Structures #1

In the midst of our Studio and Construction Systems work, our other classes continued ahead at full steam. In Structures 1, after having learned all about beams, columns, etc. and how to calculate their forces, we were tasked to construct a pedestrian bridge to scale. This bridge was to be loaded using a machine in the shop, to test its capacity. The structure also had to include a covering, but had to be open at the very center where the load would be placed.

I teamed up with classmates Jesse, Lauren, Erik, and Miguel for this project. We began with a simple arched design, with a flat path placed below rather than above the arch. Originally we had misunderstood the assignment and designed the bridge such that the load would be placed on the arch. Our hope was that the pedestrian walkway below would act as a tension member to hold the ends of the arch in place (as it had not material to brace against as it might in a building). We constructed the bridge out of laminated sheets of MDF, and this was ultimately our downfall.

First iteration of pedestrian bridge model for Structures 1. Constructed from laminated sheets of MDF.

First iteration of pedestrian bridge for Structures 1. Constructed from laminated sheets of MDF.

When tested, the bridge failed much earlier than we had hoped or anticipated – very quickly the laminated sheets sheared apart, as the glue was not able to hold them together under the stress of the load.

First iteration of pedestrian bridge model for Structures 1, being loaded for testing.

First iteration of pedestrian bridge for Structures 1, being loaded for testing.

Back to the drawing board for the second and final design, knowing this time that if we were to laminate the structure, we would need to provide some sort of cross-member to keep the material from shearing apart.

Moving Forward: Triangulation

Well, it’s been quite a while. About two months ago I said I was back for the summer, but then new commitments started up (namely work around the house as well as an internship at TIA Architects in Amherst, MA) and I just haven’t gotten around to the blog. Perhaps I should just give into the fact that I’m not going to be able to update this consistently, and that’s just how it will be.

At any rate, I would like to try and finish up my 3rd semester’s work so that I can actually move onto the 4th semester before I start my 5th in September. Holy moly, time flies! Let’s see if I can’t get back into the swing of things.

After our midterm with our artist residencies, we all returned to our desks to begin puzzling out the interior of our forms. I had created and organized the exterior shell around a basic space plan, but now I needed to actually see how I might  fit things into the interior. How would I address vertical circulation? Windows and other apertures? What about entrances? These are things that I really hadn’t thought about, and now I needed to.

My first step was to figure out how to meld the existing form for more shape and solidity. Up to this point, the project had been constantly fighting the “mayonnaise plague”, i.e. becoming so goopy as to no longer be recognizable. I needed to really grab the building and lift it out of the slop once and for all. The most drastic change I performed on the structure was a triangulation of the entire surface. I left the surrounding landscape smooth, and allowed this smoothness to bleed onto the surface of the structure at various points, namely the roof (which I imagined as a green roof, a grass-covered extension of the hill behind it. I was trying to capture the idea of “viewing platforms” that the owners of Storm King had mentioned during our tour) and the interior tunnels through the building.

I then created a sectional model to illustrate this change in exterior, as well as to give a peek into how the interior might unfold. Since I wanted the structure to smoothly rise from the landscape, I kept the entire “skin” of the model one material.

Sectional model of Storm King Residency, after triangulation. MDF skeleton with Bristol Board skin.

Sectional model of Storm King Residency, after triangulation. MDF skeleton with Bristol Board skin.

I also began to think about how I might lay out the interior, particularly the upper level with the living spaces for the artists. Following a similar triangulation pattern, I came up with this basic layout seen below. Though this ended up being adjusted fairly drastically as I created windows and vertical circulation, it was a decent place to start.

A preliminary plan for the 3rd floor, where artists in residence would live.

A preliminary plan for the 3rd floor, where artists in residence would live.

Phase 1, Part 3 – Generative Model

At  this point in our 3rd semester, we had all not only thoroughly examined and interpreted our sculpture, but also explored how it might be changed and reformed to meet different needs. Our various analytical and generative drawings attempted to express this clearly. But of course, only so much can be communicated through drawing, and so it was time to build our first model of the semester. In my section, we were all tasked to reconstruct the generative 9 ” cube models we had designed digitally, using a particular stacking method. This was most easily done through a program called “123D Make” – a free program put out by Autodesk that allows you to create physical models from digital ones with relative ease. Import a .stl file (which can be exported from either Rhino or Maya), and the software calculates one of various construction methods, including stacked slices, interlocking grids, and polygonal folding (click here to check out the website). The program will then create the cut files, ready to be sent to the laser cutter (although the settings on our laser cutters require a bit of tweaking first).

For this particular model, all 15 of us in Prof. Saunder’s section used a vertical stacking method, with 1/16 Taskboard as our construction material. Using this software coupled with the laser cutter, model preparation time was extremely quick. And given the ease of assembly with this particular system, actually constructing the model was even quicker. These models were prepared and then constructed in probably 2-3 days. The hardest part was finding a time when the laser cutter was open!

