Creative Engineering by Kristjan Plagborg Nielsen

the history of architectural engineering for which competences the engineer should possess to actively contribute to architecture. In the first century B.C. Vitruvius described how architecture was concerned with stability, function and beauty. But as Ove Arup asks: What is beauty? And how does one become creative? According to Walter Gropius Creativity cannot be taught. Creative engineers have the possibility to contribute to architecture if they only dare. Both creativity and beauty can be hard subjects for young engineering students to grasp when only being taught science. It is the intent of this book to demystify those topics through a historical perspective on the creative engineer. When you create something you are building on the works of others. One cannot imagine an architect that does not know of the history of architecture. Engineers function in the space between humanities and science, but are only being taught the latter.

end of the master builder era and the rules of proportion derived from the Ancient Greeks’ study of beauty; through Galileo, Hook, Coulomb, Navier and others on the development of the scientific theories used today; how the industrial revolution gave birth to the structural engineer as an independent race which lay ground for “Structural Artists” as Nervi; to Ove Arup, Peter Rice and the modern architectural engineering company Bollinger + Grohmann. It is investigated which competences the engineers succeeding in good design possessed. Through four different scaled case studies of present day projects of Bollinger + Grohmann, it is shown what is asked of the engineer today; whether being designing a stadium for the World Cup or developing an art piece with modern day technology. This book was written as Kristjan Plagborg Nielsens Master’s Thesis in Architectural Engineering from Aarhus University.

HOW ENGINEERS CONTRIBUTE TO ARCHITECTURE

This book takes the reader on a journey through the history of engineering; from the

CREATIVE ENGINEERING

Creative Engineering - How Engineers Contribute to Architecture is a search through

CREATIVE ENGINEERING HOW ENGINEERS CONTRIBUTE TO ARCHITECTURE

Kristjan Plagborg Nielsen

CREATIVE ENGINEERING HOW ENGINEERS CONTRIBUTE TO ARCHITECTURE

Kristjan Plagborg Nielsen

Masterâ&#x20AC;&#x2122;s Thesis in

Architectural Engineering at

Aarhus University Contact: Kristjan Plagborg Nielsen

+33 (0) 6 5207 7608 / +45 2614 1209

kristjan_pn@hotmail.com

Study number: 20108816 Main supervisor (engineering): Lars German Hagsten (lgh@iha.dk) Project supervisor (architecture): Morten Reimers Knudsen (mrk@iha.dk)

August 2013

Introduction

I have been writing this thesis while working for the engineering company Bollinger + Grohmann in Paris. I ended up in Paris, as I was searching for an alternative to the 3rd semester in my Masterâ&#x20AC;&#x2122;s studies, and I was given the chance of an internship. After the six months internship ended in August 2012, the company wanted me to stay and I thought it would be a good idea to write a thesis while working â&#x20AC;&#x201C; when everything comes to an end both work and thesis are on the same topic: collaboration with architects.

WHY THIS THESIS I always wanted to do a thesis on the architectural engineering collaboration. Architecture fascinates me and I have come to realise that this is why I am doing engineering. To create something beautiful is what I seek to do. My education in architectural engineering matches the company I ended up in. Bollinger + Grohm-

ann describe themselves as:

The passion for high-quality architecture and innovative structures has been the motivation for Klaus Bollinger, Manfred Grohmann and their team since the foundation of the office in 1983. As responsible engineers, our prime focus is the strengthening and enhancement of each individual design. We see ourselves as partners within an interdisciplinary planning team. We develop tailored solutions together with architects and clients, construction companies and industry. This is an integral component of the complete concept and not an end in itself. Besides technological innovation Bollinger + Grohmann Ingenieure prides itself in openminded dialog and fair respectful dealings with all stakeholders. We are convinced that sus-

tainable solutions can only develop in unison with technological and social progress. 1

Working in an office that collaborates with some of the best architects in the world is a great privilege. It was an opportunity that not many students have, and it was a unique chance to use my projects in my thesis as case studies. When I started out considering the topic for my thesis, I wanted to do something on concept design. I thought, and still do, that I was really good in making structural concepts, but it was hard for me to formulate the statement of the problem. I started to read a couple of books on the subject, and I thought that these books

were not very good, or at least not what I had expected. It became clear to me, that one cannot generalise what the best concept is. Creativity cannot be taught. I explained my frustration to Lars G. Hagsten, my supervisor, who told me that this thesis could either be a chance for me to learn something new, something that I felt was missing in my education, or it could be used to investigate something that he could use prospectively in the architectural engineering education. In extension of my initial idea â&#x20AC;&#x201C; and perhaps his obsession with the late engineer Peter Rice â&#x20AC;&#x201C; he proposed to investigate if the successful engineers throughout time had something in common. 1

