12 Replies Latest reply on Feb 13, 2015 8:48 PM by Harold Brunt

    The lukewarm reception to Conics

    Kevin De Smet

      So, I just came back from the local launch event for SolidWorks 2013, and I was the only person who seemed genuinely excited about the introduction of the true Conic sketch entity. Rumor has it "big shot" Austin O'Malley kept blocking Conics year and after year... is this why? If so I can see why. People were more.. confused rather than anything else!

       

      However I just wish to say that to those of us who really care about Conics and its uses, it has not gone in vain and it has not been received lukewarm. I just hope the interest level is substantial enough to warrent further development of Conics. Three things come to mind: g2 end conditions, conics with boundary surface and conics in fillet.

        • Re: The lukewarm reception to Conics
          Roland Schwarz

          Conics are a must-have if SW wants to be an engineering tool, not just a design toy.  OK, maybe the fat-middle-part-of-the-bell-curve-dwelling designers don't get conics or their importance, but SW has spent way too much effort in making things easier for people who know or do too little.

          • Re: The lukewarm reception to Conics
            Anna Wood

            Kevin,

             

            We need to remember that a very big majority of SolidWorks users create prismatic models.  Blocks with holes.  The need for conics is not that great when everything starts from a block. 

             

            For those doing industrial design, surfacing and optics, conics are the bees knees.  We are happy they have been finally added to SolidWorks.

             

            I am sure SolidWorks is very aware of what part of their customer base will appreciate conics.

             

            Cheers,

             

            Anna

            • Re: The lukewarm reception to Conics
              Matt Lombard

              I appreciate the addition, but how often would you use a conic just by itself? or only have a tangency requirement at the ends? Wouldn't you want that connection to be curvature continuous?

               

              partial ellipses and parabolas also don't allow c2, but I'd never use those inbetween two splines.

               

              I'll probably continue to use workarounds for this. I was hoping conics would allow me to have a sketch element that wouldn't flip convexity through some transition, but I also need it to be smooth in the other direction. 2 or 3 point splines will continue to stand in for conics until they have full functionality.

                • Re: The lukewarm reception to Conics
                  Jerry Steiger

                  As Matt said,"I was hoping conics would allow me to have a sketch element that wouldn't flip convexity through some transition, but I also need it to be smooth in the other direction". I'm hoping Kevin is right and we just have a year or possibly even less to get G2 on the ends. I'm still thinking that 2013 is going to be good enough (after a few SPs) to get us to update from 2010.

                   

                  Jerry S.

                • Re: The lukewarm reception to Conics

                  Hi Kevin,

                   

                  Hope you had the chance to read my October blog post on this subject. Anna's correct in that the majority of our market is still machined design and "we" as consumer product modelers need to keep that in mind - I'm certainly aware of this - more than you know being a product manager inside SW

                   

                  I don' t normally do this but I feel as thou I must put an end to the rumor right now... Austin, to my knowledge (and I think I'm pretty much in the know about this), had nothing to do with "blocking" conics getting into SW earlier, my blog post sheds some light on the history of conics etc. So I hope we don't have to hear about that again.

                   

                  Regards

                   

                  Mark

                    • Re: The lukewarm reception to Conics
                      Harold Brunt

                      Mark - I assume you know how much I have been anticipating the addition of conics to the SW sketch tools. I am doing the wave right now (and my associates are again wondering about me). I am not able to download new releases until my add-in is updated so I have not been able to use SW2013 as yet. Haven't made a demo either as I have been traveling a lot. A couple of questions I have regarding the new interface:

                      • In addition to changing the rho value, is there a means of changing the radius of curvature value?
                      • Is it possible to generate the curvature using the interface prior to sketching the approximate curve?
                      • What is the accuracy of the surface generated by the sketch (comparable to a spline, arc, or...)?

                       

                      Thanks again for being part of the solution for getting conics into SW.

