      Thanks

        • Re: Modelling post simulation - centre of rotation
          Bill McEachern

          Well, not that it works all that reliably but if it is a solid element model you can export the defomed shape to another configuration in the file. then you could try fitting a circle or a shpere and see at which point it looks equi distant. Or more rigorously if you have a single surface or say a split line that would work you can export the nodal coordinates and the component defomation and then do some magic in excel/MatLab/MathCAD and fit a sphereical surface to the data. I have no idea what your leg looks like but in general most structural curvature is of second order and typically parabolic for bending though that may not stand up to a complete and rigorous argument.

          • Re: Modelling post simulation - centre of rotation
            Jared Conway

            export a deformed version using sim pro or higher

            like bill said, you need solid elements

              • Re: Modelling post simulation - centre of rotation
                Alexander Lam

                Thanks, I managed to export the deformed shape into a separate part, but I've now come across the problem where I can't accurately determine the location of the centre of rotation in solidworks. What I'm now doing is using a set of lines that are fixed to points on the surface and connecting to an arbitrary point. And then letting them be equal length but clearly its impossible to get all the points to be exactly the same. So is there a way to set a tolerance to what i would like length differences to be between. I do know of another method that can acquire the centre of rotation, but that requires me to get actual running data - which I don't have yet sadly.

                 

                Thanks

                  • Re: Modelling post simulation - centre of rotation
                    Jared Conway

                    if you had a cube, how would you find the center of rotation?

                     

                    if you were going to setup an experiment to determine the center of rotation, how would you do it?

                     

                    if you were going to determine the center of rotation on paper, how would you do it?

                     

                    i don't have any specific experience with it, but maybe if i know more, I can come up with something.

                      • Re: Modelling post simulation - centre of rotation
                        Alexander Lam

                        Well, the goal of the research was to sort of validate that there was there was no fixed centre of rotation rather that it moving around. So Im guessing that the Solidworks method won't in this sense. So what I need to do now is velocity data which I could then use Finite Helical Axis to determine the centre of rotation. So I guess in the mean time, I just need to read more about joint articulation and FHA which has some math that I need to learn.