1: The Bollinger + Grohmann office in Paris. The main office of the firm is located in Frankfurt, Germany, but also have offices in Vienna, Oslo, Melbourne and Berlin. Image courtesy of Bollinger + Grohmann

STATEMENT OF THE PROBLEM I have always disliked the cliché of “bridging the gap between architecture and engineering”. I think it is a bad attitude to have before engaging in a team, that we are going into this team to bridge a gap. Why not instead assume that there is no gap? I also have the idea that architectural engineering is about bringing the best from both worlds to reach a higher unity than the sum of both. Engineers should, in my opinion, add something positive to the design, and not only make the architects design constructible. Engineers should contribute to architecture. We ended up agreeing on the problem formulation:

“How can the engineer contribute to architecture?”

with the sub question:

“Which competences should the engineer possess in order to successfully contribute to architecture?”

In a historical perspective, I wanted to investigate if there were any time-independent competences that all engineers from the past had possessed. Through case studies I wanted to research

the competences an engineer should posses at present day and finally try to preview; what will be the future competences and engineer should possess?

METHOD I wanted to investigate which competences the famous architectural engineers possessed and why they succeeded in creating good design. To be honest, at the time I only knew very few engineers which could be categorised as “architectural engineers”: Pier Luigi Nervi, Santiago Calatrava, and Peter Rice who only was introduced to me by Lars G. Hagsten. I started to realise that I did not know much about the history of my own profession. And when asking fellow students, neither did they. Here was in fact a gap in my knowledge – a chance to use my thesis to learn something.

“Unfortunately, civil engineers [...] have no historical understanding of their own profession, in contrast to architects. Hardly any faculties teach the history of engineering, but it is important to be aware of the development of one’s own profession. It would be inconceivable to train architects without teaching them anything about the history of architecture. [...] Knowing which developments were made and when gives us a completely different basis for reflecting on our own activities and developing them further.” 2 -Manfred Grohmann

My other supervisor, Morten R. Knudsen, proposed me the idea of creating a book â&#x20AC;&#x201C; a book that could be sold to students enrolling in the architectural engineering education. I was keen on the idea from the beginning. It would fulfil the other of criteriom of Lars G. Hagsten; to make something that he could use prospectively in the architectural engineering education.

I would also like to thank my friends and family for support and help during the last one-anda-half year that I have been living abroad. I would especially like to thank my colleague Marie Boltenstern for reading, commenting and correcting all my text.

So I began to read...

ABOUT THIS BOOK The present book is therefore targetting new engineering students who possess an interest in the history of engineering. Despite that, my aim when writing this book has been to make the majority of the content understandable for laymen. I have chosen to write it in English so more people could read it and make use of it, including my colleagues, but also because my projects have been done in English as the main language.

ACKNOWLEDGEMENTS I would like thank my supervisors Lars G. Hagsten and Morten R. Knudsen for letting me go my own way and for some very cosy and inspiring meetings the few times we have met. Thanks to my colleagues who were nice enough to leave me alone when I was finishing this thesis.

Table of contents 6 Introduction 6 Why this thesis • 8 Statement of the problem • 8 Method • 9 About this book • 9 Acknowledgements

12 Preface

PAST 18

Ars sine scientia nihil est… 19 Vitruvius • 20 Pythagoras • 23 Villard and the Gothic style • 28 Brunelleschi and the end of the master builder era • 32 Alberti and Palladio • 38 Coda chapter one

Science and Engineering 40 Hanging chain and arch structures • 42 Galileo and the idea of stresses • 44 Bernoulli, Euler and deformation • 47 Plasticity and the ultimate state • 50 Coda chapter two

Structural Art 55 Thomas Telford • 64 Isambard Brunel • 72 Gustave Eiffel • 82 Viollet-le-Duc • 86 Coda chapter three

The modern craftsman 88 François Hennebique • 94 Coda chapter four

Modern Architects 101 Frank Lloyd Wright • 104 Mies van der Rohe • 112 Le Corbusier • 120 Coda chapter five

122

Architectural engineers 123 Maillart • 130 Freyssinet • 136 Torroja • 144 Nervi • 154 Coda chapter six

156

Engineering architects 156 Gaudí • 166 Buckminster Fuller • 172 Frei Otto • 176 Coda chapter seven

178

Ove Arup 179 The outsider • 182 Christiani & Nielsen and the first years in London • 187 J. L. Kier & Co. and connection with avant-garde architects • 192 Arup & Arup Ltd. and The Second World War • 194 Ove N. Arup Consulting Engineer – an engineer working with architects • 199 Ove Arup’s legacy • 204 Coda chapter eight

