Portal Cube and Portals - Click Here
My approach to modeling my sculpture was to first model the portals so I had an accurate sizing guide for the portals and the hands. I knew the laser cutter bed was 32" x 18" so I sized up the portals to be as large as possible while still fitting in the laser bed.
Modeling the portal cube took around 10 hours total, which is a surprisingly long time for an object that appears so simple at first glance. My main approach was to model just one section of the cube and "Mirror" it to the other sides. However, using the "Mirror" command sometimes created errors where part of the cube was off-shifted just slightly from the other cube. I spent a long time working on the detail in this model, adding small little gaps between the 27 different sections in the cube that gave it more of a Rubik's Cube look. One of the biggest issues that I encountered was that I was unsure of exactly how far to extrude some of the surfaces to give the whole cube accurate proportions.
Once I had created a 3D model of the cube that I was happy, I had to think about the practicalities of 3D printing it. I came up with a mechanism to rotate the cube using a dowel with end caps that rotates freely through the center sections of each of the 3 layers of the cube. I needed to model this mechanism so I could "BooleanDifference" it inside the cubes. My plan was just to 3D parts, but because I had forgotten about accounting for tolerances, I ended up having to remake these parts by hand with wood.
I also had to hollow out all 27 cubes to conserve on filament used and 3D printing time. My original plan was to later fill in these hollowed sections with wood, but I figured that would add too much weight to the total model of the cube, so instead I ended up just gluing the hollow sides together so they were completely hidden, even when the layers rotated.
My plan was to print all 27 cubes over Thanksgiving Break, when no one else was 3D printing. I had so many issues with printing these cubes though, and it was quite frustrating. Each print took anywhere between 15 hours and 2 days to print depending on the cube size and number of them. In total, these prints took around 1.5kg of filament to print, so I had to buy my own to not use up the entire class' supply.
The main issue that I faced was that the filament did not adhere properly to the bed, even when the bed temperature was raised an extra 10°F. I had tried printing with a skirt and a brim below the prints, but both of these did not adhere to the bed properly as well. Because of this, I had to reprint some cubes from my first couple print batches, until I discovered that adding a raft helped fix the bed adhesion problem and made removing supports significantly easier. I don't know why I hadn't tried using a raft at the start, but I could've avoided so much frustration and wasted filament.
The quality of the prints on the new printers were exceptional, yet even with the raft on the prints, I still had some areas on the cubes that weren't perfect. To fix these I used wood filler which helped fill those errors in. Before sanding every surface of the cube, I also painted a layer of wood filler over the whole surface of the cubes which made them smooth and allowed for me to paint them with an even coat. An unintended side effect of using wood filler was that it gave the cubes almost a wood finish, like I had somehow crafted them all from wood by hand.
After spending my entire Thanksgiving break 3D printing, and with all the frustration, I was incredibly overjoyed when I first assembled the Portal Cube. It is roughly 8" x 8" x 8" in size, and all of the detail that I had modeled was captured perfectly by the new 3D printers. This marks the moment that my project turned from just an idea into something that was feasible to actually make.
The worst part of this project was painting 27 individual cubes. In total it took me around 15 hours to paint the whole Portal cube, and a large part of that time was spent just making sure that the lines that I had painted were clean and precise. I could've masked off the areas that I wanted to paint, but that would've taken even more time, so I just binge watched an entire season of Game of Thrones and spent my weekend painting the cubes.
I could not be happier with the final Portal cube. It looks exactly how I imagined, with each side painted a different color to represent a side on the Rubik's Cube. I ended up swapping the white side with pink because I felt the white would not have stood out on the base colors of the cube. I could've just stopped here and called this my final project, but I was committed to the idea of making my sculpture. I wanted it to be by far the most impressive art piece I've ever made, the shining star in my portfolio.
Mechanical Hands - Click Here
My original plan to 3D model the mechanical hands was to take advantage of one of the events in the BTU by getting my hand 3D scanned. The resulting mesh was composed of over 200,000 polygons, but it captured every curve and the shape of my hand with incredible detail.
