I used this project as an opportunity to explore non-humanoid animations, drawing on concepts from the first stop-motion assignment.
I used Blender to create a model of an LED, and put together a quick rig with a silly idle/walk/jump animation set. I had hoped to refine the animations, but spent so much time troubleshooting the Blender to Unreal pipeline that I did not have time nor energy to go back and refine.
I spent a good hour and a half reading up on and experimenting with the Blender to Unreal export/import process. After learning a decent amount about the Unreal animation engine I discovered something was off with my bone structure; Unreal was adding an extra “root” bone from the origin of the world that offset all the other bones in the armature.
After even more digging, I realised that my problem I was having was a direct result of a previous bug between Blender and Unreal, where Unreal removed the first bone of any skeleton named “Armature” (the Default in Blender). My best guess is that the recent bugfix has overcompensated and, at least in my case, is adding an extra bone to my “Armature” skeleton.
The simple fix was to change the skeleton name from “Armature” – I chose ‘Skeleton’. After that everything worked fine. It’s a simple solution but took over 2 hours to discover.
In my first assignment for Fabrication I created a hand-cranked flash light with parts from the junk shelf. For this final assignment I thought it would be wonderfully fitting to expand on that concept and create a hand crank that could power my other assignments, or any projects in the future.
Since discovering my partner’s sister makes plant stands out of copper I’ve been wanting to try use copper in a project, and since seeing Ben Light’s 2×4 enclosures I have been wanting to explore that technique. With this week’s prompt – “use 2 materials” – I thought I could bring those two goals together.
Materials & Connectors
I purchased some copper pipe, connectors, and a pipe cutter from Home Depot. I found a piece of 2×4 in the scrap shelf. I used M3 and M4 machine screws and magnets as fasteners.
As shown in the 2×4 instructional video, I used the band saw to cut off a small section of one of the larger sides. Since it’s the band saw the piece I cut was awfully wonky, but I figured it wouldn’t matter too much since the grains on either side would match up regardless.
I used a forstner bit to core out the inside.
And a knife to straighten out the edges a bit, making room for a bulky arduino.
The base was a little warped so I thought that adding some aluminium supports might help straighten it out, as well as add a bit of aesthetic charm. I had some magnets lying around that I wanted to try use to hold the lid in place. Installing them was a matter of carefully measuring the holes, finding the right size drill bit, and a frugal application of superglue. I thoroughly enjoy the effect.
I went through a couple of design iterations for the copper structure on top. My primary motivation was that I wanted some sort of light display coming from the structure, but beyond that I was simply playing around. The first was far too simple for my liking:
But I eventually found an arrangement of pieces sort of reminiscent of a Star Trek ship (I say sort of with a lot of leeway)
Drilling holes in the side of copper, for the plastic light rods, was a little difficult. It seemed to me as though the copper was so soft that as I stepped up in bit size the larger bits had a tendency to eat away big chunks of the metal, leaving wonky and irregular holes. Perhaps for version 2 I’ll use the drill press for more accuracy.
The final step was to wire the electricals, align the LEDs with the drilled holes (that was a tedious nightmare), and assemble the structure. The copper is soft enough that I could slightly deform the ends of each section so that the rods simply pressure fit into their connectors. This was a lot quicker and safer than trying to learn how to braze copper within a week.
Since there are no external inputs other than power, all I needed was a small hole for a DC jack I had. Sadly it was not a panel mount so I had to figure out a way to attach it (superglue).
A fellow student saw the completed version and said “it looks great! What is it?”
Honestly, I don’t know.
In another class, Basic Analog Circuits, we’ve recently been covering the 555 timer. One of my favourite ridiculous ways to use the 555 timer is in an Atari Punk Console – basically a weird noise machine. Two 555 chips are used to generate interfering square waves, the result being a ‘techno punk’ sound reminiscent of the music and sounds on an early Atari console. The small, annoying console is a perfect circuit to enclose for this week’s assignment.
I wanted to create the circuit in an unexpected place – to the junk shelf! The first potential housing I found was a broken audio mixer. I eventually decided it was too big for my liking, but I did salvage some sweet potentiometer knobs from it.
My second find was a clearly broken Macbook charger. I was attracted to it instantly; there’s something about the sealed, pristine box of mystery that immediately makes me want to hack it. At first I didn’t even know if it was possible to open it without completely breaking the housing, but after a bit of careful prying I managed to snap the plastic apart.
There’s an incredibly well-packed circuit inside, those Apple engineers are magicians. I discarded it immediately; I had my own electronics to put in.
