Arch 5650 - Graduate Studio - Professor David Newton- GDII

Collaborative design between Aaron Wilson and Tim Ogren

 

Introduction

The beginning of an inductive design process is reliant on extensive experimentation in a search for a component that's provocative in form and material qualities. In searching for this component we went through a bottom up design process where seemingly insignificant decisions on the individual and local component scale created a large effect on the regional panels and global design. The final design is understood on the global scale, but through investigation and understanding of the individual parts their influence on the whole is understood.

The process began with a series of iterations in the search of the starting point that would lead a scientific process of material and formal discovery. One week into the process desks were covered with various tests, drawing, photos and ideas. Material pieces of various sizes and shapes ranging from metal to plastic to paper were scattered all about. Ssomewhere buried in the iterations was an understanding about the limits and properties of materials, the adjustable characteristics of material and fastener parameters, and the limits of the patience of those involved.

 
 
     
 
 

Process

Before arriving at the component that would define our assembly, we went through a considerable analysis of the Mobius strip. The Mobius strip was provocative shape that attracted attention because of the possibilities of connections on the various angles and the reflective light qualities it produced in the tested materials. The strength of the unique form was eventually overshadowed by the difficulty of its manipulation as a component and its time consuming proliferation into an assembly. It did, however, project qualities that were successful, which were abstracted into a simpler component. This led to a series of cones that comprised the shape of flower. From that point, parameters were introduced in the form of twisting force and distortion that would lead to three components with features that created various conditions of lighting, airflow, and porosity. This fed the design process and it quickly began morphing into a larger system that was applied to a global design of a pavilion for an art gallery. The installation proved to be a valuable final stage of development because construction at life scale revealed how stacking the system and forming spaces would be possible in larger installations.

  
 
 

Conclusions

Throughout our tests and trials the design was infatuated with flexible forms. The creations would flop, fall over, and wiggle uncontrollably. Persistence, however, paid off as eventually the knowledge base grew and the design was able to take on a form, assembly, and scale that created an installation that used flexible forms and glowing, diffused lighting to draw attention and open minds to embracing flexibility in construction applications rather then fighting it to make a rigid form.

Our trials explored issues such as airflow, light diffusion, porosity, direct sunlight diffusion, movement, and form. In the end our design process began to emerge along with the design and we began considering every move on the individual, local, regional, and global scales. The lessons learned from the hands on process that follows led to a provocative design that could have only been developed through this form of thinking and exploration. Reverb uses the intense investigation and material understanding that transformed our own visions of what a wall assembly is.

 
 
Defining A Local Component

The local component serves as the congregation of individual moves and the beginning to the understanding of variable pieces. The seemingly small decisions made on the individual level have their differences multiplied in later steps, creating the final form of Reverb. The change of just a couple of hole placements in the individual creates the distortion of the cups seen in local component B and C which defines not just the way the single component looks, but the connections between components, the resulting formal geometry, and the aesthetic light qualities that varies between each of the pieces. While the individual and the congregation forming the local component may feel like it is making only a small variance from the neighboring community, the results of all the small variances is an unforeseeable form with qualities whose value can not yet be understood on the level of the individual and local congregation.
 
 

Products of Distortion

Each cup type has a different hole placement taking advantage of the material’s flexibility, allowing distortion in the formation of the circles (local component). Component B is formed by the top circle of the cups pushing out, forcing the bottom in, and creating a much lower edge angle than seen in component A. Component C does the opposite, bringing the tops together, and creating a more vertical edge. This variation allows for the plan and section in the global geometry to have varying curves in the formation of wall planes and the resulting spaces.

The flexibility of the PETG material will manifest itself on the global scale by varying the formal and aesthetic qualities of the spaces that are created. As the geometric and material qualities of the local component get multiplied out, the end structure and form will vary in ways that are not yet apparent on this scale. This creates unique qualities and variations in the regional components formed by these local assemplies.
 