From there I used the "MeshIntersect" command with a ton of single plane meshes that I had positioned throughout the hand. This gave me a model of my hand composed entirely of curves that I could then "Loft" together.
The resulting model created from loft accurately recreated the shape of my hand and my arm, but when I was trying to give it a mechanical look, I struggled because the model was too organic. I felt that I couldn't accurately transform the model into the vision of a mechanical hand that I wanted, so I ended up just modeling the arm and hand from scratch, and using the 3D scan as a sizing and proportion guide. I was able to incorporate a part of the 3D scan in the form of panels that went around the mechanical arm to shape it more into an arm.
The model of the arm is loosely based on one of the Portal characters "Atlas," with a few redesigns because in the beginning the model was more reminiscent of a rocket ship than an arm. I used a ball as the wrist joint so I could slide a dowel through the center of it to give the wrist some articulation because my main goal with the mechanical hands were to have them be fully pose-able, and wrist articulation added to the ability to shape the hands into any position I wanted.
From the start, I knew that I wanted to use LEGO click hinges in the fingers. This solution was much simpler than trying to model a working joint myself, and the LEGO click hinges were a ratchet style with enough friction that I knew they would be able to hold the finger in whatever position I put them in.
To model the fingers, I first had to semi-recreate the LEGO click hinges by taking a ridiculous number of measurements and recreating their complex shape with simple polygons. The fingers and the hands themselves were made by creating a blocky shape and using "CageEdit" to emphasize the divots in the palm and to curve the edges of the blocks until their shape was in between organic and mechanical. I am very pleased with how the 3D model of this hand came out, because from the beginning of this class, my whole goal was to be able to model a mechanical hand. My dream job is to make prosthetics, so being able to design one in Rhino with simple, functional joints is a huge accomplishment towards that goal.
I couldn't be happier with how my 3D printed hands came out! We had just gotten white filament the day that I tried to print them, which worked out in my favor because it meant I had to do almost no painting on the hands themselves. The best part about these hands was completely unintentional. Because I had created such complex curves and shapes with "CageEdit" the 3D printer could not accurately capture the slope of that curve in a single layer, so the resultant effect gave the fingers and the palms the texture and design of a fingerprint, which made the model that extra bit more impressive.
I was also happy to find out that taking the time to remodel the LEGO click hinges was worth the effort because they fit perfectly into the negative space on the 3D prints. The black also contrasted the white of the hands, and I later dry-brushed over the black with some silver paint to give the LEGO hinges a metallic feel.
I had seen a spool of ribbon cable lying around in the BTU back in August, and since then I had wanted to incorporate it into my projects in some way because it was aesthetically pleasing to look at. I ended up fashioning some connectors to the hands out of wood that the ribbon cable extended out of, with my plan being to wrap them around the arms to create some sort of circuitry effect.
Plaster Hands
Even though I had the scan of my 3D hands, I didn't want to print that scan out because I knew it wouldn't seem as realistic as plaster casting my hands would have. I have plaster cast my hands several times in the past for other art projects, so I was familiar with the process. Unfortunately, I had originally bought only 0.85lb of casting alginate, which was only enough to cast up to the wrist of one hand. Because of that, I had to wait a week for another 3 lb of alginate to arrive. In that time I had frozen the original cast I made to preserve it, and then I just filled that frozen mold up to my arm when the new alginate arrived. Unfortunately, because that mold had been frozen, it lost a lot of the detail of my hand, and I ended up having to scrap the first plaster cast I made. In addition, the plaster cast made my hand look at a weird angle, so I was very happy to scrap the first on.
I knew that the plaster hands would weigh about as much as a brick, and including the fact that they had to support the portal cube, I decided that the best way to mount the hands on the portal was to inset a carriage bolt inside the plaster cast while it was still wet. This meant that the bolt was incorporated into the structure of the hands, and that all the weight would be on the bolt instead of on the plaster itself.
Although the second set of hands I cast had much crisper detail, when I removed them from the alginate mold, several of the fingers on each hand snapped off. This was easy enough to fix with superglue and some leftover wood filler to fill in the gaps. Unfortunately, it was not the first time that the fingers broke off. The plaster hands need about a week to fully dry and set up, but due to time constraints, I could only let them sit for a day before I had to mount them, which meant their structural integrity was low, and that the fingers were more likely to break again and again.