206

Peter Rice 207 Arup and Sydney Opera House • 212 Centre Pompidou • 221 Materials • 232 The role of the engineer • 238 Coda chapter nine

242

Bollinger + Grohmann 245 Computational design • 254 The character of the engineer • 258 Coda chapter ten

PRESENT 266

Kaliningrad Stadium 269 Competition • 280 Schematic Design

304

Antwerp Provinciehuis 305 Tower • 318 Congress

334

Hermès Miami 340 Interior structure • 345 Glazed façade • 350 Sun shading façade

362

Planetarium Sorrow 366 Tuileries 2013 and other boundaries • 368 Form finding through generative Optimisation • 369 3Dprinting and design of the nodes • 379 Choosing the final shape

FUTURE 386

Future of architectural engineering 386 • Digital design; 390 • Conditions for succeeding; 390 • Competences

396

Statement of the problem

398 Notes 410 Bibliography

Preface

When you are creating something, you are building on the work of others. Our ability to think of new possibilities is always influenced by our existing beliefs. Engineers are not scientists neither are they artists. The engineerâ&#x20AC;&#x2122;s role is to design structures in constructions that have a function; anything from a bridge to a house. Engineers take their scientific knowledge and apply it to find the right solution.

â&#x20AC;&#x153;Engineering is not a science. Science studies particular events to find general laws. Engineering design makes use of these laws to solve particular practical problems. In this it is more closely related to art or craft; as in art its problems are underdefined, there are many solutions, good, bad and indifferent.â&#x20AC;? 1 -Ove Arup

The engineering education is primarily with a scientific approach, and there is rarely more than one solution to a problem. It is black and white; based on natural laws and facts. As a newly graduated engineer you only know the one part of your field: the scientific. But engineers act in a field between humanities and science. One cannot imagine a newly graduated architect who is not aware of the history of architecture. Why do engineers not know of their own history? The engineer emerged from the master builder, who at the time functioned as architect, engineer and builder at the same time. Today, engineers should possess knowledge of both architecture and construction techniques. Although it is of much importance, the latter will not be a main topic in this book, as according to Peter Rice:

“[…] the work was really rather simple, no more than applied common sense.” 2

tioned as the engineer. I have chosen to use the term architect rather than engineer for the first chapters, as it will be shown later how the engineer emerges as a separate race.

The engineer should possess common sense. It is a time independent competence, and it is assumed that all successful engineers described in this book possessed common sense. It is also assumed that the engineers investigated in this book did know the scientific engineering knowledge of the day, unless otherwise described. The history of the engineer starts with the breakdown of the master builder. In the beginning the breakdown was that of the architect and the craftsman, as the architect still func2

2: The master builder, here characterised by Eddie Redmayne in the screen version of Ken Follet’s book The Pillars of the Earth. Image courtesy of Channel 4

PART ONE

PAST

ARS SINE SCIENTIA NIHIL EST… PAST

Chapter one

Ars sine scientia nihil est…

… declared Jean Mignot while arguing with an Italian lodge constructing the Milan Cathedral in the end of the 14th century. Ars sine scientia nihil est can be translated in different ways: “Practice is nothing without theory” or “Art is nothing without knowledge”. Mignot was arguing with the Italians workers about the shape and height of the vault of the new cathedral. The workers claimed that “theory is one thing, practice is another”, arguing that Mignots propositions for the cathedrals roof was not the best way to construct in practice. A movement towards specialisation started to arise in the beginning of the 15th century, where the professions of the architect and the engineer began to diverge, ending the era of the “master builder”. The master builder had always been equally skilled in art, engineering and craftsmanship. Mignot lacked this combination, and was more focused on the Gothic theory of how the pro-

portions of a cathedral should be rather than how to construct the cathedral. Mignots “theory” came from his secret lodge book - a book on how to construct. In it were rules of geometry and proportions, like the diameter of columns as a function of their height or the space between two columns as a function of the column diameter. The rules were scale independent and could be used for everything from a lintel to a whole cathedral. All construction sites could use their own unit, but on one construction site the same unit of measurement – the great measure – had to be used everywhere. The master builder being the single person responsible for the construction, had his measure rod, and everything from each construction element’s dimension to the geometry of the ground plan was found from this single rod. The knowledge had been passed on from one generation of building workers to the next by

word of mouth or occasionally been written down. Succesful designs from the past were noted in the master builder’s lodge book. The rules were in sense laws of construction; the buildings constructed after these rules were proven stable, as the buildings in which they were applied had not failed. 1 But the rules of proportion originated allegedly also from something else than to just keeping the building stable.

