537722 merrick morley airjournal

ADS: AIRJOURNAL

Merrick MorleyS e m e s t e r 1 2 0 1 3

CONTENTS

Merrick MorleyADS: AIRS e m e s t e r 1 2 0 1 3

4 -Introduction

8 - EOI: I CASE FOR INNOVATION10 - A.1 Architecture as Discourse16 - A.2 Computational Architecture20 - A.3 Parametric Modeling26 - A.4 Algorithmic Explortations28 - A.5 + A.6 Conclusion

32 - EOI: II DESIGN APPROACH34 - B.1 Design Focus36 - B.2 Case Study 1.046 - B.3 Case Study 2.054 - B.4 Technique Development58 - B.5 Technique Prototypes62 - B.7 Algorithmic Sketches66 - B.6 Technique Proposal68 - B.8 Learning Objectives and Outcomes

70 - PART C: PROJECT PROPOSAL76 - C.1 + C.2 Concept and Tectonics84 - C.3 Final Model92 - C.4 Algorithmic Sketches95 - C.5 Learning Objectives and Outcomes

22. West Gippsland.Squatter.

Hi, I’m Merrick. I used to hail from the ranges near Mount Baw Baw but one day I decided to come down to explore. So far the journey has taken me to Monash Caulfield where I studied a year of sculpture. Scared of the prospect in a career of fine arts I decided to leave: My love for sculpture explodes out of the studio and into the streets. This has lead me to The Bachelor of En-vironments at Melbourne where I happily design large sculptures. They are more satisfying and interactive then anything I could ever make as an artist. One day I hope to make someone happy as they interact with the spaces I made for them.

During 2011 I participated in the Virtual Envi-ronments subject which introduced Rhino as a source for design manufaction. Developed not just for architectural purposes, Rhino is a incred-ible piece of software that can be used for a breadth of design purposes. I appreciated this

early exposure to complex NURBS modelling as it has become a common train of thought within my design repertoire. From my experiences since V.E I have often been caught pondering the involvement that local buildings, shop fit-outs, jewelry, and sculpture have had with software such as Rhino. Some examples that come to mind are the recent Southern Cross Station re-design, arbitary web pages that contain complex geomet-ric patterning, and an automotive inspired shop front for Sportsgirl located in Chadstone Shop-ping Centre.

Although strictly the only subject that has fo-cused on Rhino, I have been able to implement the techniques used from this in my other design subjects. In Particular, my final design for a proj-ect in Designing Environments incorporated a lofted surface with geometries based upon point attractors to alter shadow and light character-istics during the day. Very primitive to say the least, this small experiment demonstrated how Rhino can be translated from the screen into a model across a different subject.

4

ENGENDER

The process of coming up with a concept for virtual environments was never really hard: I have found nature to be an easy topic from which to draw upon. Focusing on the communi-cation between honeybee’s, different mediums of model sketching, drawing, and rhino experimen-tation were all utilized to reach the final design. I looked at different ways of analyzing the com-munication: film stills, abstract diagrams, and researching the meaning of the dance. This lead to a design that focused on the intricacy of the dance, where the slightest change in angle of the bee in relation to the sun and its audience would alter the bee’s message. This process came to discover an undulating form that spread from one ear to the other, wrapping around the back of the head to symbolize the relationship between human and bee (which is mostly via hearing the bees quick movement of its wings and legs).

Learning how to construct, separate, and then reassemble the individual parts of the model was a challenging task and required perseverance and motivation. Reflecting on the final product I am quick to realize that it failed to acknowledge the materials it was made from: the card couldn’t form to the Rhino model I had developed and thus buckled, broke, flexed, and failed. Addition-ally the patterning I placed on the planar surfaces (circles with fluctuating radius’) was too large for the card strips to maintain structural integrity. If I were to do the subject again I would have paid closer attention to the physical restrictions of the materials available and propose a new method of construction. Furthermore, I would have made many more test models, which would have iden-tified construction issues earlier on in the design process. Overall however, this learning experi-ence provided the foundations of hard work and dedication that is still being built on today.

6

7

EOI: 1CASE FORINNOVAITON

This section of the journal identifies and commences the challenge of creating a design for the Wyndham Gate Design Project. Explicitly it will discuss architecture in 3 respects: architecture as discourse, computation in architecture, and parametric modeling. These three short discus-sions will form the overarching notions in how the design will develop and evolve during the course of the semester. In general, this case for innovation will highlight:

> The potential of architecture to contribute positively to the design of the Wyndham Gate Project

> The breadth of innovation and discovery that lies within computation in architecture

> How algorithmic processes and parametric design can bring exciting, brave designs that celebrate new typologies inside the architectural realm.

The journey of these 3 intertwined discussions will accumulate to a logical argument for the rest of the semester. This argument, based within reason and rationality, will form the basis of my own journey in computational design for the Wyndham Gate Design Project.

9

EOI: 1CASE FORINNOVAITON

LET STRUCTURE DO THETALKING

10

LET STRUCTURE DO THETALKING

‘Any serious “rethinking” of architecture at the start of this century cannot be undertaken without upsetting the structure and emphases of the traditional profession, of traditional typologies, and of traditional modes of envisaging the architectural subject.....’

Vidler, Anthony (2000). ‘Review of Rethinking Architecture and The Anaesthetics of Architecture by Neal Leach’, HarvardDesign Magazine, 11, pp. 1-4, p. 3

This semester will focus on ‘approach’ rather than ‘appropriateness’. Good architecture how-ever is never too far from either of these terms: An architect is trusted to create space in a man-ner that is contextualized and knowledgeable. But what the architect has not faced until now is a concept that asks to ignore all past preconcep-tions and demands to be heard. I am referring to the computer, an intelligent and burgeoning platform whose potential has not been reached yet. Architecture is once again changing form, although this time it is being pushed conceptu-ally by the architect’s ability to translate comput-er programming into design. The architectural subject is thought to be less about relevance in style and more about how it chooses to interact and speak to its environment. Computers help to offer advanced thinking and direction that has never been seen before.

Already discourses in the architectural realm have been large and diverse. ‘Blobs’ offer a new typology in considering, analyzing,

and designing architecture; Greg Lynn describes it as manifestation of the general and particular in a form that can’t be broken down into singular elements1. This contradictory summary is better explained when considering the architectural practice of Shoei Yoh. Using Pre-fabricated steel members and combined with site constructed bamboo, Yoh manages to articulate the sinuous roof shape into a feast of energy and architectur-al wealth to the patrons who inhabit it. Another project demonstrates the breadth of exploration within architecture. Aegis Hyposurface by Mark Goulthorpe searches for a way to dissolve the static form through external forces such as light, movement, and sound2. The result is a program of myriad triangular metal shingles attached to a flexible rubber membrane substrate and is dictated by mechanical pistons. Dynamic, unpre-dictable, impressive: This project challenges and asks: what is architecture? Although this semes-ter is not intended to explore such a question, the use of algorithmic modeling informs a genre of unprecedented form finding.

11

DRAGON SKIN PAVILLIONEmmi Keskisarja, Pekka Tynkkynen, Kristof Crolla (LEAD) and Sebastien Delagrange (LEAD)

New materials are exciting as, although they may have an ancestor to relate to, most of the time they present exciting new possibilities for exploration and development. This project fea-tured Grada Plywood, a nascent material that has the capabilities of being moulded with heat and without ruining the structure and beauty of the wood3. The architects, including students from Finland and later Hong Kong, used parametric modeling in relation to the Grada Plywood to create an undulating form. Prefabricated with machinery and bent with a custom-made heat press, the final pieces all differentiated from one another with algothrimic generation4.

What drew me to this project was the brutal simplicity in the form: there is nothing hidden nor is it pretentious in appearance. The material-ity is attractive to the eye and the overlapping of the members creates a sense of movement and tranquility. It feels happy and content in its environment, something that the architects clearly were searching for in their brief. This project is a pioneer, using the specific material and stretching its limits in usability, functionality, and assemblage techniques.

Although I could not develop such a project within ADS: Air, there are many things that can be taken and injected into the Wyndham Gate-way design. Balancing materials with form, pat-tern, and construction is something that is very successful in the Dragon Skin Pavillion. Experi-menting with permitting light in and out of the shell is another aspect which makes this project sing; Diversity is a major influence in all of the algorithmic architectures to date.