 

Airflow Analysis

The formal variation between the components has the abiliy of a global assembly to affect the enjoyment of a space. One of the first things studied with the local assembly was how each could have the ability to affect the airflow of a larger piece. What kind of potentials could a system like this have for controling either interior airflow or wind resitence and funneling? By modeling the pieces in Solidworks the components were able to be analyzed and compared to one another. In all three case, as expected, a head on breeze would flow through the component with relative ease. The airflow through components B and C showed a slight increase in velocity as well, a resultent of funneling the same amount of air through smaller apperatures. As the components were rotated they showed a less expected result of having the ability to resist some breeze and instead push it around an overall geaometry. It became apparent that in a global assembly it would be possible to control some spaces and limit certain directions of airflow from entering an interior space.
 
 
 

Light Meter Analysis

By varying the form of components it’s possible to control the level of light transmitted, creating a variation between spaces. In this study, as expected, we found that the smaller the aperture sizes the less light that would be allowed to pass freely through the component. While at this level this variation between pieces was minimal, panels of these components would presumably created larger variations and have the ability to control direct sunlight and the effects of electric lights.

Using translucent PETG in local components allows light to transmit through the material and continually diffuse as the layers of material add up. This creates a glowing effect that will have potential in a global assembly, giving a different presence during the daytime and nighttime.

 
 
 
 

CNC Process

Through each stage of development for this project, efficiency in fabrication was an important goal. Using an anisotropic material, we were given the flexibility of orientating the individual cutouts as necessary on the page to achieve minimal unused area while not worrying about compromising the bending characteristics of the PETG.
One cone comprises 91.51 sq. in. The larger bed can nest 15 components in its 1,728 sq. in, while the smaller bed can nest only 4 components in its 512 sq. in. The resultant efficiency for the larger laser bed was 79%, and for the smaller laser bed 71%. The final decision, however, to fabricate with the smaller equipment didn’t come from analysis of efficiency in material waste, but in the efficiency of personal economics in relation to the costs of using the beds. For us to use the larger scale, professional service laser bed we would need to pay a labor fee of approximately $600. This fee was far too high for our expense account while the additional two sheets of PETG that needed to be purchased to make up for efficiency loss plus approximately $250 in laser time on the smaller bed were not of equal value to the $600 dollar fee. While efficiency would improve, saving material and money, on a larger laser bed, the labor costs compared to using our labor made our decision to use the smaller bed.
 

Rules of Assembly

As the design progressed from local to regional levels of assembly, a simple linear understanding of the connect angles and resultant geometry became necessary. A study of similar local components connected in rows and varying components revealed the linear geometric possibilities of the system. This provided a knowledge base for the design process to begin taking the local components and connecting them to produce meshes of regional panels.

The diagram on the bottom right defines the organization of type and orientation of the local component for regional panels. This is used to define mappings of various assemblies, keeping the meshed in the regional and global assemblies understandable. There is also a consistent connection strategy employed at the regional and global levels where each individual cone is connected to a neighbor through two fasteners on adjoining planes. Each local component therefore has eight connection planes and sixteen fasteners if it is surrounded on all sides. This creates strong connections between local components as they come together to form the regional panels.
 
 
 
   

Regional Panels: Set 3

As the three versions of the local component were analyzed for their strengths and weaknesses, we began to explore a system of assembly that would bring the local units together to from regional panels. The first step in this process was to use the linear connection study of local components to come up with a regional mesh. Many variations of meshes were tested, resulting in a variety of regional panel sizes, curvatures, and distortions. As a continuation of the studies of the local component, it made sense to studying the regional panels separately to understand the possibilities of form and light through panels that were exclusively A, B, or C local components.

We tested a variety of options for using a single component to make a nine component mesh, but found that the best barrel curvatures came when a panel of components are all facing the same direction except for the center component, which is flipped. The curve in all three regional panels was easy to work with, allowing for easy connection to other panels that minimizes the distortion necessary to connect the components of one panel to different components in the next panel.

 
 
 

As regional panels were created the porosity of each was one of the first variables tested. Each panel was viewed from straight on (red) and thirty degrees to each side (left is yellow and right is blue). These color codings were overlaid to explore the patterns created by each panel. In Set 3 it was common that the blue and yellow kept mostly to their own sides and didn't overlap. This study reveals how the panel will provide a changing view with each step as visitors pass by, creating opportunities to hide areas if the right combination of panels are assembled together.
 