VITRUVIUS In Vitruvius’ (c. 80-70 – c. 15 BC) first book from the 1st century B.C. he describes three essential basics of architecture: • Firmitas - strength • Utilitas - functionality • Venustas - beauty A building should first of all have strength: be solid and stable – in short it should not fail. Next it should be useful and fulfil the needs of the users. Last, Vitruvius claims, to be architecture a building must be beautiful. The last aspect is a difficult one. What is beauty? Vitruvius thought that a timeless notion of beauty could be learnt from the ‘truth of nature’. He stated that nature’s designs were based on universal laws of proportion and symmetry. He believed that the body’s proportions could be used as a model of natural proportional perfection. He wrote about the way

ancient scholars examined many examples of ‘well shaped men’ and discovered that these bodies shared certain proportions. He showed that the ‘ideal’ human body fitted precisely into both a circle and a square, and he thus illustrated the link that he believed existed between perfect geometric forms and the perfect body. In this way, the body was seen as a living rulebook, containing the fixed and faultless laws set down by nature. Vitruvius’ books on architecture were one of the first texts in history to draw the connection between the architecture of the body and that of building. Vitruvius believed that an architect’s designs must refer to the unquestionable perfection of the body’s symmetry and proportions. If a building is to create a sense of eurythmia – a graceful and agreeable atmosphere – it is essential that it mirrors these natural laws of harmony and beauty. In his first book he also describes his idea of architecture and engineering:

“[Architecture and engineering] is a body of knowledge comprising many disciplines and sciences which can also be applied in other arts. Finished works are born of skill in manufacture and design. Manufacturing skills come from the constant study of craftsmanship and the working materials to create whatever the desired result. Designing is the ability to convey the scheme for the finished object to others and to provide a rational explanation for the scheme using engineering knowledge and scientific principles.” 2

ARS SINE SCIENTIA NIHIL EST… PAST

Vitruvius describes throughout his books how to design structures: sizes, relative position and orientation, how to select and use suitable materials, and in particular focuses on the importance of economy. He advocates for a proper management of the construction site and materials, and to use common sense to reduce costs. He uses the example of primarily using materials that are available locally, and he stresses the necessity in matching the cost of the work to both the client’s purse and needs. The rules of proportion in Mignot’s master builder’s lodge book had not only been derived through a trial-and-error history of constructions stability, but also from aesthetical reasons and the need for different functions. According to Vitruvius, a building needs to be functional, stable and beautiful, and beauty came from the right proportions. In the Roman conception of architecture, everything influencing the users – the people – should be taken into account, both physical and psychological influences. The beauty of proportions were not Vitruvius’ own invention, he was highly influenced by the ancient Greeks thoughts on the topic. 3

PYTHAGORAS Beauty and proportion in architecture has often been compared with music. What make one find a building beautiful? Why does the ear appreciate consonance over dissonance? The ancient Greeks sought to explain beauty in buildings and music. According to the legend of the Pythagorean hammers, Pythagoras discovered the foundations of musical tuning by

listening to the sounds from four blacksmiths hammers. The legends says that one day Pythagoras (c. 570 – c. 495 BC) went by a blacksmith where he from the outside could sometimes hear the most beautiful harmonies. He ran inside and discovered that the weight of each of the blacksmiths four hammers was different. When struck, each hammer therefore produced a different tone. When striking two hammers at the same time, some created consonance and others created dissonance. Pythagoras then went home to his monochord, an instrument with a single string. When he divided the string into two equally lengthen pieces he had created the octave, or the ratio 2:1. When dividing the string in the ratio 4:3 he could play the fourth of the base tone and dividing his string into 3:2 ratios he played the fifth. When a person is normally listening to music, he or she doesn’t know the string length – but the person can still hear if the tones are consonance or dissonance. The ancient Greeks thought of these mathematical relations as divine, and to Pythagoras the numbers 1, 2, 3, 4 were special. He realised that he could calculate the octave by multiplying the fourth and the fifth (4/3 x 3/2 = 12/6 = 2/1). Another mathematical relation also caught Pythagoras’ interest: the pentagram. Connecting the five points of a pentagon creates a regular five star polygon, the pentagram. In a pentagram the relation between the edge of the fivestar and the edge of the inner pentagon gives

The only books on building design surviving from Greek and Roman times are Vitruvius’ De Architectura – ten books on architecture published in Rome about 25 B.C. In his books Vitruvius collected the knowledge from several decades, his design guidance may date back three or four centuries earlier. His books have been available, though for centuries more or less forgotten, for more than two thousand years. Vitruvius was born sometime between year 80 and 70 BC. He grew up between Rome and Naples and joined Julius Cesars army as architect and engineer at a young age. He built a couple of bridges and a single building that is no longer existing. He retired shortly after turning 50 years old. Vitruvius then began writing an architectural testament, ten books covering all topics of building design, as it was known in the Roman Empire of the time. The books covered town planning, building design (architecture and civil engineering), construction materials, functions and proposed layouts of different type of buildings, pavements and facades decorations, water supply and aqueducts, science influencing buildings as geometry, sundial and astronomy, and finally machines.