In my knowledge of architecture history I have been exposed to the debates of morality in ar-chitecture: AWN Pugin, William Butterfield, and Norman Shaw are 3 architects who, in the 19th century, advocated structure being bare, uncov-ered, and celebrated. It is only recently when the built environment has had to include more and more into its system (think of piping, air-condi-tioning, technological advances) that structure is forced to hide behind walls and suspended ceilings. The Dragon Skin Pavilion, although admittedly not a functional building for commer-cial purposes, continues a line of debate in which architecture should celebrate the structure that keeps it upright. Could this Grada Plywood be used on a much larger scale for office structures? Possibly, only time will tell…

12

‘Architectural features are continually tacked on buildings which they have no connection, merely for the sake of what

is termed effect; and ornaments are actually constructed, instead of forming the decoration of construction, to which

in good taste they should always be subservient’

Pugin, A.W.N (1836). ‘Contrasts and True Principles of Pointed or Christian Architecture’ Spire Books; Facsimile of 1841 ed. p.3

SUNRISE TOWERZaha hadid architectsmalaysia

What I have noticed from looking at multi-story construction is that there is always an indica-tion of where the floor level changes from one to the next. With Zaha Hadid’s design this concept is eliminated and made way for this webbing aesthetic that binds itself to the fabric of the facade. More like a futuristic building, the design appears to break Le Corbusier’s model of suspended floors on point loaded columns into a seamless display of changing floors and environ-ments. The exterior skin therefore is an indicator of what lays behind the glass and steel, revealing the anti-rigidness of floor plates and modular layouts.

When I think of established multi-story build-ings across the world, none appear to have a functional ‘partitioning system [that] becomes interchangeable, constructing a flexible vertical landscape capable of assimilating change over time’.5 The Sunrise tower claims to be able to alter its floor plates to adapt to any social/func-tional/aesthetic changes that occur

over time. This is intrinsically connected to the building envelope where components are varied depending on orientation, program, and struc-tural load.6

Hadid has clearly taken advantage of parametric modeling and experimented with its adaptabil-ity to the largest forms of architecture. She has managed to calculate an innovative network of geometrics that act both functionally to provide for all the needs of the buildings users and to awe the general public of Maylasia. What most impresses me is the delicacy of puncturing two holes through the shaft of the tower and manag-ing to solve the geometric problems for this. It is truly amazing to imagine that architecture and engineering can come together to create such a project.

However, my point of criticism in this project is that it intends to ‘[blur] the boundaries between building and landscape’.7 One glance at the CGI’s and any one can confirm that the structure has a rigid point of departure from its landscape. It thrusts upwards into the sky away from the Kuala Lumpur streets and celebrates itself, not the city in which it lives. I would agree with this statement if the building were placed in a high-density landscape such as Hong Kong, where the landscape has largely been adapted to multi-story construction.

‘Programmatic synergies are created by blending common programmatic aspects – creating a system that sepa-rates public from private, yet allows seamless transitions between environments’-Zaha Hadid Architects

15

COMPUTATIONIN ARCHITECTURE

COMPUTATIONIN ARCHITECTURE

A Grasshopper model devel-oped by Spanish architect Angel Lineras. The algorithm, generated by extracting data values from images of Mars, creates an accurate model of the planet’s topology. How could this be used by designers across the world?

‘If nothing else, eventually the sheer number of digitally-produced projects will bring about a new way of thinking about architecture and its proper place within the building industry’

Yehuda E. Kalay, Architecture’s New Media : Principles, Theories, and Methods of Computer-Aided Design (Cambridge, Mass.: MIT Press, 2004), p. 26

When I read the preceding quote again and again in my head I think about the colloquial story of The Ugly Duckling. The young duck, considered not as attractive (mind you, it depends on your personal taste) as its fellow fowl, begins life as a victim of rejection and judgment. As we all know, this duck grows up to be a beautiful swan that shadows over its fellow brethren, to their disgust. If we think about computation in this same essence it’s no surprise that innovative, algorithmically generated structures can divide humans. It is the ones (in the little ducklings case it was its mother) that see past the aesthetics and look at the inner beauty, intelligence, and potential of the given architectural program. It is the less accepting ones that hang their heads that I worry about. To be closed minded and pes-simistic is dangerously daft in the 21st century.

Personal memories of digitally driven architec-ture like Federation Square, the Southern Cross Station redevelopment, and the Guggenheim Museum in Bilbao all are controversial projects for their appearance. I have family and friends who object to this style, condemning its alien appearance and unusual characteristics. I recall one relative saying, ‘I just don’t feel it’. I imagine avant-garde artists and architects like Le Corbus-ier, Pablo Picasso, Mies Van Der Rohe, Vincent Van Gogh, and Louis Sullivan have, at one point or another, been subjected to such idle remarks.

Their works are exemplified as challenging the perception of art and architecture, but this took time, just as it will for the burgeoning computa-tion genre of construction. These are the ugly ducklings I am referring to; initially put into the spotlight for being audacious and experimental, It is only after an extended period of time when they are recognized for the influence they have had on newer generations. It is not just the critics but additionally the consultants and contractors who largely oppose to the use of computation in architecture. Who could blame them? They genu-inely present an unfamiliar scenario, with the possibility of ‘unmanageable complexities’ in the assembly of such innovative design.8 However, these designs do not come without precedence to the constraints of the real world: architects have a degree of intuitive feeling to what can be poten-tially created. There is a difference between delu-sion and vision and I feel a lot of the construction industry confuse the two with computation in architecture.

Moving the discussion along, it is time to think about the importance of computation in archi-tecture: why it is necessary and why it enables smarter, more efficient architecture. To deduce it to a nutshell, computers have progressed ar-chitecture in an exponential fashion. Aiding the modern day architect in creating construction drawings, communicating between client and

18

consultant, and computerizing design to create faster solutions: computers have been a great tool to use. However, like John H. Frazer points out, many are still to adapt to the computer phenomenon because it is only a tool and doesn’t increase design capability.9

The idea that a computer is a tool is completely true, but where would a builder be without their hammer, or a painter without their brush? Like these cases, the tool helps to enhance the design capability of the user. The computer itself has therefore become the architects’ tool, and cre-ativity can be generated due to state-of-the-art software within it. To suggest that the computer is manufacturing design is both naïve and igno-rant: A computer cannot work itself...

‘Computers, by their nature, are su-perb analytical engines. If correctly programmed, they can follow a line of reasoning to its logical conclusion…But while they can follow instructions precisely and faultlessly, computers are totally incapable of making up new instructions: they lack any creative abilities or intuition’10

-Kalay E. Yahuda

Computers have excelled at such a fast pace they are now being used by architects at the begin-ning of the design process, even before any sketch designs have been made. This is due to computation: using the power of the computers analyzing skills to produce design. All they do is in fact follow instructions, guided by parameters set by the software designer. In the case of Rhino, AutoCad, Sketchup, and Adobe design programs, the design tools imbedded into the interface are what architects use to create their design solu-tions. In other programs like Grasshopper, It is up to the user to create the parameters to produce design. The difference between these two forms of design is that the latter requires in depth knowledge of how to connect parameters and the information required to do this.

As a Modelab Webinar explains, Grasshopper is built upon Inputs, processing, and outputs.11 Users more or less have complete control over the inputs and outputs, while the brains of the computer achieve the processing. The processing itself involves complex analysis of data, whether that be a geometric surface, a list of data, or understanding the relationship between many geometries. A human could achieve none of these processes in the time a computer can complete the task: that in essence is why computation is so vital in architectural practice today.

19

PARAMETRICMODELING

PARAMETRICMODELING

There have been many attempts to articulate what boundaries parametric modeling is re-stricted to. Patrik Schumacher prescribes it as a way of organizing complex life situations within the post-Fordist network society.12 Daniel Davis prefers to coin it as a group of expressions that function dependently, with independent vari-ables.13 Robert Woodbury by far offers the most vague explanation:

Personally, the challenge of deducing parametric modeling into a contemporary description seems perplexing. If we consider the phrase in two parts, ‘parametric’ and ‘modeling’, then it eases the pain of understanding what the scholars are discussing. Parameters have existed long before

computers were imagined: ancient architects would use reason to juggle material, tectonic, and cultural parameters to build structures. Modeling is a term as old as parametric as it is an ageless way sculptor and masons could represent space with tool and stone. Collectively, the phrase represents a system that is reliant on space and processes. The three explanations of the term are esoteric and isolating; I find it best to think that parametricism, in the realm of architecture, involves having a set domain of space interacting with a set number of actions.