Painting the hands was the most important part because it gave them the realism that I wanted to achieve in order to emphasize the contrast between the organic set of hands and the mechanical ones. I was very pleased with my painting job, and I think the hands look almost exactly like my own.
Assembly
One challenge that I faced was how to mount the portals on the walls. In total, the weight of the sculpture was around 20 lb, which meant my original idea of using command strip hooks was unfeasible. I knew that I had to mount the portals in a way that was temporary and movable so I could take the project home from the BTU where I was working on it. My solution was to build a fake section of wall and floor. I bought a large sheet of Pine wood from Home Depot and was able to stain it almost the same color as the walls I had originally rendered. I am quite pleased with the color of the wood, because it contrasts quite well with the portals placed against it, and it gives the whole piece that sculptural quality that I was going for.
Due to time constraints and the problem of the fingers continuously breaking off, I decided to glue the hands onto the Portal cube directly. I had originally wanted to be able to easily remove the Portal cube so that it could be displayed on its own, but gluing it to the hands helped the stability of the whole model as well.
For the portals themselves, I added a mirror film on the pieces of acrylic that I had laser cut to give the portals a reflective, more portal-like appearance. The LED lights around the side were also a nice touch that made the portals give off a glow as if they were active, helping the whole fantasy appearance of the sculpture.
Mounting the portal assembly with the plaster hands and the Portal cube was quite the tricky process, and it require two other helpers to hold up the whole assembly while the portal was being drilled into the wooden wall. Some fingers had snapped off again in the process, so I was worried about the strength of the plaster hands, yet when we all stepped back, the portal stayed up, and the carriage bolts in the plaster hands served their purpose and kept the Portal cube and the hands properly suspended. I was elated with the result, and it was almost exactly like the render that I had made almost a month prior. You may notice that the portal is a solid blue instead of the transition that I had painted on. That's because I accidentally glued the wrong side of the portal down to the mirror earlier on.
Tragedy
Unfortunately, these were the last pictures i took of my sculpture before tragedy struck. I had the other portal mounted and the mechanical hands were in position too. I was literally drilling the last hole in the wood before the blue portal fell off the wall, taking the plaster hands with it where they shattered on the floor.
The portal fell off the wooden wall because the screws weren't drilled through properly. This is due to the fact that I had a drill bit that was too small for the screws, but I did not have the time to get the proper sized one. I really felt like I was rushed with the last half of this project. Even though I had been working on it continuously for the past month, when it was the day before it was due, and it still wasn't assembled, I was frantic. I spent all night working on the project, making the fake wooden walls, painting the plaster hands, mirroring the the acrylic, etc. I had to make some sacrifices in detail and my precision, which is why the whole sculpture ended up being so fragile. When it broke 5 minutes before it was due, I was crushed. I had poured time, money, and my soul into making this piece the greatest accomplishment in my art portfolio.
Future
I plan on rebuilding this project sometime in the near future. My goal is to get it featured in the Spring 2020 Atlas Expo, so hopefully I can finish this project again by then. I was able to salvage the portals, the Portal Cube, and the mechanical hands, so the only major thing I have to redo is the plaster hands. Next time, I plan on using resin to cast them so the fingers will be much more durable than before. In the process of making this project, I also learned about a few things I cold improve, such as getting the mirror film on the acrylic to have less bubbles in it to create a smoother surface.
The Portal Cube's paint job is very scuffed from wear and tear, but I can easily repaint it so it looks as good as new. With the mechanical hands, I opted to shorten the arm section, and I cut the exposed mechanical wires so to fit that new length. Currently, the mechanical hands almost have this fading effect because of the cut wires and tubing, which I like as a standalone model. I'll extend the wires and tubes in the future when I reassemble this project.
I am excited to rebuild this project in the near future, and I can't wait to see it completed, with the LED lights turned on and the Portal Cube suspended by the hands alone. Although it sucks that my project broke, I know that I can fix it, and without a deadline I can spend more time getting those finer details that turn the project from something that's cool to something incredible.