3: Pages from an illustrated version of Vitruvius “De Architectura” published in 1567, on left illustrations on how to solve the geometrical problem of halvfing a square. Image courtesey of New York University Department of Mathematics.

ARS SINE SCIENTIA NIHIL EST… PAST 4

the golden ratio. By connecting the points of the inner pentagon a new pentagram can be made by the rule of the golden ratio. The golden ratio can be derived in several ways: the basic principle is to divide a line into two parts, a and b. If the line is divided with the golden ratio, the length (a plus b) is to a what a is to b:

It must have been a problem for Pythagoras to solve for j, as the decimal system was not yet known. He was forced to round it to rational numbers. Pythagoras and Euclid (fl. 300 BC) were the first in history to describe the golden ratio, but it can be shown that it was used in the basic geometry of earlier buildings, like the Giza Pyramid from around 2560 BC.

The ancient Greeks transferred the rules of harmony in music to not only their buildings, but also to describe everything from the human body to the astronomy of the Greek night sky. They believed the whole universe must have been created according to this divine harmony. The harmony of the ear being the consonance tones: the harmony of the eye being the golden ratio. 4 Vitruvius’ ten books summarised the building theory of the Roman times, combining engineering, art and craftsmanship. Although Vitruvius’ proportional rules did not follow the golden ratio completely – as noted before, at his time only a system of rational numbers was known – Vitruvius collected the knowledge passed on by word of mouth for generations, combining the needs for stable, well functioning and beautiful buildings. Vitruvius might have done

4: Human proportions related to the golden ratio as Pythagoras imagined it, Fibbonachi spiral also drawn in on the right. Image courtesy of Piotr Sadowski

the same as Le Corbuiser (1887 – 1965) did with his “Le Modulor” measurement, in which the dimensions found through the golden ratio are refined to a round number. In many ways Vitruvius’ ten books covers the same topics that concerned ancient Greeks, almost everything influencing the user. The master builder of Roman times must have possessed knowledge on multiple topics, something that took a lifetime to master.

VILLARD AND THE GOTHIC STYLE The remaining sheets of Villard De Honnecourt’s sketchbook from 1235 shows an example of what a master builders secret lodge book could have contained. In the same way as Vitrivius, Villard describes how to design the proportions of a building. Villards manuscript, similar to Vitrivius’, describes a solution of the geometrical problem of how to halve a square. Plato theoretically solved the problem, but it contained the irrational number of square root two. Vitrivius described and Villard sketched how such a length could be constructed. For medieval builders this was a problem to be solved, and the answer was not shared with layman – it stayed a secret within the lodge.

In the thirteen century the Italian mathematician Leonardo Pisano Bigollo, better known as Fibonacchi, introduced the Hindu-Arabic numeral system to Europe. Fibonacchi published a book of calculations, Liber Abaci, in 1202. He describes an example of a number sequence, later to be known as Fibonacchi numbers. In the Fibonacci sequence of numbers, each number is the sum of the previous two numbers, starting with 0 and 1. This sequence begins 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987 ... The sequence has that interesting feature, that the higher through the sequence, the closer to the golden ratio. For example 2/3 is far from, 3/5 is closer, but already at 5/8 is relatively close. 5 5: Le Corbusiers “Le Modulor” based on the human proportions of a average height British policeman Drawing by Le Corbusier

ARS SINE SCIENTIA NIHIL EST… PAST

ARS SINE SCIENTIA NIHIL ESTâ&#x20AC;Ś PAST

A master builder known by name is William of Sens (c. 1130 â&#x20AC;&#x201C; 1180), and his is one of very few master builders whose existence is recorded in the annals. He was born around 1130 and was from a wealthy family, his dad being a respected craftsman, mason or stone carver. William of Sens is a good example of a master builder of the Gothic period. He grew up in Saint Denis during the construction of the abbey church, worked as a child under the master builder until the completion of the church around 1144 and then moved to Sens where the first Gothic Cathedral was being built. He was enrolled in a mason lodge at the age of twelve. William became the first to design a Gothic choir in the Sens cathedral, and his design was copied to several other churches throughout the Gothic period, among them the Lausanne Notre Dame completed in 1240. The form and technique developed by one master builder in France was distributed to the rest of Europe by the network of mason lodges. William took over the job as the master builder of the Sens cathedral, and after the completion of the choir he took the task of rebuilding the choir of the Canterbury cathedral in 1174. For the Canterbury cathedral William further develop his structural ideas for the choir, making it even higher and more daring than in Sens. He even invents new machinery to help speed ud the construction process. In 1178 he falls down from the scaffolding and get seriously injured. From his bed he still runs the construction site for another year, before returning to France where he dies in 1180. Allegedly he had the whole Canterbury cathedral in his head, and was able to run the construction site from a distance. 6