I will not lie: I am skeptical of the importance that parametric modeling has within the ar-chitectural world. It will one day have its place safely within society, but it should not consume society in the way that terracotta roof tiles domi-nate Melbourne’s suburbia. It is both exciting and scary as it spreads and develops across the con-tinents. My main concern is efficiency of such a process in design: does it live up to the elaborate and spectacular form it creates? Could the time, effort, energy, material use, and space be better used for something else? Does parametricism only speak of architects trying to out design one another: who can come up with the most ground-breaking, cryptic parametric model?

‘Parametric modeling (also known as constraint modeling) introduces a fundamental change: “ marks” , that is, parts of a design, relate and change together in a coordinated way. No longer must designers simply add and erase. They now add, erase, relate and repair.’14

-Robert Woodbury

22

Furthermore, the process of creating the para-metric model is difficult and individualistic: changing one process in a complex model may require the additional work of altering many more processes which, in turn, makes the model difficult to use.15 Daniel Davis highlighted his four points of major concern in parametric mod-eling techniques…

>Front-loading, where the design needs to be carefully designed as opposed to spontaneously made with intuitive action

>Difficulty in altering complex models after its inception (similar to Woodbury’s point)

>Confusion when sharing work with another (the models are extremely hard to follow and understand if you didn’t create it yourself)

>Even when the model is working, changing the variables may have little effect on the model in question.16

This bleak picture of parametricism is biased without considering its positive contributions to society. The most intriguing part of this style is the ease in which complex curves and surfaces

are created: amateurs now produce intricate curvilinear forms that couldn’t be imagined decades ago.17 Constructed pieces of parametric geometries are beautiful, sublime, inspiring. As a student there is nothing more motivating than looking at a piece of construction like the South-ern Cross Station Redevelopment: this is why one enters the game of architecture. Innovation in the construction industry creates vision and direction. It also sparks communities and imbeds energy that otherwise wouldn’t have existed. However this is reliant on the community em-bracing the innovation, which doesn’t necessarily happen...

If considered correctly, it presents the endless possibility of effective use of material, structure, geometry, tectonics, and design. If brought into the wrong hands, it can deliver ill-informed, shal-low, self-indulgent architecture that only speaks of itself.

23

Engineering and Architecture are two worlds that must co-exist in the construction industry: In many ways they are the ying and yang of the built environment. Since the inception of the modern engineers role since the Industrial Revolution, it has been a tug of war between the two parties to reach a realistic design that is both structually based and tectonically sound. Recently the two professions engage one another earlier in the design process and together con-sider possible solutions to the given brief. What this achieves is a building that is more success-fully resolved: ‘The Birds Nest’ by Herzog & De Meuron in conjunction with ARUP engineers is a test to this.

The synthesis of structure, function, form, and pleasure disguises this national stadium as a large sculptural monument, establishing a truely inspiring piece of construction. The complex intricacy of the steel facade smoothly wraps around to create the exterior envelope of the design. With only movement joints positioned to allow material expansion and contraction, the design team created the steel shell to be com-pletely connected and static.18 This is the perfect example of dissolving the recent seperation of engineers and architects roles in structure and tectonics.

Parametrically, the entangled grid, which creates the stadium facade, is composed of structurally acting girders juxtaposed with intuitively placed steel members to create the effect of wooden twigs. The 24 girders, composed to create wedge voids (diagram below) enable the large hole in the stadiums roof to be unsupported by ad-ditional columns.19 However, this would not be achievable conventionally without the aid of steel and ETFE membrane panels to support such an innovative structure.20

All these processes were brought together with parametric programs and BIM softwares. Without either of the two there would be no conventional way to produce such stadium. It is inspiring to think of the work that Herzog and De Meuron would have put into creating such a wonderful design.

24

NATIONAL STADIUM OF BEIJINGHerzog & De MeuronOve Arup & Partners

Structure is the basic requirement for any design that is to be built. Because of that, most architects acknowledge that it has in some sense to be confronted, incorporated, and, quite possibly, expressed. In those terms, an architectural design that does not address structure (and there are many such) is incomplete or illogical.Saint, Andrew., Architect and Engineer: A study in Sibling Rivalry, New Haven and London: Yale University Press, 2007. pp 2

25

Here point attractors have been implemented with geometric meshes to create these fluid patterns. I am curious to see if this line of thought could be addressed when considering struc-tual efficiency in the design. Could Kangaroo help to identify this?

Voronoi meshes have been applied to a distribution of points addressed by mathe-matical equations of sin and cosine. The results are concentric pat-terns that are beautiful. Again, can Kangaroo be of assistance here?

26

GRASSHOPPER,LUNCHBOX, AND KANGAROO

Like a young child placed at the deep end of an Olympic sized swimming pool, this semester will feature a constant struggle of confronting unfamiliar territories. With this in mind, I may have bitten off more than I can chew, challeng-ing myself to find a design that has structural efficiency through LunchboxTM and Kangaroo PhysicsTM tools.

Kangaroo manages to emulate characteristics of gravity and implement them into Grasshop-per. In turn, this can actually test the structural integrity of a design without engaging high-tech engineering software. Lunchbox, more of a tool then an analyzer, has created grids, trusses, and panels that are more difficult to create with the standard tools of grasshopper. With a simple surface (like the one shown below) I am able to apply a grid of geometries to re-create a surface. Although I was not able to show this in time, this structure can then be tested with Kangaroo to

measure its successfulness at remaining stable.

Collectively these two plugin features will offer a greater breadth of exploration than what grass-hopper already provides. For the remainder of the semester time will be spent experimenting with the implementation and characterization of grids and structures to formulate a design suit-able for the Wyndham Gate Design Project.

27

To gather all previous ideas to one focal point, this semester will embark on creating a design for the Wyndham Gate Design Project. Using the featured precedents, a design that is promi-nently based on structural performance and algorithmic generation will be developed. This innovative style of design is best suited to the brief, as it will:

> Create a structure that is eye-catching and exciting

> Enhance the physical environment

> Form a new icon for the Wyndham City and Greater Melbourne

> Generate and develop a new architectural discourse

The modern day architect uses, to the best of their ability, innovative materials and techniques offered by science and technology. The Wyndham project is no different to this and therefore a structure that incorporates computation and nascent materials will be designed. The public, ranging from Wyndham locals, to Melbournians, and even the world community, will look at this gateway and discover the possibilities that architecture can continually provide.

28

C O N C L U S I O N

29

Similar to before, this series of Grasshopper sketches addresses relationships between sin and tan, as well as List Culling and Voronoi Meshes. Even slightly changing the cull alters the geometries severly.

1 - Greg Lynn, (1998) ‘Why Tectonics is Square and Topology is Groovy’, in Fold, Bodies and Blobs: Collected Essays ed. by Greg Lynn (Bruxelles: La Lettre volée), p. 178

2 - Branko Kolarevic, (2003) ‘Architecture in the Digital Age: Design and Manufacturing’ (New York; London: Spon Press,), p. 51

3 - Emmi Keskisarja, Pekka Tynkkynen & LEAD, ‘Dragon Skin Pavilion’ / (10 Mar 2012). ArchDaily. <http://www.archdaily.com/215249> [Accessed 20 March 2013]

4 - Emmi Keskisarja, Pekka Tynkkynen & LEAD, ‘Dragon Skin Pavilion’ / (10 Mar 2012)

5 - Zaha Hadid Architects (2013), ‘Sunrise Tower’. < http://www.zaha-hadid.com/architecture/sunrise-tower/> [Date Accessed 20 March 2013]

6 - Zaha Hadid Architects (2013), ‘Sunrise Tower’.

7 - Zaha Hadid Architects (2013), ‘Sunrise Tower’.

8 - Branko Kolarevic, (2003) ‘Architecture in the Digital Age: Design and Manufacturing’ p. 29

9 - John H. Frazer, (2006) ‘The Generation of Virtual prototypes for performance optimization’, in GameSetAndMatch II: The Architecture Co-Laboratory on Computer Games, Advanced Geom-etries and Digital Technologies, ed. by Kas oosterhuis and Lukas Feireiss (rotterdam: episode publishers), pp. 208-212

10 - Yehuda E. Kalay, (2004) Architecture’s New Media : Principles, Theories, and Methods of Computer-Aided Design (Cambridge, Mass.: MIT Press), p. 20

11 - Ronnie Parsons & Gil Akos, (2012) ‘MODELAB - GRASSHOPPER WEBINAR’< http://lab.modecollective.nu/lab/introduction-to-grasshopper/> [Date Accessed 25 February 2013]

12 - Patrik Schumacher, (2010) ‘Let The Style Wars Begin’ < http://www.architectsjournal.co.uk/the-critics/patrik-schumacher-on-parametricism-let-the-style-wars-begin/5217211.> [Date Accessed 3 April 2013]

13 - Daniel Davis, (2013) ‘Lecture 3 - Parametric Modeling’ Powerpoint. < http://www.lms.un-imelb.edu.au> [Date Accessed 28 March 2013]

14 - Robert Woodbury (2010). ‘Elements of Parametric Design’ (London: Routledge) p. 2

30

15 - Robert Woodbury (2010). ‘Elements of Parametric Design’ p. 23

16 - Daniel Davis, (2013) ‘Lecture 3 - Parametric Modeling’ Powerpoint.

