Overview

The Cardboard Eames Chair is a chair designed as the IKEA alternative for the constantly moving college students. As a college student I was simultaneously frustrated by my lack of furniture and by the bundles of clean cardboard that gets thrown out everyday on the streets of NYC. 

Made entirely of cardboard, the chair was designed to be treated just like any other chair with the exception that it could be dumped straight into recycling at the end of its life cycle. The prototypes were built using exclusively reclaimed cardboard meaning no additional waste was created by this project. The final results are intended for anyone with a laser cutter to be able to produce their own chair, either by using my files or by following the same steps. 

The Why

I’ve thought about building solutions from cardboard since my first cramped NYC dorm but considering that I didn’t have access to a laser cutter, building anything was really impractical. Instead, the idea stewed in my head and it turned into this grand idea of saving all the wasted from the streets to turn them into flat-packed IKEA like furniture that would require minimal assembly on location. Considering that I didn’t have any ideas to do my senior project on, I jumped at the opportunity to finally execute this dream.

When I had previously built cardboard pieces for my own space, I was used to balancing the needs of the space to match the cardboard I had available to minimize waste. But for this, I needed to design it be broader reaching. I wanted to find methodologies that could be used universally. And do I picked a cabinet and a chair as representative for a majority of furniture. I started with the chair as I perceived to be more difficult and it ended up taking up all of my time.

Examples of cardboard furniture online either look like for children or look like human sized pieces of origami. The examples were frequently clunky and boxy, the opposite of the more minimal furniture that is currently popular. There were always exposed edges and an overall lack of polish. It was always obvious that it was built from cardboard and cardboard furniture on Kickstarter always had to prove and overcompensate how much weight the product could support. Throughout the project I always conscious of how my choices would affect how the end product would look. 

Research

I researched different building methods and found 3 distinct ideas to try out; stacked slices of cardboard like that of the Youtuber Kikomoda, a grid like the ones from Mike Sheldrake’s cardboard surfboards, and a variation of the paper shells made from Pepakura Designer. 

The highlights of Kikomoda’s methodology is that strength is created by laminating all the pieces together and that they create a clean shell by folding in between the flutes of the cardboard. The result is a smooth and strong cohesive piece of cardboard with sharp corners. To emulate this method, I separated the hex shelf into six equal pieces and cut out the base trapezoids using a laser cutter. I attempted to also laser cut the outer shell but if the lines didn’t match up the trophs in the flutes of the cardboard, it wouldn’t create that crisp corner that I required. Building the shells by hand did achieve cleaner corners but took a fair amount of time. Combined with the process of gluing everything up, for what I was intending to build this method was unnecessarily time intensive and just inconsistent. I didn’t end up finishing the shelf. 

Next I recreated Sheldrake’s smallest cardboard surfboard at a 75% scale to better understand how it was built and how it was designed. Despite having been created in the early 2000’s, it still holds up over time and there were even people that I knew who had either built their own or made a variation of one. From building the surfboard, I really appreciated the numbering and notch system that allowed the individual pieces to be cut in any order, allowing for maximum efficiency and limited waste of raw cardboard. I was surprised at how well it held together without any glue but what I wanted to emulate the most was his “quarter isogrid” as it not only looked nice but also supported different directions of motion. However I had no way to recreating it since Sheldrake had used a custom program and the closest

someone came to recreating it used the CAD software Rhino. Around this time I also learned about Slicer for Fusion360 and after building a prototype of the base of the chair, it turned out that a simple right angle grid would suffice. 

Going back to my original intentions, while the grid was sturdy enough, it wasn’t enough. While Kikomoda’s method naturally created a smooth polished exterior, I had to find a way to achieve a similar look. And because of the nature of cardboard, the joints were weak when the force wasn’t spread out and someone could rotate with the chair but at the expense of the individual joints. To take care of all these problems, I had to have a shell to cover the grid.  

For the shell I attempted both Slicer’s folded panels option and Pepakura. Considering that Pepakura was designed for paper models, Slicer ended up being the better of the two because it was better equipped to handle materials of different thicknesses. It was also much easier to bring into the laser cutter and since I had chosen Slicer for the internal gird, it meant one less software to deal with, especially since Slicer and Fusion 360 are free for students. 