                       

                      Harold

                        • Re: The lukewarm reception to Conics

                          Hi Harold,

                           

                          Yes, you were instrumental in helping us - in fact your name is listed in the original product spec! As to your more specific questions: A) no, there is no means of changing the radius of curvature value in this first release - do you mean as an alternative to Rho? B) Not quite understanding your question, but yes, you an do a "show curvature" on the conic, but you'll have to expand on "using the interface prior to sketching the approxiamate curve?" C) I believe it is 1.0e-5(m). I'll have Rob Jost chime in here (although I believe he is out of the office part of this week visiting customers.)

                           

                          Regards

                           

                          Mark

                            • Re: The lukewarm reception to Conics
                              Harold Brunt

                              "...listed in the original product spec!" I'm printing that email out and posting it on the lab door. Good luck keeping my ego in ckeck today!

                               

                              A) Changing the radius of curvature is in addition to changing the Rho value. Typically I use the formula driven sketch with this form:

                              y = (x2/R) / (1 +[1 - (1+K) (x/R)2]),

                              Where: The axis lies along the y-axis and represents the height as a function of x.

                              The vertex is at y = 0.

                              K = Conic constant Rho

                              R = Radius of curvature.

                               

                              If there is no input for R the how is the radius established?

                               

                              B) Not sure I can quite explain what I am refering to breifly but this is the least of my concerns so lets just let this one just sit.

                               

                              C) That would be the same as the standard bspline in SW, correct? I am looking forward to putting one in an optical system and comparing it to the Add-In.

                                • Re: The lukewarm reception to Conics
                                  Eric Herman

                                  Harold - Have you figured out a way to define the conic constant using the typical equation (y = (x^2/R) / (1 + sqrt[1 - (1 + k) * (x/R)^2])) in the form that is necessary for the conic tool? I have spent a fair bit of time trying to translate the terms necessary into optical references.

                                    • Re: The lukewarm reception to Conics
                                      Harold Brunt

                                      hello Eric.

                                      Nope but then I haven't needed to try much either. The last time I messed around with the conic sketch tool in SW I found the interface a bit lacking. The tool was geared more for industrial design than optics to the point that the radius of curvature was not included as an input. My understanding was that the formula also deviates from the standard / common conic section curve in that the Rho value is always less than or equal to 0 (zero) making translation from optics software prescriptions (like Zemax) a bit more difficult.

                                       

                                      I'll take another look when I start using 2015 but until then I have the curve driven sketch tool. If you generate lens surfaces frequently for imaging systems you might want to consider an add-in. The interface for OptisWorks for optical design and specifically the lens tool has improved a lot over the several years I've been using it. Feel free to send me an email if there are other questions you have about modeling optics in SW. I'll do my best to get back to you asap.

                                       

                                      Harold

                            • This reply has been hidden. This can happen if the message has been hidden by a moderator, or has been reported as abusive.
                              • Re: The lukewarm reception to Conics
                                Kevin Quigley

                                Well I am very happy with conics in SW. No it does not have G2 yet, probably because SolidWorks license the D-Cubed 2D constraint manager from Siemens. The ability for the DCM to handle any tho controlled conic was only added in 2011. The ability to apply G2 end conditions was only added in April 2012....which is very late in the release cycle for SW2013.

                                 

                                I suspect the reason conics were never added was more to do with keeping the 2D sketch system entirely in the D-Cubed system than anything. If conics were not in it then it cannot be added can it? Other apps we use that utilise the same engine have conics but no constraint control for those elements.

                                 

                                I am quite sure SW will incorporate all the features of the DCM as it releases service packs or new versions.

                                 

                                The reality is this. SolidWorks now has a conics tool. Which is more than other competitive products have.

                                  • Re: The lukewarm reception to Conics
                                    Kevin De Smet

                                    Kevin,

                                     

                                    I think you nailed it on the head. I suppose the rumors about Austin sounded a bit to much like rumors anyway, and I for one apologize for bringing it up in this thread. It makes sense that you wouldn't want a separate constraint solver just to support Conics.

                                     

                                    Mark,

                                     

                                    Great blog post! it clears up some of the history, and how to use them in a boundary surface.