Tuned Surfaces

Role: Researcher
Research Advisers: Jim Moore, Alex Terzich, Marc Swackhamer
Program Coordinator: Renee Cheng

In contemporary architectural discourse, focus has expanded from “making form” to “finding form.” In form-finding, geometry grows out of a careful analysis of building program, user behavior and other “performative” standards such as sound. Architectural acoustics are an important characteristic in many of the spaces that we inhabit, but the tools and processes for designing acoustically-optimized spaces aren’t developed and documented as thoroughly as other dynamic natural phenomena like light. Phase one of the research involved a survey of published literature on architectural acoustics and interviews of the architectural and acoustical design team of the Ordway Center of the Performance Arts and Northrop Auditorium. Information from these sources was combined to evaluate the current state of practice for the use of computer simulations, the representation of acoustical data, and physical mock-ups in the design of acoustically driven surfaces. At the end of this phase, a workflow was proposed that architects could use to design a surface whose geometry is determined by its desired acoustic performance.

 



 The second phase of the research involved testing that process in the design of a diffusing side wall panel. Two issues were identified as the work progressed that necessitated a modification to the original workflow developed during the first phase of the research if this process was going to be applicable to architectural practice. These issues were; the need to design the manufacturing process for the panel in order to make the varied geometry more economical and the need to incorporate acoustical testing of the panel to verify the diffusive performance of the designed panels. The study concluded with a modified workflow that incorporated lessons learned through the prototyping process as well as a physical artifact that demonstrated what an acoustically optimized surface could look like.