6: (Previous page) Picture of the inside of the Saint Denis Basilique. Image courtesy of Gilles Messian 7: (Above) The high nave in the Canterbury Cathedral, England. Image courtesy of Chris Beckett

Villard was – also like Vitrivius – not a great architect of his time. He travelled and collected information for his lodge in the “golden age” of Gothic architecture (1140 to 1284). 6 One of the pieces of the Gothic period is the abbey church of Saint Denis, located just outside Paris. This is one of the first Gothic style buildings. The name of the master builder is unknown, but it is known that abbey Suger (c. 1081 – 1151), as the head of the church being the client, was greatly involved in the planning and administration of the construction. The Gothic style differs from the previous Classic or Roman style by a new structural principle – it is no longer the walls that carry the ceiling and roof, but a system of pillars and crossrib vaults. It is a self-supporting construction, which renders the heavy walls of the Roman style redundant and allows space for huge windows. The lines of the forces are clearly readable in the vault, where the lines of the normal forces created two simple arches. In the same way Nervi (1891 – 1979) shaped flat concrete structures slabs according to the isostatic lines of the bending moment in the beginning of the 20th century. The Gothic designers must have had a developed knowledge of how the forces in the structure worked to be able to create the, for this time, enourmes buildings.

to stay stable a thickness of 20 % of the diameter, or the span, is required. The Gothic designers introduced the parabolic arch, which could reduce the required thickness to 10%. This, of course, reduced the weight of the structure and made higher structures possible. Until the first railroad bridges in the middle of the 19th century the governing load for structures was the self-weight. Another great weight reducing effect came from the use of barrels intersecting at the right angles. This created a three-dimensional structure, which only needed support at four points. When erecting the structure of the vault, the 8

The Gothic also separated its style from the Roman in the geometric shaping. Almost all Roman arches were of semi-circular shape, probably to prevent strong horizontal forces at the support but also easier to construct with simple tools. For a semi-circular construction

8: The ceiling in the Gatti Wool Factory, Rome, designed by Nervi, 1953 Image courtesy of Vasari, Rome

ARS SINE SCIENTIA NIHIL EST… PAST

barrel intersection lines were normally built first using heavier stone ribs, then afterwards the in-between surfaces were constructed with lighter bricks. These more dense lines were visibly and very prominent in the impression, helping the impression of a “readable” structure, but even though the ribs collected more forces – as they are stiffer than the brick - they were not structurally necessary. The construction sequence therefore had an impact on the design, with less scaffolding needed when erecting the ribs prior, speeded up the time of construction. Constructing higher and less dense also introduced a new dilemma for the designer: lateral stability. Buttresses were introduced at the lower sides of the church nave, stabilising the

higher nave. 8 In 1400 the Milan lodge rejects the Gothic rules for proportion given in Mignots secret lodge book and continue the work by their own rules. The master builder’s theory was dismissed in favour of the craftsman’s and the tripod of art, engineering and craftsmanship was coming to its end.

BRUNELLESCHI AND THE END OF THE MASTER BUILDER ERA At almost the same time, 250 km south east of Milan, the city of Florence was building their new Cathedral. In 1418, when construction had been ongoing for almost a century, the time had come to finalise the design of the dome in

9: Panoramic view of present city of Florence, with the Cathedral still dominating the city’s skyline Image courtesy unknown

the Santa Maria del Fiore Cathedral. A dome spanning 42 m in diameter had been anticipated in the project, but no realisable design had been made so far. After Pantheon in Rome, the dome would be the biggest in Europe of the time, and the existing church structure did not have the big walls that were normally needed to support a shell dome. Neither did the design leave room for Gothic buttresses, as the city fathers forbade those. A competition was held to answer the technical problem. Filippo Brunelleschi (1377 – 1446) was a sculptor educated in the goldsmith’s guild. Not much is known of his transition from sculptor to architect and engineer, although it is told that he was well founded in mathematics in school. He entered and won the competition, although

the jury questioned if his design was possible to construct: the dome was to be erected 52 m above ground level and reaching 90 m at its top. Normally the construction of a dome required heavy scaffolding until the key stone was put in place, and no raft was long enough to support a dome at 90 m height. Brunelleschi claimed that he could build the dome without centring, but unwilling to reveal his idea as the jury would find it childishly simple. A model of wood and brick was created to illustrate his proposal, but the model was incomplete to insure him control of the project. Brunelleschi’s ingenious idea was to use tension rings of stone and iron to keep the dome from spreading. The dome was constructed as a double shell – a lightweight hollow structure