I started to play around with the built-in openings, and to my immediate delight found that our own ITP potentiometers fit well with the hole for the charging cable.
What’s more, the washer and nut I’d salvaged from the mixer secures the pot in perfectly.
I took this as a sign: my Atari Punk Console simply had to go in this charger. The space inside is cozy; I could tell my circuit would be a tight fit. No room for large breadboards or stray wires; everything would have to be soldered and placed carefully.
Since the Atari Punk Console requires 2 variable resistors for best effect, I had to think about where the second would go. Luckily, the little metal nib thing that keeps the wall adapter in place popped off quite easily, revealing another hole. Unfortunately, that hole was slightly too small for a pot. A quick spin on the drill sorted that out.
I got tunnel visioned while soldering so unfortunately I don’t have any process pics of that part, but here’s the before and after. The first pic is the prototyping board, the second is the soldered circuit, and the third is all the pieces fitting snugly together. I’ll be honest that the way the internal pieces fit together is half design, half luck.
This week’s assignment called for us to use one of ITP’s laser cutters. Though I booked the 50 Watt engraver, it was in use so I used the 60 Watt machine for this project.
I knew from the get go I wanted to play around with Total Internal Reflection acrylic, or as best I could find in Canal Plastics. I went through a few idea iterations, including a sort of light-up piano with custom cut keys. Although exciting, that idea took a back-burner when I thought a little about scoping and the timeline that we had for this project (one week). I settled on a simple but (hopefully) elegant design for a light up object, using a short strip of Neopixels.
The design was simple enough to translate from terrible hand drawings to nice clean vectors on Illustrator. I used the digital calipers to accurately measure the thickness of the material (it was definitely not exactly 1/4″ as advertised), so as to know the size of the hole to cut. I also used the calipers to determine the distance between each hole based on the distance between each Neopixel on the strip.
Since we have to be careful about the width of saw blades when cutting, I figure we should also be aware that the laser has a width. I did a little searching and found that the area the laser vapourises is called the kerf. I shaved a few millimeters off my hole size to accommodate the kerf.
I wanted to explore how different shapes and cuts affect the propagation of light within the piece. To be honest I wasn’t particularly meticulous about this as I was mostly just curious, but I gleaned some interesting information nevertheless.
Essentially I drew a few interesting lines within a few rectangles, and a few odd shapes as well. once I cut these out I help them up to the LED light on my phone to see if they’d behave differently.
I found that each different shape creates a different… ‘hotspot’ at the edge of the acrylic – what I’m calling the point where the most of the original light source is concentrated. With no shape (or straight perpendicular lines), that hotspot appears on the plane directly opposite the original light source.
Images of experiment results still to come
Once I had played around with different shapes, I settled on – for this project at least – plain empty rounded rectangles; the light seemed to have the most even spread across the sides, and I liked the clean, unmarred look. It reminded me a little of Superman’s fortress of solitude.
One thing I found interesting; when I purchased the black sheet of acrylic, the side I looked at was the glossy side, and I just assumed that the reverse side would be the same. The reverse was in fact a matte finish – which worked out well for the look of my project. I can’t rely of happy accidents every time, though, so I’ll definitely check both sides next time.
I made sure to include 3mm screw holes for the M3 screws I have lying around. The (roughly) 1/4″ acrylic required a bit of testing to get the right settings (ended up as 8/100/5000 with three passes).
A few weeks ago I helped out a friend who lit up the Cornell clock tower using sequences from various human DNA for the light patterns. My part in the project mostly involved advising on the types of lights to use. They went with Adafruit Pixies.
Since then I’ve been thinking about ways to go bigger than that. As all LED enthusiasts know, brighter is better. I had exactly 5 spare LED 120V bulbs lying around at home (for general household use), so once the project prompt to “Make 5 of Something” was announced, I thought it’d be a good opportunity to experiment with much brighter lights than our typical ITP fair of breadboarded LEDs.
I also have a (pretty rational, I think) fear of using 120V wall power in my projects, so this project was a good way to delve a little further into going above the usual 5V – 12V.
Materials & Tools
Since this project involves dealing with relatively high voltage, I did not want to skimp on the electrical components (unlike last week, when everything came off the junk shelf). For the electricals I used a combination of purchased and owned components:
After sketching out a basic design of what the build could look like, I briefly browsed the unclaimed materials bin to see if there was anything useful. There was! An unclaimed 5 1/2″ wide board of pine (?) was perfect for me.