   
   

Regional Panels: Set 4

While Set 3 was chosen for global assemblies because it creates regional panels with a barrel curve, Set 4 was selected because the panels it creates are nearly flat. Each local component has a little flex and bend to it, but because regional panels are assembled of like parts, by flipping every other panel backwards the components would theoretically fit together relatively easily and without distortion of the local components. It is clear from looking at the panels that they are made of three different local components, despite having the same mapping pattern and a similar panel shape and curvature. The porosity divides these three panels.

 
 
 
Set 4 is different than Set 3 or other test sets because from all three viewing angles apertures on the extents of the face provided view, creating panels that hold less of a flare for the dramatic and private. Comparisons between the regional meshes reveals a clear pattern of progression in the porosities; the regional mesh becomes less porous as the individual and local components become more deformed. This creates panels that can be used to provide less view and a similar geometric shape in the formation of a global assembly.
 

Regional Curvature: Set 3

The second parameter to the consider when comparing the regional panels is the sectional plane. How does the connection and resultant distortion of the local components cause Set 3 to curve? The first way this was analyzed was during the initial assembly process. As each local component was added to the mesh and fastened into place we could feel the components distorting very slightly to fit together. The hands on assembly of material to form local components into panels allows the assembler to see the distortion and transformation as each addition changes the system. As the regional panels were assembled a greater understanding of the strength and flexibility of each regional panel provided a knowledge base for construction of regional panels into global assembly.

The second way the curvature of each regional panel was understood was through digital analysis. Each panel was modeled in Rhino and the radius of curvature was color coded for easy reading and interpretation. This reveals exactly how and where in the panel the barrel curve was formed and where the holes were showing through and creating a quick, violent curve. It also revealed how the curve, which seemed slightly barreled in C was forming from the corner. It also revealed how the curve, which seemed slightly barreled in C was forming from the corner. While we were not able to deduce a way that this analysis could be used to adjust the individual or local components to control curvature, deeming this math outside of our realm to understand on this short time frame, it did reveal exactly where and in what direction the curvature was forming.

 

Regional Curvature: Set 4

The curvature in Set 4 proved to have a wider variety then was expected. As discussed previously with the mapping of Set 4, by flipping the orientation of every other panel, the angles were expected to offset and create nearly flat panels. This worked with regional panel A and to great degree in panel C ( As the photo shows the nine panel mesh creates a slight double curvature, but this is due to the edges of the panel being unconnected and therefore uncontrolled. In the global assembly this curvature disappears). Regional panel B, however, didn't follow this trend. Part of the reason is due to a similar effect as mentioned in panel C where the edges are uncontrolled, but it is also effected by the distortion in assembling the regional components. This gave this set a bit more variety to play with, providing some geometric variation rather then having three flat panels with only porosity influencing decisions over which to use in the global assembly.

The Gaussian curvature for Set 4 was an interesting study because the local component holes were much more apparent and their curvature allowed us to understand how the deformation was occurring. This revealed that our front oriented components were showing a positive Gaussian curvature, while our backwards were showing a very slight negative curvature in the regional panels shape.
 
   

Direct Sun

In studying the local components there was a noticible difference in the diffusion of light between components A, B, and C and whether they were oriented frontwards or backwards. As the regional panels were assembled and tested it was a priority that transmitance of sun would be studied. Understanding how regional panels behaved in sunlight has an impace on where panels are used to creat spaces in a global assembly. As the explorations in regional assemblies drew to a close our project received a site in the form of a competition entry. The competition was for a pavilion on the patio of a site in London, UK, which allowed our system to be studied using actual sun angles for a proposed project site. Angles studied were the maximum (61o ) and minimum (15o ) altitudes for the United Kingdom. The results allowed the design to now control diffussion of light or accept direct light depending on its placement and angle to south in the global assembly. The regional panels 3B, 3C, and 4B all allowed less than 18% direct light through in the combined two study angles. Using these three panels it would now be possible to create a surface structurte that would allow the minimum level of direct light through.
   
