12 Replies Latest reply on Mar 25, 2011 5:19 PM by RICH PAUZA

    Any idea how to approach this?

    Brian McVeigh

      Hi guys, I'm pulling my hair out here trying to model this!  I have attached images of what I am trying to achieve, but not really sure how to go about it.  I tried using boundary surfaces as shown, but I really need to be able to vary the tangent length so they're not working out very well.  Any suggestions would be greatly appreciated!

       

      Cheers!

        • Re: Any idea how to approach this?
          Philip Blair

          I would split it into the top and bottom surface.  I assume since you tried the boundry surface with the shape on the cone that you already have a cut out in that cone.  I'd draw to "guide curves" or boundry lines at the front and back of the airfoil.  Then you could create two seperate boundry surfaces for the top and the bottom.  You could also ignore the tangent to face option and put that in with a second operation as a fillet.

            • Re: Any idea how to approach this?
              Brian McVeigh

              Hi guys!

              @Charles, The first images are from a 3D modeler called Modo, they are just a rough subdivision surface polygon model I mocked up to give myself some reference images.  Yeah, I mean the strength of the tangent handle.  Could you please explain a little bit what you mean by "Then trim each sketch-on-edge down to the proper size you need"?

               

              @ Phillip, if I didn't use the tangent to face option, would the curvature not only match near the guide curves I had drawn?

               

              Thanks for the help guys, will give this a try first thing tomorrow!

            • Re: Any idea how to approach this?
              Charles Culp

              Brian,

               

              What are those first images from? Is that your model, or is that just a reference? When you say the tangent length, I assume you are talking about the strength of the tangent handle?

               

              I suggest splitting this into 4 seperate surfaces. Do the top, then the bottom, then the front and back edges. You will have to create a 3D sketch, and use convert entities on the open edges. Then trim each sketch-on-edge down to the proper size you need. This will make controlling the flow much easier.

               

              *Edit: Or as Philip mentioned, even just splitting it into 2 might help. It depends what reference sketches you have available.

              • Re: Any idea how to approach this?
                Jerry Steiger

                Brian,

                 

                It looks like your curve on the body needs to be much fatter, at least on the upper side, to get the kind of flow that you have on your preliminary model. It looks like the upper surface is being force under the lower surface.

                 

                Jerry Steiger

                • Re: Any idea how to approach this?
                  Brian McVeigh

                  Thanks for the help guys!  I think I have managed to get it quite close!  I'm pretty new to this kind of surfacing and really appreciate all the input!

                  Struggling with 2 new areas now!  Does anyone know of a way I could improve the flow (?) of the surfaces? - zebra screen attached.

                   

                  I am also trying to cap off the wing tips, much like a spitfire.  I have used a boundary surface between the top and bottom halves of the airfoil, with a projected curve in between them, which is giving me the shape I want coupled with 'tangency to face' however when trying to knit the surfaces together, the gaps are pretty big!  Anyone know how I could improve upon this?

                   

                  Thanks again folks!

                    • Re: Any idea how to approach this?
                      Charles Culp

                      Brian,

                       

                      I have attached a model of my method.

                       

                      WingtipsAndCurves.png

                       

                      The center fuselage is a revolved surface. Then the front fuselage is a boundary surface from the existing revolve to a sketched circle. I used a "Curvature To Face" to keep it nice and smooth. Then, to cap off the nose, I used a surface fill with Curvature for the edge. I added a constraint curve, which only consists of a sketched point (and a reference line), this way I can control the height of the fill. I then trimmed off the back edge, and also replaced it with a fill. I knitted it all together, and since the model is symmetric, I trimmed it in half.

                       

                      Next I created the wings with a boundary surface. Notice that the top and bottom are seperate sketches. Generally, open splines work better than closed splines, so this is good. Note that for the second (bottom) surface I used a tangent constraint on each of the secondary direction edges. This ensures good tangency between the new and old surfaces.

                       

                      Now for the good stuff, and the answers to your questions. I trimmed away the fuselage using a single closed spline. note that at the front, it is not symmetrical (top to bottom). This is an artifact of using a closed spline, and why I generally avoid it. They are just harder to control. In this case, to match your question, I need to keep it as a single spline. So ignore the error and lets take a look at what I did.

                       

                      I did it in two parts, top and bottom. For the top, I needed to only select half of the fuselage edge. But as it is a whole, single edge, I cannot select that. It wants to select the whole edge, which isn't what I want. So I made a 3D sketch, used convert entities to make the edge a sketch object, and then trimmed the sketch line back to just the top half. Now I can use that for the boundary surface. I also added sketched transitions from the wing to the fuselage (I discuss how to do that more below in bold). I didn't just want sketches, though, I want to make sure that the entire leading and trailing edges are perfectly vertical along those lines. So I extruded those sketches with the surface extrude tool. Then, when I made the boundary surface, I can select those edges, and make the surface tangent. So I created the boundary surface, made the Dir1 curves Curvature to Face with 100% tangency influence, and made the Dir2 curves with just Tangency to Face and 0% influence. Then, I added connectors (right click on one of the green lines, and select "add connector") to help control the flow.

                       

                      WingtipsAndCurves1.png

                       

                      Then, to make the bottom transition surface, I just hid the hold reference surfaces, and used the new edge of the top half boundary surface for that Dir2 curves. I knit the existing surfaces (that way I don't have to do the 3D sketch thing again). Then just repeat the same method as the top.

                       

                      To make the wing tips I just made a transitionary sketch on the Front plane. I used a 2 point spline (just the endpoints). This makes for a very sharp transition, but you may want to make it come out more. To do that just make it a three point spline, and move that centerpoint out as far as it takes. Instructions for face-to-face transitions: To make the sketch also maintain good curvature, you need to select: the spline, the edge, and the face. This will bring up the constraint Equal Curvature, which is what you want. Do this for both the top and the bottom faces.

                       

                      WingtipsAndCurves2.png

                      Now, use this to make a boundary surface from the top to the bottom. This is a bit tricky. I actually created a video, so let me just show that here (it will take about 30 minutes from when I post it for vimeo to process it, so please be patient).

                       

                       

                      So. Does that give you the results that you want?

                      • Re: Any idea how to approach this?
                        Jerry Steiger

                        Brian,

                         

                        Looking at your zebra stripes, I would reinforce what Charles said about curvature continuous transitions. It looks to me like you didn't even set the transitions up with tangency, let alone curvature continuity.

                         

                        Jerry Steiger