I adjusted my sketch slightly (from 5″ wide to 5 1/2″), but otherwise this looked like pretty much all the material I’d need (except for a few details later on).
The first thing I needed to do was find a way to cut the right sized hole (about 1 1/2″) in the center of the block. Amazingly, the ITP toolbox had just the right sized drill bit. Less amazingly, the bit was incomplete and useless. Even more amazingly John let me borrow his (complete) version of the tool to make the cuts:
Cutting things down to size
The end of the board I had was a bit funny-looking, so I cut it off – with a bit of space so that I could experiment with the hole-making tool, screw holes, and such. I suppose that piece was my first pancake.
I cut a test hole in the pancake, and the lightbulb housing fit perfectly.
Pluribus E Unum? – Out of One, Many
Once the basic premise of my contraption – that the lightbulb housing could fit snugly in the hole – was confirmed, I turned my attention to making many of them.
Using the compound sliding miter saw (incidentally the second coolest sounding tool next to the LAZER cutter) and a stop block I made quick work of five 5 1/2″ x 5 1/2″ squares and ten 5 1/2″ x 1/2″ rectangles.
After that, I aligned and cut the center holes. No fancy jig for that, each hole was measured manually.
Although measuring 5 center points was doable, the prospect of 4 screw holes for the legs on each of the 5 boards, so 20 careful measurements in total, seemed tedious. At this point I fashioned a rudimentary jig to help align the screw holes the same distance from each of the 4 corners of each square:
It worked marvelously and saved me at least minutes!
With the screw holes in place the next step was to attach the legs I’d cut earlier. Each leg needed to be quickly held in place as I drilled into it, but that did not require a lot of clamping force. As such I used the speed clamps so I could align with one hand and clamp with the other.
I found a cheap countersink bit at a hardware store so tested it out here. It sucks (i.e. creates the beveled hole slowly) but does the job.
Even more legs
One problem that arose later in the process involved the intended height of each stand. I had hoped to bend the metal pins of the E27 socket out so that the stand could rest on its wooden legs, but when I tried the metal pins wouldn’t budge without cracking the plastic housing.
A quick solution to this problem was to raise the platform a bit higher. I didn’t have enough more of the 5 1/2″ board to make more legs, so I yet again raided the junk shelf in desperation. I found a perfect set of dowels that provided an adequate solution:
I intended to raise the platforms a little higher using the dowels as legs. This required drilling repeatable holes in the legs I’d already attached to the squares. Time for another jig! If you can even call it that. Really it was just a piece of scrap wood cut to size with evenly spaced holes for a center hole punch:
I used the band saw (unpictured) and stop block to cut the dowels evenly. I thought they seemed thin enough that the band saw wouldn’t make too much of a hash job of their size. I was mistaken. The dowel pieces came out a bit uneven, but because they’re such a tight fit in the holes I drilled I could compensate the depth each dowel went in.
Voila, 5 light stands ready to be filled with electricals.
I created one circuit as a proof of concept, then turned to creating all the rest at once.
The real tedium of this project manifested in the cutting, stripping, and soldering of all the wires involved. Added to the mix of tedium was the anxiety involved in dealing with 120V AC wall power, a brand of electricity you may recall I am less confident with. Incorrectly wire a 5V Arduino circuit and you might fry your component. Incorrectly wire AC wall power and you might fry your building, or your brain.
That’s all to say I was slower and more careful than usual, and as such this process took quite a while. Nevertheless, I was able to port some of the mad new repeatability skills I’ve gleaned into this process as well.
With wire cutters in hand, I cut all the different wires down to size.
Then, in one go, I stripped all the wires. Finally, most of the soldering was done in one go as well. It wasn’t exactly a mindless process; as mentioned above I had to be on alert so as to prevent miswiring-related fires or electrocution. It was, however, handy to perform all similar tasks at once, and save some time from changing tools and thought patterns.
Voila again, 5 ready-to-flash 120V AC powered, Arduino controlled lights.
For the finishing touches: a liberal application of cable ties.
The inspiration for this project came from another project I did a while ago, on a different continent, with far less electronics understanding than I have now. In short, I needed to light up and make “interactive” a sign, with no access to wall power. The result was A Hand Cranked LED Sign.
That project started me thinking about different sources of energy, and how we so easily take for granted electricity that comes out of our walls. I have thus been thinking about sustainability in Interactive Media for a while, and this week’s assignment was a wonderful catalyst to get started experimenting. My goal is to create a flashlight that is both portable and powered completely by the user. Continue reading “Junk: The Green Flashlight”