Parametric Building Skin_Final


Closeup
Panel Types2

Background

Throughout the last century, mass production has been the guiding rule of modernism. The premise has been that through repetition, we can cut building costs. Unfortunately, the result is often sterile building surfaces of constant regularity across every facade—regardless of extreme differences in both internal and external conditions. These monotonous buildings are not only displeasing to the eye, they are completely unresponsive to their own needs. However, with the aid of digital fabrication, complexity no longer equates to higher cost. Architects and designers alike are using parametric tools to organize the surplus of data available to us today and use it to efficiently manage complex forms and tasks. This allows for designers to fine-tune every portion and panel of a building in response to its exact conditions.

ScWe must look no further than the scales of a reptile to see this same principle in nature. Much like our skin, these rough external plates protect the body of snakes, lizards, crocodiles, turtles and more. Interestingly, the parameters (size, shape, flexibility, etc) of each scale change depending on its location on the body. For example, scales near the eyes of snakes actually protrude out to shade and protect the eye while scales that touch the ground are flatter and smoother to allow for easy sliding. I believe this system of geometric divergence for improved functionality is a perfect metaphor for how a building skins can be fine-tuned to meet the specific needs of their exact locations.

 

Concept

Data Set Elevations.pngUsing this metaphor as a guiding logic, I developed a skin for an existing building that responds to varying internal and external conditions at the location of each individual panel. Much like a reptile, the building envelope is systemically organized into a diagrid of diamond shaped “scales” or panels with a window offset inside of each one. The articulation of these windows and their outer frames caters to two data-sets simultaneously: internal program and solar radiation. The internal program of the building controls the size of each window within the frames. Additionally, the varying angle, location, and intensity of the sun’s rays (quantified by solar radiation) controls the formation of the frames that surround these windows. Therefore, the parameters of each panel will vary according to its location on the facade. For instance, frames on the south facade will have larger extrusions at the top, while panel frames on the east and west facades will have larger extrusions on the left and right sides. The result is a coherent and unified system that can be applied across differing facades in an extremely responsive way. In this system, functional and formal variation go together. The movement of the sun translates into a gradient transformation reflecting the panels evolving formations. Thus, the performance-driven shape of the parametrically generated panels becomes heightened into an artistic concept.

 

Design Process

GoalsTo achieve these goals, I knew I had to find the total radiation of all the windows using Ladybug and then somehow run Galapagos to find the optimal frame extrusions for shading them (but only in the summer). In order to reduce the number of different frame types, I divided each facade into 4 different regions and placed a panel and frame in the middle of each one. Next, I ran Galapagos to optimally form each frame to equally minimize solar radiation in the summer and maximize solar radiation in the winter. As a result, each facade has four potential frame types to use. I then used the same objectives in Galapagos again to optimally populate the facade with these frame types. The result is essentially a gradient of the 4 different frame types across each facade, for a total of 16 different frame types.

Each of these frames is then populated with one of 4 different in-fill panels. They range from completely transparent to completely opaque. The specific designation for each panel is based off of the internal program of the building. I assigned different levels of “porosity” (openness) to each programmatic space based on its use, need for daylight, and desired level of privacy. These values were then plugged into an Image Sampler in Grasshopper, which ultimately produced the resulting pattern of panels.

Panel Types

Control Points

Test Point Locations

DtsWall SectionExploded Assembly2

 

Fabrication

IMG_2767The final step was fabricating the system at different scales. One of which was at a building scale where the patterns could be seen and evaluated across each facade. This presented several problems as the only logical method for re-creating all of these unique shapes was by 3D printing them, and there are many limitations with that technology. Finding an appropriate scale that was both economical in size, yet big enough to print the fine detail was extremely challenging. Moreover, the medium resolution and limited bed sizes of the 3D printers here at UC weren’t going to cut it. The solution was to outsource the fabrication to a company that uses a high-resolution PolyJet printer with a 20″x16″x8″ bed. The other scale of fabrication was that of a 1′ to 1″ model. I selected three different frames from the south facade and had them milled out of high density foam. Next I laser cut the diagrid frame and varying panel types out of MDF to experiment with patterning combinations.

IMG_2798_editIMG_4028_edit2 IMG_4041_edit2 IMG_4022_edit1

 

Final Renderings

Night4

closeup 2

Atrium2_2

Closeup3

 

Comments are closed.