17 - Robert Woodbury (2010). ‘Elements of Parametric Design’ p. 39

18 - DETAIL Magazine, (2008). ”National Stadium in Beijing.” Issue Five, p. 484

19 - DETAIL Magazine, (2008). ”National Stadium in Beijing.” Issue Five, p. 492

20 - Jacques Herzog & Pierre De Meuron (2013), Official Website, < http://www.herzogdemeu-ron.com/index.html> [Date Accessed 28 March 2013]

31

PART B:DESIGNAPPROACH

PART B:DESIGNAPPROACH

DESIGNFOCUSWhen glancing across all the projects featured under Design Approach and Case Study 1.0 I am both in awe and excitement. To me structure and architecture, although it cannot be abused in the built environment, creates a landscape that could not be imagined otherwise. The Eiffel Tower, Sydney Harbor Bridge, The Crystal Pal-ace: these iconic and monumental pieces would not exist without one of the two professions. Too many of these designs and society would drown from toxic amounts of beauty and industrial elegance. The use of expressed structure has its place in the world and the Wyndham Gateway Design Project is can be one of these.

The Birds Nest, Canton Tower, and The British Museum Roof all posses an energy that can’t be generated without impressive engineering. In turn this activates particular spatial qualities

such as place-making, monumental sensitivities, and cultural richness. This could be achieved in a variety of ways with different architectural programs, but the elegance and completeness of expressed structure is something that cannot be denied.

Adolf Loos promoted the denial of decoration and 100 years after publishing Ornament and Crime these words still hold true. The saturation of décor through daily life needs to be counter-balanced with objects of beauty and simplicity; this could not be more truthful in a society that is constantly growing and becoming more connect-ed to the global market that yearns for a faster, more congested lifestyle. Structure potentially offers the beauty that other architecture achieves but, if considered intelligently, does it in a fashion that elevates it from the rest.

W H Y S H O U L D I T B E

S T RU C T U R E ?

34

DESIGNFOCUS The interior view of

the roof located at The Great Court in The Brit-ish Museum. Designed by Foster + Partners. this undulating struc-ture combines elegance with efficiency.

“I will not subscribe to the argument that ornament in-creases the pleasure of the life of a cultivated person, or the argument which covers itself with the words: “But if the ornament is beautiful! ...” To me, and to all the culti-

vated people, ornament does not increase the pleasures of life. If I want to eat a piece of gingerbread I will choose one that is completely plain and not a piece which rep-resents a baby in arms of a horserider, a piece which is

covered over and over with decoration. The man of the fifteenth century would not understand me. But modern people will. The supporter of ornament believes that the

urge for simplicity is equivalent to self-denial. No, dear professor from the College of Applied Arts, I am not deny-

ing myself ! To me, it tastes better this way.” Loos, Adolf, Ornament and Crime: Selected Essays, Ariadne Press,

Riverside, CA, 1997, p. 27

35

CASE STUDY 1.0VOUSSOIR CLOUD

CASE STUDY 1.0VOUSSOIR CLOUD

The definition of voussoir is specifically any group of truncated wedges that are shaped into a vault or arch1. Cloud, a word that is used frequently across many societies and languages, is therefore a contradicting term to pair with the previous term as it creates an oxymoron. Howev-er, the architects have created a condition where the two terms collide within a three-dimensional space and challenge the perception of its users and their 5 senses.

One can only imagine how it would feel to move through this installation with the varying levels of light, open space, protection, as well as touch and smell of the wood laminate wedges. The designers envisaged the construction to be experienced inside and from above, possibly to identify to the public how one should not assume that the outside language of the installation has a clear connection to the phenomenology of the inside2. When observing images from the walk-way above and another of a user interacting with the cloud, it is obvious IwamotoSCOTT wanted to create a feeling of surprise and confusion as well as efficient use of materiality and structural efficiency.

Essentially the Voussoir Cloud is an exploration into finding efficient form. This has involved

a series of processes that considered form, mate-rial efficiency, aesthetic appeal, fabrication, and assembly3. Computational modeling of catenaries was imperative to create the compressive vault structure. Delaunay tessellations, a fundamental type of triangulation for anyone’s computational geometry repertoire, were taken advantage of to create the structural efficiency needed for the compressive vaults and to vary the amount of porosity in the cloud4. I can only wonder how much of this was thought of by the Buro Happold engineers who were partners in the project.

This short introduction to Case Study 1.0 helps to understand the projects intentions and the most important features, both technically and con-ceptualy. The next group of pages features an at-tempt to emulate the project to create something unexpected and to analyse the work considering the multiple design spaces learn in EOI:1. Additionally I will consider how these new geom-etries could be used in the architectural realm.

I w a m o t o S C O T TB U RO H A P P O L DSouthern California Institute of Architecture, LOS ANGELES2008

‘Voussoir Cloud explores the structural paradigm of pure compression coupled with an ultra-light material system.’-IwamotoSCOTT ARCHITECTURE

37

T O P

P E R S P E C T I V E

F R O N T

S I D E

B A S E G E O M E T RY

A B C

Firstly it must be stated that the algorithm gen-erated for this project (as far as I am aware) does not correspond to project that IwamotoSCOTT and Buro Happold created. The given algorithm therefore is creatively acting in the same way that the installation would in the GH environ-ment.

Following the tutorials I decided that it would not be in my best interests to create the panels and tab system as it doesn’t concern my groups explicit design focus on structure. Although it is incredibly interesting to watch, the ideas pre-sented in it don’t have a place for my exploration this semester. The previous ideas of form genera-tion and kangaroo physics are points of depar-ture that I could use thoroughly for the groups final technique.

Mutations A, B, and C are not really mutations at all: they simply mimic what the Voussoir Cloud looks like. However what they do show is the importance of changing parameter values and the effects it places on the algorithm. A, B, C all have the same base geometry (planar surfaces formed by the lofted relationships between the top voronoi grid and the bottom offset of these initial grid boundaries) and anchor points for kangaroo to read. The only difference between all three is the stiffness of the mesh edges which, through the intelligent power of kangaroo phys-ics, changes the characteristics of the edges so they act as springs which all have unique tensile and compressive qualities. The lower the stiff-ness (which is easily changed with a number slider), the more the edges can expand according to the applied forces on it which in this case was positively in the Z direction. This one example, where only one parameter has changed amongst many, illustrates how easily it is to change form within the grasshopper environment without

U N D E R S T A N D I N G T H E V O U S S O I R C L O U D

38

B A S E G E O M E T RY

D E F

having large ramifications. When the stiffness was reduced to zero the lines would extend end-lessly in the Z direction and shows that there is a limit to the design of this one parameter. Unex-pected results can come from myriad situations so it is important to explore everything that the algorithm contains…

Exploring the algorithm involved analyzing which parameters effect the overall design of the cloud and secondly changing these give param-eters to their limit; the limit being when the model breaks (the computer cannot process

‘Search processes involve two steps: (1) producing candidate solutions for con-sideration, and (2) choosing the ‘right’ solution for further consideration and development.’ - Yehuda E. Kalay

the changes in the algorithm) or the model fails (it not longer functions with the given changes in the algorithm). In respect to Kalay’s reading this style of search processes is closely aligned to depth first, being that each parameter was expvlored thoroughly before another was exam-ined5. This explicit design space was efficient as no parameter was left unturned and each part of the Voussoir Cloud was altered to some extent bringing with it unexpected (and sometimes unexplainable) results. These unexplainable out-comes are due to my lack in GH knowledge and is something I knew would be expected. However, even architects like Daniel Davis acknowledge this is inevitable within the computational world of architecture.

The following diagram helps to highlight the differences between parameters that could be altered and others that were less susceptible to change.