ARS SINE SCIENTIA NIHIL EST… PAST 10

– making room between the two for an access to the lantern on top. Brunelleschi used the Gothic principle of the intersecting barrels to divide the dome into eight pieces, having eight major ribs, and use more lightweight bricks in between. By equalising the horizontal forces within the dome with the tension rings, Brunelleschi avoided the need for buttresses or massively, thick walls to carry the spreading forces to the ground. In between each of the eight major ribs, which are visible from the outside, two secondary, invisible ribs were placed. The ribs were locked in place by four tension rings made of sandstone – 43 cm in diameter and each 2,3 m long – and one made of wood, locked to-

gether by iron and eventually creating a lightweight skeleton. Brunelleschi’s idea also incorporated the construction sequence, using the five tension rings to equalise the horizontal forces as the works progressed towards the top. Brunelleschi’s design also had the advantage, that most of the structure was stable during construction – in comparison to a simple arch, which is only self-supporting after the keystone has been put in place, a dome structure is stable when a complete ring has been put in place. The dome does not need an arch in two directions to be stable, as seen in the Pantheon roof. Brunelleschi used this to his advantage, and a minimum of scaffolding was needed to construct the dome.

10: The dome of the Florence Cathedral, Santa Maria del Fiore Image and sketch from “Brunelleschi’s dome” by Ross King

Concrete was not available at the time of Brunelleschi. Mixing aggregate, cement and water have been known for many years, the earliest recognitions of cement dates as far back as twelve million years. In the hands of the Roman Empire, concrete influenced a structural revolution. With concrete structures the Romans perfected arches, vaults and domes. Underwater engineering was also developed, which was perhaps more important to the Roman civilians, leading to the creation of aqueducts and dams for clean, running water inside houses. The Romans used volcano ashes as cement, and maybe the lack of volcanic ashes elsewhere in Europe made craftsmen forget the recipe during the sixth century. Concrete was only rediscovered more than a thousand years later. 9 In the year 126 Pantheon was constructed in Rome. For a very long time the largest dome of the world, and still the largest unreinforced dome, the Pantheon is a concrete colossal and a symbol of the advances of Roman construction. The dome, or Rotunda, is shaped after a perfect sphere, with an internal diameter of 43,3 m. From the 9,1 m diameter hole at the top, the concrete thickness varies from 1,2 m to 6,3 m at the base. The dome is constructed together with building at its rear which acts as a buttress stabilising the dome together with the massive walls. The dome is estimated to weigh 4500 tons and 20th century finite element calculations of it show that probably only very small tension stresses exist within the dome. 10

11 11: Cross section of Pantheon (right page) from Pallatio’s “Quattro libri dell’architettura” Image courtesy of The Library of Congress

ARS SINE SCIENTIA NIHIL EST… PAST

ALBERTI AND PALLADIO While Brunelleschi was the practitioner of Florence in the 15th century, Leon Battista Alberti (1404 – 1472) was more the theorist. Also from a Florence family, Alberti received the best education at the time; studying classicism from 1414 to 1418, later law and in his spare time composing music and allegedly being a superb horseman. Alberti even wrote a Latin comedy and the classicism was a great inspiration for him. He saw mathematics as the common basis of art and science, and took mathematics as the basis for his study of perspective, published in 1435. Alberti had a clear idea of what he saw as beauty and how to achieve it. He was very inspired by nature, claiming that the process of

learning should be sought from nature. He believed the ultimate goal of an artist was to imitate nature subjectively, creating beauty with harmony between all parts in relation to one another. Alberti believed that beauty could be:

“realised in a particular number, proportion and arrangement demanded by harmony”,

a theory reflecting the one of Pythagoras. In 1452 Alberti published De Re Aedificatoria, “On the Art of Building”, with the same structure as Vitruvius’ ten books. Alberti’s book is believed to be the first treatment of Renaissance architecture, and it is perhaps Alberti’s treatment of Vitruvius that made Vitruvius pub-

12: Design of a villa from Pallatio’s “Quattro libri dell’architettura” Image courtesy of The Library of Congress

13: Geometric comparison between Palladio’s and Le Corbuiser’s villas, from Colin Rowe’s essay on “The Mathematics of the Ideal Villa”, published in “Arhcitectural Review” March 1947

ARS SINE SCIENTIA NIHIL EST… PAST

licly known today. Alberti stated the book was written “not only for craftsmen but also for anyone interested in the noble arts,” and throughout his book, Alberti brings the theories of Renaissance to architects, scholars and others. 11

Although the Gothic might have broken some of the tradition of the master builder, Albertis book now made it possible for laymen to construct according to the rules of proportion.

