It's been awhile since I've done any real personal design work, especially CAD work. I've never been that great at using CAD and have looked into "optimal" practices when drawing something up in the computer on and off over the years. Some examples of questions I've had are like what order should different bodies be created; when should we fillet on the sketch first or after the extrusion on the edge, how to create effective "dynamic" designs that change with parameters, the best way of creating either movable or rigid joints between objects, good flows for large multi-part assemblies, etc. There are always more questions the more I see other designs online and the more I personally design.
I've definitely picked up tips along the way like using boolean intersection and cuts more often (e.g. for making tabs), using a dog-bone plugin for modifying sharp internal corners for CNCing, and other more general workflow optimizations. But in the end, just CAD'ing more is the best practice for being faster and identifying poor workflow; if possible, manufacturing the design after CAD is even better as you quickly learn what is and isn't possible to make with your tooling (my time with my crappy CNC has taught me a ton).
But back to just CAD'ing more; I had some free time recently and decided I should try to CAD my glasses for some practice. Using only a simple ruler for some basic measurements (so I was designing near the right order of magnitude) and the judicious freedom of my eyeballs, I whipped up this model:
Perspective top view. |
I modeled half of the glasses and then mirrored everything at the end for the other half. I started off with the lenses. I created them by creating a spline for the front facing outline of the lens, then created a series of arcs and lines for the top down outline of the lens. I extruded both profiles, and used a boolean intersection to get the final lens. For the groove around the edge of the lens that holds the wire holding the lens to the frame, I sketched a groove into one side of the lens, then lofted the profile as a cut all the way around.
For the frame, I used a very similar procedure to the lens: I created both front and top down sketches of the profile of the frame near the lens, extruded one and cut with the other. The hinge fingers I cut in with another sketch on the outer face. The bridge in the middle was extruded very roughly then cut out with refining curves on separate sketches, and then fillets were added to smooth out joining edges.
Next up are the annoying legs. I had the most trouble with the legs of the glasses since they had the most complex shape and had to form a joint with the rest of the frame. Given that it is mostly rectangular and follows a fairly smooth 3D path, I used a loft. I first made a top down sketch with a spline that would be the top down curve shape, then made a side view curve sketch. Then, using the top down curve as a reference, I moved the control points of the side view curve out of the original plane of the sketch to make a 3D spline. Finally, I added some work planes along the path to draw the profiles that I wanted the legs to sweep through. It's still not perfect though; I was a little lazy and didn't refine the hinge as much as I should have. I had another sketch to cut away near the hinge the excessive material from the loft and make it transition smoothly, but I neglected to cut the extra area needed for the hinge to actually not self-interfere when closed all the way. The underside of the hinge isn't perfectly clean either:
The leg isn't cleanly cut from the frame. |
The last bit of modeling was the nose supports which was straight forward; I used a pipe along a spline to make the wires and use standard 2D extrudes for the rest of the geometry.
Comparing the model to my actual glasses, the front is not curved enough but I think it's good enough for now.
Here's a video going along the timeline for making the model:
And a gif of the glasses closing:
And a nice render:
Along with fixing up the joints, some other fixes would be to add some more "texture" (there are some grippy indentations in a grid-like pattern) to the inside of the legs and adjust the middle of the frame some more to better match the thinner parts. Good practice overall and maybe I'll do some more in the future.
No comments:
Post a Comment