39

1 - Input Curve/s2 - Input Point/s3 - Offset Distance of Voronoi 4 - Distance to Move Offset Shapes in Z Direction5 - Stiffness of Mesh Edges6 - Unary Force in Z Direction7 - Unary force in Y Direction - Output Mesh - Output Anchor Points

- Kangaroo Physics Plugin

V O U S S O I R C L O U D A L G O R I T H M

40

To elaborate, the parametric diagram identifies the strategies I used to extend and manipulate the Voussoir Cloud design. Each provided copi-ous amounts of leverage and creative explora-tion.

As mentioned before, mutations A, B, and C play with the stiffness of the mesh edges. D. E. and F are examples of limiting the offset distance of the voronoi grid and in turn making wider vaulted bases. As well as giving the design a bulkier weight, I also experimented in applying multiple unary forces in the X and Y directions. This transposed the design towards a new point of focus, almost as though it is being pushed or pulled in certain axis’s.

D, E, F, G, H, I, and J mutations all implemented new geometries (as opposed to a basic rectan-gle). These were merely arbitrary in choice and reflect how the cloud can change in form dra-matically. Iterations G and H explore an increase in vault bases and the effects it has on a) the amount of structure that is thrusted vertically and b) how the distance of the offset voronoi grid in the Z direction affects the proportions of the cloud. I think these mutations would feel isolating if experienced from inside.

I and J employ an oval form in plan but decrease in vault size and stiffness of springs. The result almost mimics some geometry of a camel. I can imagine J to be scaled to span across multiple road lanes and fulfill a brief like in ADS AIR. They do feel alien like and again isolating.

41

T O P

P E R S P E C T I V E

F R O N T

S I D E

B A S E G E O M E T RY

G H

42

B A S E G E O M E T RY

I J

43

The four following iterations were forms that in-tuitively felt successful in upsetting and creating a new language from the Voussoir Cloud dia-lect. Accompanying each is a short description identifying the key parameters affected from the original algorithm.

These new languages could be used for different purposes and spaces: what once seemed unsuit-able using the Voussoir Cloud tectonics could now be quite adaptable to another scenario. Al-though these are only for the purpose of advanc-ing my skills in GH, these algorithms could one day prove handy in solving a new design brief.

M U T A T I O N S O F T H E V O U S S O I R C L O U D

Limiting the anchor points to only the edges of the geometry enabled a more fluid, tactile impression of the Cloud. The bases have turned into a series of connected arches with pointed bases which I find pro-portionally pleasing to the rest of the model.

M U T A T I O N O N E

Elminating the anchor points of bases permitted the entire vault base to thrust into the positive Z direction, creating a very scupltural iteration

M U T A T I O N T W O

44

Initially appearing very differ-ent to the original form, this design has merely eliminated any distance vertically between the original voronoi grid and the internal offset version. Creating a lower stiffness of the mesh edges created a alien like design that is precedented by no other design I can think of. I could imgine this being de-signed at a monumental scale and used for exhibition spaces on the inside.

M U T A T I O N T H R E E

Similiar to the preceding design, this iteration decreases the stiffness value of the mesh edges even further and then al-most all of the exterior anchor points are removed. This has create tunnel like forms that allow movement underneath (as opposed to mutation three where the exterior was com-pletely closed off). Again the kangaroo physics has morphed the once stiff geometry into something that appears very fluid and translucent.

M U T A T I O N F O U R

45

CASE STUDY 2.0

KINGS CROSS STATION

CASE STUDY 2.0

KINGS CROSS STATION

Costing £550 million, the redevelopment of Kings Cross Station in Camden, London, reveals that sometimes architecture needs to be ex-pensive. The design intent of John McAslan + Partners was to transform not only the station configuration but also the wider infrastructure and commercial activities surrounding the 160 year old station. Coining the project as defined ‘layered interventions’ in the station landscape, the architects needed to bring life to the Grade I listed station through re-using, rebuilding and restoring the architecture6.

Although many sections were homogeneously designed by John McAslan + Partners since the late 1990’s, it is the western concourse which is of importance to ADS AIR. Located within the 150m span of open space the architects, in con-junction with Arup engineers, have created

a 20-metre high vaulted diagrid structure that wraps itself to the interior layer of the station7. Elegant, minimal, and expressive are words that can be closely associated to this project. I interpret this part of the station transformation poetically as the designer has tried to melt two distinctly different periods of architecture into a synthesized form. It is beautiful to compare to the contrasting spatial qualities of the existing brick façade and the new structure in front. It is neither domineering or quiet in its presence: The architects have been able to find a good hierarchy between the old and the new.

For such an ambitious project, innovation through engineering was required to create such a large span from steel without the need for internal columns. 50metre piles are driven into the ground to act as the footing system but are undetectable, creating a sense of illusion as the steel floats harmoniously above the hoards of people8.

J O H N M c A S L A N + PA RT N E R SCamden, London2012

‘Our ambitious transformation of the station creates a remarkable dialogue between Cubitt’s 1852 station and 21st-century architecture - a quantum shift in strategic infrastructure design in the UK’- John McAslan + Partners

48

K I N G S C R O S S P R O D U C T I O N

Plan view of the Concourse. The steel structure has a large presence on the entry to the station

49

50

Featured on the left is a series of final outcomes of reverse engineering the Kings Cross Station. They are the accumulation of firstly isolating the steel structure, then secondly trying to mimic its geometry. This was relatively easy as we had the use of lunchbox that can create structure and paneling grids very similar to the case study. However, as easy as it was to produce the itera-tions of the initial structure grid, the changes architecturally are extensive.

Although it initially seems like a basic construc-tion language, the grid used for the structure has many pleasing effects and positively impact on the station concourse. The simple diagrid structure with steel pipes running in perpen-dicular directions morphs continuously around the space. What appeared like a simple structure actually is very complex and any person with basic construction knowledge realizes this. The eye is drawn along these paths of steel before be-ing distracted by another, and so on and so forth. It is these paths of steel structure that cling to the interior surface of the building envelope that demonstrate its organic form.

The iterations of the case study help to demon-strate the fluidity of the case study 2.0. Em-ploying grids such as the hexagon, its Boolean

counterpart (reversing the structural placement) and the series of braced grids with steel mem-bers connecting all nodes created impressions that are static and confused. They may represent an organic form (the overall geometry) but the structure itself is overwhelming or uninteresting.

Following these initial iterations, it was decided to create a space truss system by offsetting the original structural grid. This granted further ex-ploration but mostly iterations that failed to cap-ture the essence of the Kings Cross concourse. As the construction holds the paneled roof system, this was somewhat mimicked but with arbitrary panels, including ones that rotated away exter-nally from the structure to panels that fit snugly in between the grid it sat on. The model that has been created fails to acknowledge the immediate surroundings of the design and its concourse, so therefore its impossible to really compare it to its real equivalent.

Overall these iterations revealed the importance in the specified grid employed John McAslan + Partners.

51

This secondary set of explorations look at apply-ing the tectonic language similar to The Kings Cross station to a geometry that has multiple points of curvature. Disregarding aspects that have been already covered in the previous page, what these iterations show is the adaptability of a technique to other organic forms.

This technique could be further explored with point attractors or the like that vary the length of structural members in relation to both curvature and its distance to the attractor. Also I would like to see the generated models brought to reality and to analyze how it would feel to move through, in a phenomenological sense.

52

53

TECHNIQUEDEVELOPMENT

1 2

5

3 4

STRUCTURE

MONUM

ENALITY

FLUIDITY

CONTRAST IN SPACE

FEASIBIL ITY

Following our mid-semester presentation it was obvious that our design lacked exploring explic-itly the Wyndham Gateway Design Project. In theory we had looked at establishing a technique that could be adapted to the site, but had failed to implement it to evaluate its effectiveness in achieving the design brief. Realizing this, as a group we went back to our technique develop-ment and acknowledged there was a lack in understanding of what our design wanted to be based on and how it is achieved.

Our 5 criteria focus on the crux of the design. From here was the generation of form and struc-ture that abided to these criteria: the GH algo-rithm would be constantly altered to cater for this. It had now become a systematic approach to design where each criterion would be analysed and design generated for it.

The following pages confirm the process is which we took: it is the end result of the sample development. Within the Rhino environment we all analysed our separate iterations and evalu-ated for its merit. The only criterion that has not been considered extensively within this stage is feasibility.