 

Global Design: Competition Entry

As Reverb progressed towards the global scale we were confronted with the possibility of entering a design competition. The competition called for the design of a pavilion for the temporary showing of artwork on the patio of The Light Box in London, UK. On the global scale the design for assembly went through a few transformations. It began by looking at the purely formal creation of an assembly that ran planes together to create reducing passages for art and interaction. After this the design started taking the direct sun studies and focused on creating spaces where there were space both out of sunlight created for the art. As the global assembly created these space it too a serpentine form move visitors around the assembly by drawing intrigue by creating spaces ranging from large gathering, public spaces to intimate spaces for an individual or pair. The final design uses this to create a structure that showes the ability of this system to create spaces and situations that would work for a variety of functions.
 
  In the global application of Reverb the design intentions were led by the sectional curve analysis and light analysis. The footprint is created using the geometric studies to begin defining spaces. As crevices are created they became the spaces where the sculptural pieces would be inserted. This creates a separation between the open traffic spaces and the crevice viewing areas. The crevices are the spaces where leaning and flex will occur in the system due to the shapes and connections ofthe regional panels. This creates space uninhabitable for traffic, but ideal for framing an area for sculptures and installation pieces. On flat sections of the wall are the spaces for fixed hanging artwork. With this system it is assumed the art will be attached to the regional panels so that it is able to adjust with the form of the wall as the weight, elements, and human influence cause the assembly to react.  
 
  The studies of porosity influenced the design as we considered the definite placement of the wall footprint and how a visitor would move around and be intrigued by the assembly. As they entered on the southwest side a porus wall greets the visitor, allowing a direct view through that gradually becomes obstructed towards the edges. As the visitor moves around to the north side the view opens up at different angles allowing an increased understanding of the space inside, creating intrigue as they move around the assembly.  
 
  If the visitor instead moves to the south of the assembly they find a narrower path leading them between the rock wall of the courtyard and the porous assembly. The path they follow would wind a bit as they round the most intimate of spaces, large enough for an art piece and one or two people. These spaces allow only a narrow, direct view in from any angle, raising intrigue while allowing visitors on the inside privacy. This is a time to enjoy the separation from others in an intimate space while being surrounded by a public place. In between this string of intimate spaces are unreachable zones, where the assembly’s boundaries lean together. These spaces are lit from the bottom with large flood lights, diffusing light as it moves away from the source, through the components, creating a glowing sensation at night. The assembly of reverb when applied to a public art gallery, provides the opportunity for discovery and intimacy as visitors move through.  
 

Installation

The final step in the development of Reverb was to build a portion of the global design as an installation to discover what we knew and what wasn’t considered. The installation focused on creating one of the intimate spaces with a light well. This allowed us to study the diffusion of light across the system, the amount of allowable sight, and the flex of the assembly. The light diffused as it had in many of the previous tests and created a glowing beacon that attracted as much attention as had been anticipated. The flex drew the same level of attention mainly because this isn’t normally a feature that one wants to achieve in our field but instead quite the opposite. The flex was more prevalent then we had anticipate but was also an inspiring feature as we considered the next steps of progress for this project. Because only a portion was built the structure leaned more then intended, but this can be attributed to the fact that these tight curves were designed to rely on the rest of the structure to provide the mass for stability. The installation increased our understanding of the system considerably and improved the ability to predict effects in section of any future application.

Conclusion

The successes of this half semester project was the ability to create a six foot high, flexible, light structure that stood temporarily on its own and afterwords with the help of an internal tensile tie. The lighting within Reverb is also a success, giving off the exoected glow and new shines as visitors moved around the installation. The semester provided valuable hands on experience about materials and fasteners and an understanding of a bottom up, growing design process. The process took a creative thought process that focused on a trial and error method of testing. If we continue work on this design, learning more about the flex and finding a more elegant way to deal with the leaning would be the first steps to take. Reverb opened our design approach up to a process of discovery, reactionary thinking and assembly. Continued work on this process could potentially changing the way pavilion or interior, non- structural design is theorized and applied.