Solid Network explores a novel fabrication system for producing an architecture with porous, lattice-like partitions. The system presents an alternative to the use of solid walls with punctured openings as well as modern space-frame structures, such as those by Buckminster Fuller (1895-1983) or Konrad Wachsmann (1901-1980). The proposed system can produce structures with varied porosity, ranging from almost solid sponge-like morphologies to slender frame structures. This tectonic concept is realised with a novel fabrication technology for casting non-standard geometries. The invention and development of this technology is the core achievement of the thesis research.
The experimental research results are demonstrated architecturally in a case study proposal for a pressurised dome which, in typological terms, refers to the tradition of Fuller’s geodesics or Nicholas Grimshaw’s Eden Project (2001). The dome is situated on the north-eastern bank of the Main river in Frankfurt. The proposal exhibits novel qualities beyond a minimal shell by using the structure and the affordance that it offers to accommodate additional programmes and supplement the internal landscape of the dome that overall serves recreational activities.
The research aimed at designing a universal apparatus for generating a variety of non-repetitive, three-dimensional spatial patterns that in turn would serve as the structural schema for producing sets of derivative surfaces. The apparatus comprised of a parametric interface which allowed control over various sets of spatially distributed coordinates and their interstitial distances within a three-dimensional design space. Based on these variables, iso-surfaces were constructed as closed meshes using the Marching Cubes Algorithm (Paul Bourke). The resulting digital simulation of a pseudo-solid body served as the basis for the design of the architectural prototype before proceeding with experiments in fabrication and physical prototyping.
Since the introduction of digital design and fabrication technologies, efforts have been made to use flexible and re-usable moulding systems, such as CNC milled or fabric formwork, for the production of novel architectural forms. With reference to non-standard, pre-cast joints, this experimental research presents a different approach to the problem of form generation. Using meltable materials, such as ice, wax and gelatine, the work explored a thermodynamic moulding procedure by heating up nichrome wires embedded in the mould material.
Schematically, the set-up for the casting process involved a network of nichrome wires that were strung out between the walls in a sealed and transparent tub. The wires were fixed to the walls of the tub, and their fastening included valves for evacuating liquid mould material from the tub. Each wire could be heated through their connections to an electric power supply. Before heating, the tub was filled with one of the liquid moulding materials. Once the material was cured to a solid block, the wires were sequentially heated so that the molten material could run out of the tub via the open valves. This produced a network of tunnelled cavities. Subsequently, the valves were closed and Ultra High Performance Concrete (UHPC) was poured into the cavities. When the concrete set, the cast parts could be excavated from the mould while the wires stayed, and the mould material could be recycled.
The research was successful in coupling the computational procedure for producing iso-surfaces with a novel moulding and casting technique. Hence, beyond the innovative aspect of moulding and casting, the thesis research integrated digital processes for modelling and manufacturing in an unconventional manner. There was a fair correspondence between the digital simulations using the given algorithm and the physical prototypes. The technology was refined and tested at different scales, producing a serial range of outcomes. At each iteration the digital methods were calibrated to better match the cast prototypes. The cast parts could be used either as complex, non-standard joints or as elements of larger, non-standard, networked solids.
The fabrication methods presents a synthetic form generation process - regardless of the fact that the Solid Network produced by-and-large “natural” looking shapes. In other words, the process presents an alternative to research based on natural formation processes. In the procedure herein, the computational processes directly informs the outcome on a fundamental, procedural level. The final building prototype inherits qualities from these processes that go beyond the precise, numerical basis of digital form production. These qualities are the result of thermodynamical energies, their flows and dissipation in the spatial medium of the mould.