The overarching driver: feasibility is the sports team we all love to hate. During our mid-semes-ter presentation we were handed a harsh reality check that our design is economically out of reach for the given budget. Acknowledging this, either we must reconsider our materiality or scale of design. To change steel, in my view, is to change the entire ethos of the design intent and therefore I should start the whole project again. From here it will take the careful articulation of creating a design that is within a realistic margin to the budget.

T H E F I V E

C R I T E R I A

55

S T R U C T U R E

M O N U M E N -T A L I T Y

F L U I D I T Y

C O N T R A S T I N S P A C E

T O P56

57

TECHNIQUEPROTOTYPES

60

Extensively our preference in materiality has always been steel. Like in Case Study 2.0 it offers the chance for seamless, aesthetically appeal-ing connections. To reach a perfect transition from member to member, welders would work between each join; this increases cost and time.

Steel is a great structural material as it provides strength in both compressive and tensile forces. It also has a high weight to strength ratio when compared to other structural materials such as concrete. However its downfall comes with the high cost and off site manufaction that means if a member is incorrect then it must be reproduced in the factory or altered on site, which is very costly.

In relation to our technique and design brief, it is obvious that processes from case study 2.0 cannot be replicated. With a low budget ($280, 000) the prospect of welding any joints on site may come underdone. Additionally if the design imbeds over the roadway so it is preferable that the members come together quickly to avoid

lengthy delays in traffic. This means that the steel must be connected in a way that is generic, to avoid confusion on site, and quickly.

From research there is an abundance of connec-tions available for our technique including: Hinge and pin, cast, reinforced and hidden as well as lap and butt join connections9. From discussions it has been decided that hidden connections are the most suitable type for our design intention and the feasibility of it coming under budget. Hidden steel sleeves hide the bolted connections under-neath. To increase assembly time, we have also suggested that the nodes be welded offsite.

We have produced two 3D models that have helped to visualize the physical form of our de-sign technique. These models were made before the mid-semester presentation and don’t reflect our current design technique. However they consistently show how one design that is virtual does not correlate to its physical counterpart. In theory it is always good to produce physi-cal models to be able to evaluate design in a 3 dimensional space.

61

ALGORITHMICSKETCHES

CREATE CONVERT DEFINE ASSEMBLE ANALYZE VIEW

Although not currently involved in our design technique, we have been exploring methods of simulating structure. Initially this was through Kangaroo physics, which enabled our grids to interact with applied forces (I have shown how this would have worked with the Voussoir Cloud exploration in Case Study 1.0). However the downside to a plugin like this is that it doesn’t reference anything: the forces that are applied are arbitrary and do not correlate to physical forces in the real world.

Karamba is a Finite Element plug-in for GH and is more aligned to engineering analysis than Kangaroo is. From empirical understanding and reading of the user manual Karamba tries to accurately measure the compressive and tensile forces in a beam. It also quantifies the reaction forces of fixed points in the beam calculated from the forces applied to it. These types of results are topics I have covered in other subjects at The University of Melbourne but the ease of getting the results is dramatically quicker as the com-puter does all the number crunching for me.

Above is a simple illustration showing the steps one must take to reach a working Karamba

algorithm10. CREATE is simply manufacturing some type of geometry, whether that be one line or a program of mesh lines. In order for the plug-in to understand these geometries it is necessary to CONVERT them into beams or mesh shells. DE-FINE and ASSEMBLE involves adding loads and supports that are easily achieved with the right parameters. They are also controlled arbitrary by number sliders and offers the pleasure of chang-ing how much force is applied to the support or load points in the model.

Additionally in this step one is able to change the characteristics of the beams from CHS to RHS, as well as generic U beams and I beams. Karamba has the intelligence of knowing how the differ-ence in structural qualities between all of these beams types and therefore the applied loads interact differently with all of them. To finally ANALYZE the model a simple parameter is added to collect all the existing data and to make it function as a cohesive unit. VIEW is added at the end of the process as a final parameter is added to reveal all the information and results. Here it is easy to toggle what information you want to view.

U S I N G A S I M U L A T O R

K A R A M B A

63

1

2

3

4

5

6

7

1 - Start Point of Curve2 - End Point of Curve3 - Point Load Input4 - Input Suface5 - Mesh Size in U and V 6 - Gravity Load Input7 - Circular Cross Section Diameter’ - KARAMBA Parameters

Here is a more visual interpretation of how Karamba works. As it shows in the diagram, Karamba Plugins form the latter half of each algorithm definition. Although it seems like a logical progression, working the correct data into each parameter was a challenge in itself and therefore it took sometime to finally reach a working algorithm.

The difficulty lay in the ASSEMBLE stage of the definition where I found it particularly difficult to connect the cross section parameter or beam characteristics to the beam generation node. The solution was to connect a panel with the beam identity, a unique cross section that is stored within the Karamba parameter, so it knows which beam to use.

This introduction to such an impressive plugin has really shown the potential of GH and the interdisciplinary connections that can exist between engineering and architecture. Hopefully it can be used in the next stage of final design generation and be used to quantify the structural integrity of the gateway design.

Although it will not guarantee any kind of struc-tural upholding, at least we have a glimpse of what areas of a structure are put under compres-sive and tensile strain and to compare parts of the model with the listed forces that are given. Therefore if we were to consult a structural en-gineer, we could have a greater knowledge even before we stepped through the door.

S U P P O R T E D B E A M

M E S H N E T W O R K

64

1 - Start Point of Curve2 - End Point of Curve3 - Point Load Input4 - Input Suface5 - Mesh Size in U and V 6 - Gravity Load Input7 - Circular Cross Section Diameter’ - KARAMBA Parameters

A simply supported beam is fixed at one point and a force ap-plied at the opposite end. The resulting forces can then be drawn from the model to test whether it will stand in reality, or at least to argue it will.

A large mesh geometry has been put under a gravity load. This helps to show that even huge amounts of steel mem-bers can be tested with the KARAMBA plugin

A closer look at the forces occuring within the beam. The two points are anchored so therefore all forces from other regions are being transferred down to these points.

65

TECHNIQUE PROPOSALFollowing on from the technique development, key samples have been noted and from here they will be advanced further. What they each possess are the characteristics of what our group wants to achieve through the five criteria we have set.

To elaborate, the technique we have created esablishes contrasting spaces on the site: this currently is through a change in volume, scale, shape, and/or orientation along the design. This ties in with the concept of monumentality as the experiences of a design that develops, rotates, and changes as you experience it leaves a memo-rable impression on the users mind. Monumen-tality is achieved through scale, but not neces-sarily through mass: our intended use of steel won’t permit this within the budget. Structure and steel is something that we currently endorse within our technique proposal as it representsthe most expressive, abstract, raw, and genuine

grandeur of architecture. Steel is an effective ma-terial to use structurally and opens the door for myriad opportunities for plausible forms.

Fluidity from structure is a concept that stretches industrial techniques to its limits as the baroque architects did with their concrete. Creating a sense of the organic, free-flowing tendencies that are held so strongly to the Kings Cross concourse project is what is in mind. Replicating this on the Wyndham Gateway Design Project will celebrate not just the architecture but the way it encap-sulates the energy and spirit of the people it borders.

Collectively the criteria are gospel to reaching the final design: without any one of them the design will fail just as a GH definition does without implementing the correct parameter.

T H E E N D O F T H E

B E G I N N I N G

66

TECHNIQUE PROPOSAL

67

First and foremost this subject has been exhilarating. Although it is not at the conclusion of the semester yet, I am confident in my work and the way it has accumulated. In relation to the learning objectives set in the AIR reader, all 8 parts highlight the fundamental aspects of this studio. To be completely fair I would say that my learning of objective 4 is lacking currently: our group hasn’t had enough exposure of physical models to understand the spatial qualities of our current technique proposal. Similarly objective 8 is constrained due to my limited exposure to GH: overtime hopefully this personal repertoire of computational skills grows and I can become more suited to parametric modeling.

I find architecture fascinating: anything that is thrown at me is taken with both hands. However it has been naturally difficult to comprehend some works that I am exposed to and their concep-tual merit. Sometimes architects are so idiosyncratic with their approach and final designs that it seems to completely alienate the project from myself.

On the contrary, I have had experiences where I feel I can connect and understand a project but classmates find it convoluting or too abstract to assess. I guess this all draws back to the notion of subjectivity in architecture and that everyone in the community has the power to have their own opinion, approach, and ideology. I would like to believe that, regardless of what one person’s preference is to the next in architectural taste, holistically architects continually try to serve the people they are designing for and not for their own gain. If this is lost, essentially architecture is lost.