Building the Chair

The chair was divided into four different parts, the internal grid of the chair, the shell of the chair, the internal grid of the legs, and the shell of the legs. I made a smaller version of each before building the larger version. 

Building the grid was relatively straight forward. In the first prototype of the internal grid for the chair, one axis was parallel to the ground while the other was perpendicular. While it worked for the top half of the chair, the seat at the bottom half was unbuildable. Changing the first axis from parallel to the ground to 45 degrees fixed the issue. Similarly was done for the legs since they designed to be slanted at 15 degrees. For the legs, one axis was parallel to the ground while the second was also angled at 15 degrees. 

In putting together the grids, if all thickness of the cardboard was inputted correctly, the joints would be snug enough to be able to hold together without glue. If that measurement was made too conservatively, I would have to force them together, frequently damaging the cardboard. But considering that I used different sources of cardboard, I typically used the largest thickness instead with the only issue being that the initial pieces wouldn’t stay in place. The easy fix was to use a little bit of hot glue as the grid became more rigid once more pieces were added.

The shell was by far the most complicated part of the whole chair as it took a lot of trial and error. Generally the shell was made by scaling up the original model. The shell of the chair was always made with a back half and a front half so that the grid could be dropped in like a box with the front part like a lid to seal it shut. The shape of the legs’ shell was straightforward and wrapped like a Christmas present.

I built the smaller model of the entire chair first to have a working concept for a presentation. Rather than doing any math, I scaled it by an arbitrary amount. After building it, the height of the shell was correct but fell short on the length. For the front part of the shell, I glued the top half to the bottom half at the wrong angle so the width didn't fit regardless. 

I did the bulk of the testing on the leg pieces as they were a simpler shape and easier to rebuild in the case of any accidents. For the first test, I built a shell for a leg using the same dimensions as the internal grid of the same price  and then measured gaps. From there I continued to increase the width of the model with Slicer proportionally scaling the other dimensions. Atl the 6th trial the shell still wasn’t wide enough to cover the circumference of the leg but had clearly become too tall. I decided on using the height from the 4th trial because it allowed the end caps to be on the outside which I assumed allowed for better distribution of force between the internal grid and the shell. I evenly scaled the other two dimensions until there was a perfect fit with the 8th trial. 

My biggest concern at this point was that I couldn’t find a rule that would allow me to translate my findings to the full scale models. It wasn’t enough to compensate for the thickness of the cardboard shell and just adding the extra 6 mm that was required to make the smaller version wouldn’t work. After much pondering and a good’s night rest, I did the math and according proper fit was 1.14x the measurement of the original model. And so for the full size, I started by printing out a shell with all the dimensions scaled by 1.1x, and then measured the gap. From there, I increased the shell by that gap until it was the proper height

 then found the height for the shell, and kept increasing the other two dimensions until it could fit around the circumference of the leg. This took about four trials, but I only printed the first shell with 1.1x the dimensions and then measured the gaps. After creating the shell using Slicer, I could measure in Illustrator using the measurement tool to see if the shell had increased enough. The result was a relatively perfect fit in the end.

Testing

While I wasn’t able to properly test the whole chair with all the components together, I was able to test the sittable part of the chair. On just the internal grid, I initially tested by placing the chair on the floor and it supported my weight. I also had a friend from the Army who weighed over 200lbs sit on the chair while it was on a normal chair and it also didn’t break. Another friend sat on the chair while balancing himself on one leg. 

While the grid didn’t crack in half or completely fall apart, a couple of the individual joints, especially in the lower half of the seat that wasn’t touching the floor, tore a bit. The chair could also rotate left and right a bit but as it really wasn’t meant to, that also stressed the joints.

I was only able to get one test of the seat with the grid together as one cohesive piece. As I only had enough time to apply heavy amounts of glue to one leg, three of the legs broke during the showcase exhibition and the last one refused to come off. I sat the leg between two tables so that the seat could rest on the tables and had a friend sit on it. While she was relatively light, the chair still held.