Quarter 1: Window Screens (styrofoam)

4.021, FALL 2018

The first quarter of 4.021 focused on the traditional design process, where a designer follows through on an idea with successive iterations and documentation to final come up with a final result. The task was to create a functional screen wall out of white styrofoam. It was phenomenal to see the variety of responses during our final presentations, as people's concepts led them to explore everything from Conway's Game of Life, to combining visual aesthetic with soundproofing, to novel manufacturing techniques like melting the foam using a hot air gun, and Moiré patterns.

Process and commentary here (below final presentation)

Gallery here


Context → Concept → Draw → Make → Iterate → Analysis/Narrative → Present




When I began investigating screens it was important to me to understand what a window screen was — what function did it fulfil? What were the traditional and modern precedents to creating designs? Two sorts most appealed to me: those based on geometric patterns, be they regular or irregular (right), and those which were only translucent, like Japanese Shoji screens. 

Moreover, I came to contemplate smocking, a pattern type I had always found appealing. Generally, smocking is taken to mean a method of gathering fabric together to create small pleats. This can be done in the form of simple ribs, or, in a style more commonly known as Canadian smocking, create a variety of textually interesting patterns. It has been used to great effect in costume design, for instance for Game of Thrones (below, right).


(The blue sample is actually one I made for GCSE four years ago. At the time I was not able to use it for my project, so it was very fulfilling to find an application for it now.)

Screenshot 2018-10-27 at 4.20.00 PM.png
Screenshot 2018-10-27 at 4.20.04 PM.png

Initial Experimentation

My first experiments were rough. Initially, we were not given the tools with which we would make our final project, so I attempted to mould to foam using everything from a hot (though I boiled it it never got hotter than "warm", so it effectively failed as a tool) to a knife, with which I created the sample to the right.

Since this was to be a screen, achieving translucency was important to me. While I could have cut straight through the foam, I hoped to create a gentle ambient light, which would only be possible if I cut the foam as thinly as I could. For instance, to the left, my experimentation with pattern compositions heavily features patterns overlaid on one another, using a mesh of light to determine the final product. Indeed, my initial idea was to create several layers of "smocked" material, which could then be moved relative to one another to generate new patterns.

Screenshot 2018-10-27 at 4.33.23 PM.png
Screenshot 2018-10-27 at 4.40.51 PM.png

Concept sketches

Initial Drawings

Based on sample photographs of smocked fabrics, I rendered three-dimensional versions of my two favourites in Rhino. The process of making these helped me gain a better understanding of the shape of the pattern, which then led to more effective experimentation. 

In the following images, I attempt to create modular units which could, given a sufficient quantity, be put together to create a final screen. In section, parts of the pattern would be thin enough to create a translucent surface.

180922_Screen A Shaded.jpg
180922_Screen A Plan.jpg
180922_Screen 2 Pen Sketch.jpg

Renderings (Rhino)


Studying modularity for respective patterns:

Left, blocks layered on top of a thin sheet

Right, tessallation of arrowhead shapes

It was almost by accident, while I tried to figure out how to reproduce a hollow, folded object using subtractive construction methods, that I happened upon the idea to fold the thinly cut foam I had used as a backboard for the blocks. 

It was not long before I realised that I could abandon the frame altogether and, given enough reinforcement, treat the foam like cloth, and truly smock it (below.)

Subtractive construction method


Textile manipulation: smocking


The gif to the left demonstrates how the smocking process works: the textile is pulled together from the back according to a regular grid, and once turned over, reveals the pattern. Initially, I attempted to gather the foam using glue and other adhesives, but found that the most reliable way to do it was traditional needle and thread. As you can see above, however, the foam resists being folded, and cracks. To get around this issue, I used cling wrap as a supporting material. (As shown below, occasionally the spray adhesive melted the foam...)

I cut the foam using a hot wire and a straight edge, such that I was able to shave off regular slices, which were then carefully adhered into a large sheet.


Pattern samples (ordinary lighting, backlight)



IMG_5753 2.jpg


As with all art, the final product is only so good as the pictures you make of it. In 4.021 we were also taught how to document the pieces using Rhino and Adobe Illustrator. Since my final outcome was so complex, it would have been impossible to model it in Rhino by hand. Hence, my first instinct was to use Illustrator to create a vector file. 

IMG_5752 2.jpg
Screenshot 2018-10-22 at 9.41.30 AM.png

Simply applying an automatic image trace function, however, turned out to be deeply dissatisfying. Even using noise reduction, it merely looked like an edited black-and-white photograph, and provided no additional information as to the construction or the system that underlay the pattern. 

Ultimately, I settled on hand-tracing the patterns to create a clear image.

IMG_5751 2.jpg

Vector files (dissatisfying)

Original photographs


Final Documentation​

With this method, I was also able to create a better visualisation of the relationship between the front and the back (below). It was also very interesting for me to be able to break down a process that almost seemed magical — by simply pulling two sides of fabric together, one could create triangles, overlapped lattices, and even (in more complicated smocking examples), roses.

The blue rendering is of the front of the pattern, with the raised sections paler than the dark sections. I used rendering so that the eye might more easily differentiate between the front and the back when the back (green) was overlaid. Once viewed together, one can clearly see how fabric being pinched together in the back (an intersection of several green lines) creates a dipped section (dark blue) in the front.

Screenshot 2018-10-23 at 4.14.18 PM.png
Screenshot 2018-10-23 at 4.14.29 PM.png

Drawing, front

Illustrative rendering

Drawing, back