L E A R N I N G O B J E C T I V E S

A N D O U T C O M E S

68

1 - Dictionary.Com, LLC, (2013) < http://dictionary.reference.com/browse/voussoir> [Accessed 1 May 2013]

2 - IwamotoSCOTT Arhictects, (2008) ‘Voussoir Cloud’ < http://www.iwamotoscott.com> [Acessed 15 April 2013]

3 - IwamotoSCOTT Arhictects, (2008) ‘Voussoir Cloud’

4 - Chen, R., et al., (2010) ‘A spectral characterization of the Delaunay triangulation. Computer Aided Geometric Design’ < http://www.math.zju.edu.cn/ligangliu/Publications/Papers/2010_CAGD_spectral_paper.pdf> [Accessed 3 May 2013]

5 - Yehuda E. Kalay, (2004) Architecture’s New Media : Principles, Theories, and Methods of Computer-Aided Design (Cambridge, Mass.: MIT Press), p. 20

6 - John McAslan + Partners, (2012) ‘King’s Cross Station’ < http://www.mcaslan.co.uk/proj-ects/king-s-cross-station> [Accessed 1 May 2013]

7 - John McAslan + Partners, ‘King’s Cross Station’ / (21 Mar 2012). ArchDaily. < http://www.archdaily.com/219082> [Accessed 29 April 2013]

8 - John McAslan + Partners, ‘King’s Cross Station’ / (14 Mar 2012). The Guardian. < http://www.guardian.co.uk/business/2012/mar/14/five-year-redevelopment-kings-cross-station> [Accessed 29 April 2013]

9 - Steel Structures Education Foundation (SSEF) (2013), ‘Fun is in the Details: Innovation in Steel Connections’ < http://tboake.com/SSEF1/index.shtml> [Accessed 2 May 2013]

10 - KARAMBA3D (2013), ‘Karamba User Manual’ 4/04/13. < http://www.karamba3d.com/downloads/> [Downloaded 15 April 2013]

69

PART C:PROJECT

PROPOSAL

This project involved the expertise of The Green Oak Carpentry Co., who helped to develop a vi-able gridshell structure for the design. This is a great example of multidisciplinary trades coming together to create an innovative architectural typology.

But what is a gridshell? According to Celine Paoli ‘gridshells are basically shells where mate-rial has been removed to create a grid pattern’. Furthermore, these structures ‘can be assembled flat on the ground so as to form a two-dimen-sional articulated mat’ with the final form reached by deforming the mat and introducing anchor points on the ground floor plane1. Mark Cabrinha’s interpretation is similar but differs by stretching the definition to designs that don’t conform to ‘straight barrel vault’ forms that Paoli is focused on2. It is the combination of the wooden laths, the connections between them, and the anchor points that are all working

together or against each other to achieve a shell form. The form desired for the museum was a large open space without any internal partitions and a lot of natural light3. The Gridshell would be a suitable design solution as it enabled this and combined the structural language of the museum with the architectural paradigm.

The Weald and Downland Gridshell was low-ered into shape, rather than lifted from supports below, in a effort to reach to desired shape. The two lattice mats at lengths of 36metres were dropped at predetermined points each day, with the carpenters fixing the structure when suited. At the end of the construction only 145 of 10,000 scarf joints (joints to create longer timber laths) failed and needed to be rectified4.

the final form achieved 50metres in length, a varying width between 12.5 - 16metres, and a height nearly reaching 10 metres from the workshop floor. costing roughly £2 million this gridshell was a success of combining structure with function in a pragmatic way. It is important to highlight that the brief of the project called for a contextualized, materially appropriate design that functioned to the needs of the museum us-ers. What the end result achieved was something much more than that: an icon for Sussex.

E d w a r d C U L L I N A N A r c h i t e c t sB U RO H A P P O L DWeald and Downland Museum, Sussex, England1996 - 2002

‘The Gridshell was formed from a flat lattice of green oak the was lowered into a three-dimensional shape with the use of gravity.’-Edward CULLINAN Architects

73

The Timber Lattice was completely assembled as a 2D mat, resting on scaffolding, and ready for lowering.

Daily the scaffolding would be lowered to allow the timber laths to bend into place. This needed to be done over an extended period of time so the laths could rest each day before be-ing subjected to changes in compressive and tensile forces.

The final structure was fastened and structur-ally bolstered at the most critical points of pressure. The final form resembled the boolean union of three spherical shell forms.

74

Gridshells have only been realized in the past 50 years due to the hard work of Frei Otto and his team. Since this and only up until the previous ten years gridshells were sparse. However with the rise of computational processing in archi-tecture and engineering the challenge of making this structure has become a lot easier: gridshell.it is a testimony to this. Above are cropped shots of various existing gridshells across the world and demonstrate the scope this language can be applied to, including zoo pavilions, warehouses,

museums, shelters, and art galleries.

Receiving a large wake-up call from our mid semester crits, it was obvious we needed to search for new sources of inspiration for our de-sign. Gridshells were the perfect fit for what we wanted to achieve with our design criteria.Grasshopper has been a tool implemented in the form-finding process of newly created gridshells, so this too was a motivating reason to use it in the Wyndham Gateway Design Project.

75

CONCEPT &TECTONICS

CONCEPT &TECTONICS

77

Considering the site plans, given photos, .DWG files, and Wyndham design brief we imple-mented our structural technique to create the final form. On the previous page is a series of diagrams highlighting the fundamental aspects of the site our group thought of when thinking of the design.

The top diagram illustrates the changing thresholds of urban landscape and lifestyle in the site. Werribee, Geelong, and Melbourne: all different across urban scales but yet equally important to our interpretation of the site. The middle diagram simply reveals our intent for the architecture to speak and interact to all these communities as well as work with the topogra-phy. Site A was therefore the most appropriate location for our design and enabled an area that could be observed by anyone in close proximity. The two emotive sweeping curves on the third picture represent the speed at which users move through the site. It was our intention to cater for only these people and not for others at the petrol station close by. It cannot go understated just how much we intuitively considered the people in locomotive transport and the possibilities that could make exciting architecture.

This series of images above were the main steps we took to achieving the final form. An elongated volume was the starting point: it would stretch across site A to instill a longer impression and provide loftier changes in the tectonics. The curve would help to create adversity in the users vision of the architecture, as they would move through the space at fast speed and at the same time be exposed to ever -changing angles of the sculptural piece. Flattening the volume to a plane and rotating the ends gives the eye focus points and let the design lightly intermingle with the ground. Curving the plane perpendicular to its length would again add more levels of interpre-tation and contrasts in space during the users experience. Finally, stretching the design to form two giant cantilevered networks would be our main attraction as the multi-curved plane felt complete. Collectively these transformations were the end result of critical thinking and trial-ing of multiple design solutions.

The search for a form that I just explained was never without the guidance of structure behind it and helped to push the formal gestures. Together they were altered to imbed our design technique onto the site.

78

Traditional construction of this design would in-volve placing all the members flat, using scaffold-ing to lift the timber laths to position and then fastening them into place. Our anticlastic form couldn’t plausibly be built like this. The designs contact points to the ground are central, almost like an inverse form of a gridshell where they are always around the exterior. This makes our de-sign more of a reticulated surface than a gridshell but that didn’t matter to us as we are trying to stretch the use of this structural typology.

So then how would our project be feasible in the construction process? Like in the precedents that use scaffolding and cranes, our design would too incorporate these methods to raise (or lower) the timber into place. Additionally it has been suggested that hardwood be used for the tim-ber laths: our initial incorporation of laminated plywood seems to not be a structural solution. All these considerations would be further examined with engineers to discover the most useful wood to use for the project.

79

Top Flange

Diagonal

Perpendicular

Bottom Flange

Complete Structure

80

Inertia is described as the property of matter when it is in a state of rest along a straight line and is a very important in regards to designing a gridshell structure5. The Weald and Downland, Japan Pavilion, and Mannheim Multihalle all had multi-layered timber laths for two reasons: a) to combat the original calculated inertia with more slender laths that would bend easier and; b) to allow the straight laths to pass each other at nodal points without complex joinery problems6. The first reason can be further explained by the similarity the gridshells have with a conventional I-beam: the two thin flanges are connected with a web to combat shear stress. In the case of the gridshell two timber laths are placed over one another and are connected at the nodes and by

timber blocks along the shell, simultaneously al-lowing the right moment of inertia and to permit the timber to bend easier than one thick lath7.

