3D Printing in AEC: 3D Printing in Cantilever Design

Today’s guest blog is from Peter Macapia, Adjunct Assistant Professor of Architecture at SCI-Arc and the Pratt Institute - http://www.petermacapia.com/

Materials Lab: Cantilever, Fall 2010
Peter Macapia and Tom Wiscombe, SCI-Arc (The Southern California Institute of Architecture: 960 East 3rd Street, Los Angeles, CA 90013)

The cantilever is one of the classical elements of engineering investigation. But rarely does its study enter into the design process as such. So what if we were to take the premise of the cantilever to design an architectural situation? In this course, Peter Macapia and Tom Wiscombe worked with a group of students at SCI-Arc to develop an architectural project using advanced computational tools and material testing in order to arrive at a project that would transform the traditional linear relationships between engineering and architecture. The goal was to develop a massive cantilever that sprung out of the entrance of SCI-Arc serving a number of spatial and programmatic functions. Using a structural growth algorithm provided by SolidThinking called Morphogenesis; we developed a diagrammatic study of potential structures. We then transformed the topological behavior of the diagram into another system. A morphogenetic process is iterative in the sense that multiple forces are repeatedly tested against the general form until an optimized structure is produced. Materially, the challenge was to convert the notion of a solid form into a surface, folding into the structural system, a system of enclosure or envelope.

Eventually we began to see a pattern of networks emerge around the panelization of the form. It became clear that if we introduced seam logic into the panels and thus subdivided the surface of the form into another structural network then we could reintegrate the original topological optimization into a new material and spatial logic. Because we were able to test this out in a scaled model with the help of Z Corporation 3D printed (ZPrinted) models, we could basically test how the pattern of panels could connect along seams that were now acting like a network of embedded structural ribs. They key difference is that the ribs were not supporting the surface; they were now an internal feature of it.