Getting Unstuck

Here you can see the results of yet another reverse engineering exercise. I whipped this one out in less than an hour one evening recently, and then worked on rendering it for about a half hour. It was another one of those, "hmm, how would I model this object?" moments that soon became "Now I've just gotta see how quickly I can get this." When my mom found me kneeling in the kitchen examining the OXO measuring cup so intently, she couldn't help but laugh.

It's inevitable to get stuck from time to time. And so I did. The handle for this cup was something new for me, but I wanted to give it a try. I tried to slog through the problem, and tried a couple different approaches. I learned a few things along the way, such as how to create a 3D spline by projecting two non-coplanar sketches onto one another, but still I wasn't able to find a satisfactory solution. I've gotten into the somewhat bad habit of talking to the designers of the program in harsh tones, even though they can't hear me, when I get really frustrated with their program. It's not very helpful to tell me that there was an error, and then not say a word about why there was an error. Sometimes it's over a simple issue that's easily fixed through a bit of trial and error in setting options (which I then go read about), but sometimes it's a much bigger problem. Sometimes after a bit of tweaking, I can get something out of it, which then shows me that my whole approach is flawed. That's still frustrating.





So, after a few hours, I knew it was time to move on. I found a cool product design forum a little while ago, so I looked through another tutorial, discovered an essential feature I had overlooked, and played around with it to have a bit of fun. A silly, improvisational design every once in a while helps me feel refreshed for getting back to my bigger projects.

Deeper Understanding

It's not often that one takes an item and really studies it, observing every curve, surface, and texture until they could be accurately drawn from memory. When I was young, I would memorize a drawing, basically a complex symbol, and reproduce it to impress my friends. I learned one angle, one small view of many animals, vehicles, and more. It didn't really improve my understanding of their three dimensional forms, or their nature. I improved my ability to draw what I see, and I improved my visual and kinesthetic memory. Now, I am studying three dimensional objects, not just two dimensional images. Also, I am studying them with the intention to understand how they were designed and manufactured, not just to be able to impress people with accurate representations. I want to know how things fit together and how they work, not just what they look like. Too many designers focus just on the surface and styling of objects without paying attention to the underlying idea. That has parallels in every field, I'd reckon. I've noticed that in comics, advertisements, and movies; heck, just about any media is prone to creative deficits.

In my modeling of this knife, I had to figure out how every grinding of a surface was applied, and in which order, because that was basically how I was able to construct my digital version. Instead of grinding, I just created a surface and clipped away a portion of the solid or split two regions. Here you can see all the surfaces I used to do this. One cut would create the geometry necessary to create and apply the next, and so on down the line until it was complete. I can tell that the creation of this knife was very labor intensive compared to your typical mass-produced kitchen knife. A premium knife uses better materials and has much better craftsmanship. In studying this object so intensely, I gained a greater appreciation for the work of those who created it. That's not to say this is exactly how I'd design a knife. Oh, no. I'm a much bigger fan of Leatherman tools than I am of fancy kitchen knives. I'd love to design a folding knife/multi-tool sometime. I'd get a chance to use the FEA add-in in SolidWorks to test parts to make sure they bend or spring in the right ways. That sounds like a good portfolio project.

Still Learning

I have been working on a bigger project the last couple of days, and I want to wait until I'm finished with that before I post it. It could be a decent portfolio piece when I'm done. I suppose it might be good to post it here in rough draft form to get a critique from my readers (both of them). Tonight I worked on reverse engineering a teakettle which was designed by OXO, my favorite kitchenware designer. They make products that look nice, feel nice, and work well. The full user experience is accounted for, not just the purchasing portion. Our previous teakettle looked nice, but was too heavy and had a few design flaws that made it very tough to use. The OXO kettle is a much more complicated design, with curvature along two axes, a linkage in the handle that opens the spout, and some tricky manufacturing details that help simplify thing and hold it together well. Due to thoughtful design, all that structural complexity leads to functional simplicity. I'd much rather struggle to model a great product than breeze through the design of a crappy one. I figured out how to use a few new features through this exercise. It seems that I'm getting faster with the program, and it's getting pretty fun to use. I'm also learning about how it's used within organizations and in industry. It has a lot of potential which I have yet to tap, but I'm glad to not have to use all of it's capabilities quite yet. Knowing of them is good enough for now.

