Abhi Chauhan: ‘Testing the Machines of a Third Industrial Revolution’ 1:100 Site Section Model

6th Year MArch student Abhi Chauhan has recently completed several models as part of his Intimate Cities project. The earlier models were used to demonstrate initial concept ideas and helped influence design changes. In keeping with the subject matter of the proposed development Abhi has put heavy emphasis on digital manufacture.

Abhi gives us an over view of the project and how this model fit in to its development:

As part of the Intimate Cities Atelier this year we were concerned stalled construction sites in the city of Manchester. These sites are unique in that their infrastructural order has been partially installed and my primary aim is the reconnection of these sites back to the city context. Situated on the Potato Wharf stalled construction site, the final scheme looks at the idea of bringing around a Third Industrial Revolution, by looking at the research and testing of an advanced manufacturing technique (3d printing) and a new energy infrastructure, (hydrogen fuel cells).

 Realised as a masterplanning strategy the stalled concrete frame on the Potato Wharf site is used a ‘live’ test-bed for 3d printed architectural components, in addition to this the scheme engages with the redundantΒ  transport infrastructure bounding the site and reinstates the canal and rail network as a distribution matrix for the transport of raw material. A reconfigurable 3d printed public park defines the edges of the new site in the overall strategy.

VISUAL LIGHT

The renders depict how the main 3d printing manufacturing hall and hydrogen exchange will look. The 3d printing facility is concerned with the research, manufacture and testing of 3d printed architectural components and as such the construction and detail is oversized to deal with a variety of different scales present on this project.

This first conceptual model depicts the main processes occurring in the 3d printing facility and follows the life-cycle of a 3d printed architectural component from its raw powdered state – stored in a material archive; to the printer beds; then for reconfiguration in a graveyard of failed components; and ultimately to its reverse engineering back to its raw powdered state.

The main frame was laser cut from 6mm MDF and designed to slot together. After spray painting grey to depict a raw concrete surface a series of powder printed material stores were fixed in place. It was decided to 3d print these stores due to their complex shape and the desired ‘layered’ construction aesthetic I was after.

The main machines in the model have all been constructed from separated components each laser cut from 2 and 3mm clear acrylic.

The 3d printed architectural components created in the facility were depicted by themselves being 3d printed. These parts were modelled in 3ds max and made ‘watertight’ ultimately for 3d printing on the ABS printer. (Abhi Chauhan May 2014)

Abi (2) Abi (5)Abi (6)

Abi (7)

One aspect of Abhi’s model work which is particularly successful is the appropriate use of different process. Having an understanding of the best suited method to achieve a desired outcome is key to an effective model. Without a clear aim as to what it is you are trying to convey many models have little practical use in conveying the key aspects of a design concept. This model of course naturally lends itself to 3d printing due to the subject matter.

IMG_0740 IMG_0744 IMG_0750 IMG_0752 IMG_0753 IMG_0757 IMG_0758