The diagrams above helps to explain that the insertion of a 4-layered structure was not for aesthetic or intuitive reasons but because of the research into constructed gridshells. The top two diagrams illustrate the range of forces that are dealt with the differing layers of timber laths. The bottom image is most relevant to what has been discussed so far with the similarity to a steel beam. Once all the members are in place the design effectively combats diagonal stiffness, deformability of the cells, shear force, and bend-ing moments8.

T H E I M P O RTA N C E O F G R I D S H E L L P H Y S I C S

81

82

The erection process places a high reliance on the quality of the connections. During this stage unprecedented forces will inevitably be applied in all directions on the node, meaning that the connection has to allow for this. From all understanding the final connection we have designed will counteract such forces. Our initial connection, which was a mere bolt connecting all laths at the node, would deform with traditional construction10.

The connections at the node of the intersecting laths are vital to ensure structural stability and strength. With a solution that is influenced by Buro Happold and the work on the Weald and Downland Gridshell (image to the left), we adapt-ed the connection to integrate with our own timber structure. The main features of the con-nection are allowing each lath to rotate and slide at their own will but being forced together by the connection of the two exterior metal plates. The middle plate has a pin fixed through it creating the approximate point of the node and holds all the laths in the same point - so they don’t move around causing severe instability9!

The 1:10 joint model was a good way to explore this connection type and to further push our arguement for the gridshell solution. It felt as though this could really be created at a larger scale and be able to withstand multiple forces.

‘The connections must also allow the two layers to slide one on top of anoth-er since inner and outer layers do not have the same radius of curvature.’- Celine Paoli

83

FINAL MODEL

‘Architecture is really about well-being. I think that people want to feel good in a space... On the one hand it’s about shelter, but it’s also about pleasure.’-Zaha Hadid

86

North Elevation

East Elevation

90

South Elevation

West Elevation

91

ALGORITHMIC SKETCHES

Reference Curves and Surfaces

Top Flange

Perpendicular

Diagonal

Bottom Flange

The final algorithm generated for our project was broken down into 4 main parts – seen above in the diagram. It’s interesting to note how the final product, which is meant to appear fluid, clean, and austere, is made up of an extremely complex system of parameters. Tom Nelson, who is intro-duced in the learning objectives and outcomes, noted that if asked to produce the same work he would have come up with an alternative way to create the multi-layered structure. This only proves how this parametric modeling platform is completely open to varying ways of reaching the same goal in a variety of ways.

The reason the algorithm needed to be broken into the four clear segments was that each range of structural laths needed to be carefully altered. This includes the well-known processes of find-ing surface edges, orientations, moving and rotat-ing the lines, and making it adhere to the curva-ture of the original reference lines. After this was achieved the design was only an amalgamation of lines: there needed to be a profile added to give the design depth. A separate geometry was added which needed to be lofted and conform to the ever changing orientation of the laths.

93

1

2 3 4 5 6

7 8 9 10 11

1 - Ratio for number of Timber Laths2 - X value (no. of Timber Laths)3 - Input Curves4 - Rebuild Curve5 - Loft Curvea to Surface 6 - Offset Surface7 - Surface Edges8 - Flip Matrix9 - Surface Normal10 - Move Point along Plane11 - Move Geometry

Taken from Beginning of Top Flange section

Taken from Beginning of Algorithm

Like the above quote verifies, the algorithm that was produced for the project was hard to interpret, as I was not the one who had produced it. Albert constructed the large majority of it and with this came the empirical knowledge of really understanding how it operated. When I came along and tried to analyze it myself, there were points of it that were very confusing to recog-nize. Still some of the algorithm sections don’t make any sense to me and tht reflects my limited experience with Grasshopper. I have picked out two of the more recognizable parts that I could easily read and identify how they helped to form the design.

‘Even after a model is created, other designers can’t easily modify the design because they don’t possess the knowl-edge about how it is created and the original design intent’. -Parametric Technology Corporation, 2008

94

L E A R N I N G O B J E C T I V E S

A N D O U T C O M E S

In respect to the eight learning objectives for ADS: AIR, I feel this journal captivates a thorough and explored understanding of all relevant material. Without going into excessive detail, this semester has focused on delivering a design that is focused on parametric design using Grasshop-per and Rhino. As a group we researched a diverse range of precedents, developed an argument based on this that validated design engrossed in structure, executed a myriad range of iterations, translated the virtual designs into physical models, and then analysed the success and failure of our journey. Our EOI was well explored for third year architecture students but it did not present a design that could be built – which will be explained shortly.

The lecture Daniel Davis gave in the third week of the semester seemed to make most sense to me in the grand scheme of parametricism and computation in architecture. By thoroughly exploring a design within gh it is easy (or at least easier if you have the training) to alter a design earlier in the design process. As the design slowly moves to its realization the cost of design changes rises exponentially, so the prospect of finding these problems and solving them before they are incredibly expensive makes sense. ADS: AIR was a good introduction to this process of design and enlightened me to the flaws in traditional design practice. Bespoke detailing can be achieved if it is done correctly. It has to be brought to attention at the early stages of the project to achieve the most efficient and cost-effective results. This is what stands out from the subject to me.

95

which were not placed in the initial analysis – we managed to achieve a displacement of 90cm. This was very exciting to see but we wanted to reduce it even further. Tom showed that steel was a more effective material to use structurally, and by altering the Karmaba algorithm we changed the profile to be made of hollow steel and reduced the displacement to 30cm.

The images above highlight the displaced areas of the design - the yellow representing successful areas, the purple trouble, and green extreme ar-eas of deflection. We intuitvely knew these would be the most vulnerable points, but it was amaz-ing to see that proved within a virtual simulator

Following the final presentations Dr Stanislav Roudavski advised that we should pursue analyz-ing our design within the Karamba plugin. We sought out Tom Nelson - a master’s student at The University of Melbourne, employee at COX architects, and apart of Alberto Pugnale’s design studio. He helped us to implement our design to be read by Karamba and some amazing results came from this.

Tom advised that the success of the structure re-lies on the displacement, with 6 – 8cm being ac-ceptable. Initially our design had a displacement of more than 2.5 metres, which was an incredible shame, but after thickening the profile to 400mx 200m and adding the diagonal timber laths –

96

Tom also informed us that the design was highly vulnerable to lateral forces such as wind. All the contact points to the ground were more or less in a straight line so the cantilevers would be det-rimental to the stability of the structure and, to prove this, columns were added to the Karamba analysis. This action brought down the overall deflection to an even smaller amount. As a group we had never considered this to be a structural flaw but it now was obvious.

We were very grateful for the help Tom gave and in hindsight it taught us some of the most impor-tant lessons this semester. If we had been able to implement such analysis earlier on in the design process then a more structurally sound model

could have been manufactured and assembled. Stanislav questioned the success of the design as we had not been able to test it standing up but with Tom’s help we had been able to prove that it could potentially stand on the site in Wyndham.

97

1 - Celine Paoli, (2007) ‘Past and Future of Grid Shell Structures’ (Cambridge: Mass: MIT Press), p.6

2 - Mark Cabrinha, (2008) ‘Gridshell Tectonics: Material Values Digital Parameters’ < http://www.arch.calpoly.edu/research/documents/research-0809/Cabrinha_3_0809.pdf> [Accessed 14 May 2013] p.3

3 - Edward Cullinan, (2013) ‘Edward Cullinan Architects’ , <http://www.edwardcullinanarchi-tects.com> [Accessed 20 May 2013]

4 - Commission for Architecture and the Built Environment (CABE), (2011) ‘Weald and Down-land Gridsell’ < http://webarchive.nationalarchives.gov.uk/20110118095356/http:/www.cabe.org.uk/case-studies/weald-and-downland-gridshell/design> [Accessed 20 May 2013]

5 - Dictionary.Com, LLC, (2013) < http://dictionary.reference.com/browse/inertia > [Accessed 24 May 2013]

6 - Mark Cabrinha, (2008) ‘Gridshell Tectonics: Material Values Digital Parameters’ p.4

7 - Celine Paoli, (2007) ‘Past and Future of Grid Shell Structures’ p.34

8 - Celine Paoli, (2007) ‘Past and Future of Grid Shell Structures’ p.42

9 - Celine Paoli, (2007) ‘Past and Future of Grid Shell Structures’ p.44

10 -Celine Paoli, (2007) ‘Past and Future of Grid Shell Structures’ p.7

98