Below are a few photographs of my generative model. All photos taken by classmates Andrew Kim and Christian Gartland.

EDIT 01/13 – Images now correctly show.

Antoine Pevsner - Sculptural Analysis 9" Generative Model (vertically stacked taskboard)

Antoine Pevsner – Sculptural Analysis
9″ Generative Model (vertically stacked taskboard)

Antoine Pevsner - Sculptural Analysis 9" Generative Model (vertically stacked taskboard)

Antoine Pevsner – Sculptural Analysis
9″ Generative Model (vertically stacked taskboard)

Antoine Pevsner - Sculptural Analysis 9" Generative Model (vertically stacked taskboard)

Antoine Pevsner – Sculptural Analysis
9″ Generative Model (vertically stacked taskboard)

Antoine Pevsner - Sculptural Analysis 9" Generative Model (vertically stacked taskboard)

Antoine Pevsner – Sculptural Analysis
9″ Generative Model (vertically stacked taskboard)

Antoine Pevsner - Sculptural Analysis 9" Generative Model (vertically stacked taskboard)

Antoine Pevsner – Sculptural Analysis
9″ Generative Model (vertically stacked taskboard)

Antoine Pevsner - Sculptural Analysis 9" Generative Model (vertically stacked taskboard)

Antoine Pevsner – Sculptural Analysis
9″ Generative Model (vertically stacked taskboard)

Moving Forward – Field Studies

Having now explored and ultimately created a basic ruleset/design mode to guide the exploration of our pavilion, we were asked to construct a portion of the pavilion using our current iteration of the 4″ x 4″ x 12″ joint. This portion had to create some form of “surface” – be it ground, roof, or wall – that ideally the rest of the pavilion would then extend from. I and my classmates found this to be increasingly difficult, as our smaller structure was just that – small, and the concept of transforming it to become a larger structure was very foreign. Ultimately the guidelines regarding the specific function of our new exploration (specifically one of the three necessary surfaces) were blurred/removed, allowing for a more open design process.

Looking back at some of the photographs I had taken of my smaller joint in various configurations formed the basis of my inspiration – I decided to quite literally stack my joint multiple times, creating a sort of smaller module which was then repeated over the course of the structure. The idea was simply to use one basic form of module (derived near-directly from the original joint), which was then slightly transformed to create eight basic modules. These parts were then stacked in various configurations to allow for openings and multiple surfaces.

Because of the construction of my original joint, I found it made the most sense to take one half of the joint and split it down the middle, effectively deriving my basic module from one quarter of the original joint. These quarters were then turned on their side, allowing the vertical nature to then become horizontal. This created various platforms across the surface as the modules were stacked and shifted. The exact function for each surface was unclear – I wanted their purpose to emerge naturally as they were used. Some surfaces were quite clearly stairs to upper levels, but all could then be used as seats or other forms beyond my limited imagining in the studio. I also decided to keep the idea of a “key” or anchor for the structure, creating three pillars that held the space together (although physically within the model they were not required to keep the rest of the parts together – but the idea was that perhaps they would).

Of course, the entire model was also to be constructed within Rhino, our primary modeling software. In anticipation of creating the actual model, I created some quick renderings of my design to get a feel of how it would ultimately look, some of which I have included below.

A sample of some of my renderings of the concept as designed in Rhino

A sample of some of my renderings of the concept as designed in Rhino

For the first time we were also introduced to the concept of space having a primary function – part of the requirements for our final pavilion was to ultimately create multiple spaces that could be used as a) public seating/conversation space, b) private space, and c) public demonstration/presentation space. While we weren’t expected to explicitly address these modes of space in this iteration of our development, they were something to keep in mind.

The final constructed model, built at 1″ = 1′ scale (so the model stood at 12″ tall and about 20″ long):

The completed field model. Like the original joint, I chose to construct the model out of a mixture of cold press and basswood, both because of ease of cutting/assembly and also to maintain a sense of continuity between the two models.

The completed field model. Like the original joint, I chose to construct the model out of a mixture of cold press and basswood, both because of ease of cutting/assembly and also to maintain a sense of continuity between the two models.

Like the original joint, the field model could also be disassembled, although unfortunately not as much as before - this time, only the wooden "keys" or anchors were removable, but the rest of the structure was stationary.

Like the original joint, the field model could also be disassembled, although unfortunately not as much as before – this time, only the wooden “keys” or anchors were removable, but the rest of the structure was stationary.

A view from the side of the field model/surface, to help illustrate how I shifted the alignment of the modules to create a more broken surface on the interior.

A view from the side of the field model/surface, to help illustrate how I shifted the alignment of the modules to create a more broken surface on the interior.

The exterior of the field model/surface. I chose to cover up various portions of the surface in an attempt to make the exterior feel more closed, to contrast with the broken/more open interior.

The exterior of the field model/surface. I chose to cover up various portions of the surface in an attempt to make the exterior feel more closed, to contrast with the broken/more open interior.