I’ve been playing with origami, lately. Specifically, I’ve been exploring how to simulate, model, and fold origami shapes in ways that could be automated to create useful mechanisms. The system I’ve come up with is designed to fold rip-stop nylon, which I’ve worked with a bit during my time at Makani Power and research at Super-Releaser.
After some experiments with programs designed specifically for generating origami patterns, I found I wasn’t able replicate the patterns I’d prototyped in paper. Since I wanted to start out with a paper prototype, do some bench tests, and move to CAD from there, I needed to consider other options. I also wasn’t able to convert the output into a format that would play with CAD for printing and prototyping the resulting forms. So, I fell back on my old standard: SolidWorks. If you’ve worked with me before or you’re a regular reader, you don’t get any bonus points for guessing I’d find a way to turn this into a SolidWorks project. This video was very helpful for understanding how to think about origami in a SW context.
I developed this static mixer design to streamline casting demos. Often times, a casting demo can get bogged down with portioning, mixing, and degassing, especially when you’re trying to have a group of students get hands-on time with the casting materials.
With this design, you load up degassed silicone, store the unit until needed, and then dispense mixed material out of the nozzle. If you’d like to build your own, you can find all of the source files on Thingiverse. This project was also picked up by Hack A Day.
Cheap grow lights are great – they’re energy efficient, ubiquitous, and the design pattern of the aluminum-backed PCB for dissipating heat is well chosen. However, unshielded high-output LED’s are a real pain on the eyes.
I bought a set of these grow lights to give the plants in my bedroom a little boost since their window faces a shaded courtyard in the center of my building. Rather than invest in something more expensive or put up with hot pinpoints of light stabbing my eyes every day, I designed a cheap printed diffuser to make the system more hospitable.
You can download the design files and read the build instructions on Thingiverse.
As a bonus, they also provide some great Miami Vice lighting, just in case I wanted to film my own episode of ContraPoints.
My partner Shayna and I decided to attend the NYC Resistor Halloween party as The Lone Wanderer and Hancock from Fallout 4. She went all-out to create an exceptional spandex costume, and I wanted to build something that would at least prevent me from looking like a schlub in comparison.
I haven’t sculpted by hand in a while, let alone made an FX prop, but this felt like a manageable chunk to bite off. I started with a face casting, which Shayna built up out of alginate. From there, I made a plaster positive of that casting, and began sculpting a ghoul nose that made it look like my actual nose had fallen off due to intense mutating radiation.
Once I had the sculpture laid out, I added a gutter around the sculpture, waxed the prop, and made a mold box around it to pour a 1-part casting. Then, I removed the sculpture from the silicone, broke down the sculpture to get the original plaster casting back, and mixed up some silicone pigment to match my skin tone as much as possible.
From there, I ran a few castings to try and get something as bubble-free as possible (with mixed success). If I tried this all again, I’d definitely plan around paths for bubbles exiting the mold in a more robust way. However, vacuum casting and the general forgiving nature of zombie makeup ended up pulling this prosthetic through.
I mixed up some silicone with a thixotropic to create a sculpting putty, patched the prosthetic, and then went on to detailing it with some tinted silicone adhesive. I applied the prosthetic using that same adhesive as well and added some basic makeup. I’d intended to stick the bald cap down too, but it was already uncomfortably hot in my lab, and I wanted to avoid becoming a hot sweaty mess during the dance party.
After testing the Flat-Pack Camera Arm I built, I was pretty happy with the results. Happy, except for one detail: the joint at the base of the arm would creep down over time. This wasn’t a problem while taking shots of projects at the bench, given how often I’d have to reposition it anywhere. The big breakdown was trying to capture time lapses. The creep was just too noticeable, and it would never stay in place long enough to keep the action of a day’s hacking in frame.
So, I set out to make some locking plates for the arm, and I think people could find some interesting uses for the process I came up with. The broad strokes of the method are that you design the part you’d like at the end in CAD, design a floor under your part with walls around it (I call this a bathtub), print the bathtub mold you designed, cast the mold using 2-part silicone (making sure it’s nice and level), and cast your final material into that mold. Once you’ve got the knack of replicating parts using 1-part molds, you can get fancier: adding vent holes for letting air escape or labels for your parts or building multiple parts for your molds for even more precise geometry.
Below you can find more information on the whole project:
I’ve been experimenting with printed flexures, and wanted to make a simple tensegrity toy to explore the concept. This design (which you can download on Thingiverse) features both printed tension and compression elements that all build together into a slightly bouncy tensegrity sculpture.
I also optimized the design to allow everything to print at once on my Ultimaker 2+ buildplate. The sculpture assembles with a handful of self-threading torx screws to make it easy for anyone to replicate.