Simple Solutions

Today I learned a few more new features, and worked on some complicated designs. I didn't make too much progress on them, but I did find a solution to a design problem that had been bugging me for a while. I finally figured out how to replicate a very nice looking design element on my Canon printer's front tray. The surface is inset a bit, but the transition is very smooth all around, like the shape of the miniscus of water in a test tube. I couldn't quite figure out where to start for a while, and then I tried something that didn't work quite right, and then I got it. It was a lot simpler than I had thought, and it gives me more confidence that I will be able to keep on finding the simple and elegant solutions to design problems.

Design of the Day: Split Mouse

Today I wanted to figure out how to create an arbitrary object that has separate pieces that join together to form smooth surfaces and line up perfectly. Upon observing the design of things such as computer mice and video game controllers, I hypothesized how this would be done, and set about to figure it out. I found that after a solid shape was created, it could be split by any surface into separate solids that could then be saved as different parts, hidden, deleted, or just assigned separate features. Since I now knew many ways to form surfaces, this opened up a lot of possibilities.
I constructed a computer mouse that was based somewhat on my Wacom mouse, and on my old MS mouse that broke. As I went along, I figured out how many of the features on my Wacom mouse must have been created. The next step would be to figure out how to model the interior parts so that it can hold together and hold all the electronics, while also being efficient to mold and assemble. Making a representation of a product design could be an afternoon project, but actually designing a product from start to finish is a very involved process. I want to be involved in that process again very soon, and that's why I'm working at this so hard.

Surfaces Experiment

I worked on better understanding surface features in Solidworks last night. I made extrusions, sweeps, revolves, lofts, holes, fills, and more. I then knit them together and thickened the surface into a solid. Then I could easily round the edges. I want to learn surfaces better because there are so many types of designs where they are essential. They seem to be most useful for making complex, rounded shapes, such as phones or video game controllers. I haven't mastered them yet, but I'll at least better know where they could be used in a design project. I had fun with the rendering, and added different colored spot lights to give it a cool look.

Canoscan 8600F Reverse Engineering

Yesterday evening I worked on modeling my scanner's lid. It was a lot more difficult than I initially predicted, although if I had to do it a second time, it wouldn't take me all that long. I think I spent about as much time pondering as I did modeling. I had a few "Aha!" moments, and found simple ways to create features that I had been going about all wrong at first. There are a lot of small, subtle design touches in this product which give it a striking look overall. Does the front face need to be slightly curved, angled, and beveled for the scanner to be functional? Heck no. But does it look really cool, and probably cost about the same to manufacture? Yeah. It just costs more to pay some industrial designer to figure out how to add those details. There is a lot more to the lid than what I was able to model, yet although I'm tempted to, I'm not about to dismantle my new scanner to figure out what it looks like on the inside. I wait until they break to do that. I still have the stepper motors from my old HP scanner that conked out a couple years ago. I'll make that egg-plotter yet.

Design of the Day (Times Nine)

Here is a collection of my "Designs of the Day" thus far. Only two were from last year, although I rerendered one of them. Each design involved an exploration of different elements of Solidworks, such as lofts, shells, patterns, flexing, assemblies, sweeps, and of course, lots of fillets (rounded edges). I included larger versions of the images in a small gallery.

I'm starting to look at common household objects differently. I either think, "Oh, I could model that using such and such features", or I think, "Dang, I wonder how I'd model that?" My goal is to have more situations where I can figure it out right away, so I want to practice more.

Design of the Day

I am working to become an industrial designer, so I want to learn to use many design tools, and greatly improve my design abilities this year. I plan to make at least one design every day; either abstract, illustrative, or reverse engineered (copied). SolidWorks is quite useful, so I'm studying that at the moment. Over the last few days, I have made several models to explore the various features in the program. Here, I chose an object in my room to model, so that I could put my knowledge of all the new techniques to the test. I had a fun time figuring out the right combination of features to use to make it work. I then spent a long time working on the settings, scene, and materials to render it properly. It's a pretty close match. This is a cutaway view of my Bose Companion speaker. I'm just guessing at what it might be like inside, as I didn't actually dismantle it. I'll post the abstract experiments I made so far in the next post.