Like Alberti, Leonardo Da Vinci (c. 1452 – c. 1519) was inspired by the proportions of the human body. Da Vinci’s famous drawing, The Vitruvian Man from ca 1490, reflects the description of the human body by Vitruvius: “For the human body is so designed by nature that the face, from the chin to the top of the forehead and the lowest roots of the hair, is a tenth part of the whole height; the open hand from the wrist to the tip of the middle finger is just the same; the head from the chin to the crown is an eighth, and with the neck and shoulder from the top of the breast to the lowest roots of the hair is a sixth; from the middle of the breast to the summit of the crown is a fourth. If we take the height of the face itself, the distance from the bottom of the chin to the underside of the nostrils is one third of it; the nose from the underside of the nostrils to a line between the eyebrows is the same; from there to the lowest roots of the hair is also a third, comprising the forehead. The length of the foot is one sixth of the height of the body; of the forearm, one fourth; and the breadth of the breast is also one fourth. The other members, too, have their own symmetrical proportions, and it was by employing them that the famous painters and sculptors of antiquity attained to great and endless renown. Similarly, in the members of a temple there ought to be the greatest harmony in the symmetrical relations of the different parts to the general magnitude of the whole. Then again, in the human body the central point is naturally the navel. For if a man be placed flat on his back, with his hands and feet extended, and a pair of compasses centred at his navel, the fingers and toes of his two hands and feet will touch the circumference of a circle described there from. And just as the human body yields a circular outline, so too a square figure may be found from it. For if we measure the distance from the soles of the feet to the top of the head, and then apply that measure to the outstretched arms, the breadth will be found to be the same as the height, as in the case of plane surfaces which are perfectly square.” 12 The points determining these proportions are marked with lines on the drawing. Below the drawing itself is a single line equal to a side of the square and divided into four cubits, of which the outer two are divided into six palms each, two of which have the mirror-text annotation “palmi”; the outermost two palms are divided into four fingers each, and are each annotated “diti”.

14: (Opposite page) Leonardo da Vinci’s famous “Vitruvian Man” Drawing by Leonardo Da Vinci,1490

ARS SINE SCIENTIA NIHIL EST… PAST

Palladio’s Four Books of Architecture were at once enourmously influental when published in 1570. Andrea Palladio (1508 – 1580) was born near Venice and served a apprenticeship as a stonecutter. His great talent was recognised by humanist and poet Trissino, who helped and supported his career. Trissino was reconstructing a villa and was greatly inspired by Vitruvius and introduced his ideas to Palladio. Trissino granted Palladio to travel to Rome to study the Classic architecture of Roman time. Palladio began to develop his own architectural style around 1541, as he rediscovered the classical Roman principles. Palladio was in comparison to Vitruvius, Villard and somewhat Alberti, a very active architect and not only theoretical. He worked all over Italy and was appointed chief architect of the Republic of Venice. In 1570 he publishes “I quattro libri dell’architetture” in Venice. The four books contains Palladio’s own designs, celebrating the idea of purity and simplicity of classical architecture. Palladianism swept through Europe in the following years, and Palladio is by many historians seen as the most influencial person in history of Western architecture. The four books was in detail a manual on how to construct and plan. Palladio created a new style and a new vocabulary of forms. Palladio was very aware of the proportional effects of the golden ratio and made great use of it in his buildings. In 1947 Colin Rowe compared Palladio’s villas with Le Corbusier’s, and how they

both make use of the golden ratio. Palladio was not only designing villas for the high society of Venice, he also was also creating bridges.

15: Design of bridges from Pallatio’s “Quattro libri dell’architettura”, courtesy of The Library of Congress 16: (Opposite page) Palladio’s bridge in Bassano del Grappa bridge, courtesy of Comune di Bassano del Grappa

ARS SINE SCIENTIA NIHIL EST… PAST

CODA CHAPTER ONE Gothic churches and cathedrals were becoming ever higher with an ever slender structure. In the preceding Classical style, the stability of a large structure as cathedrals was secured by the heavy weight of the massive stone walls. During the Gothic period, walls were replaced by slender columns and the structures needed lateral stabilisation to prevent it from turning over. The solution was buttresses, external additions that were incorporated in the architecture. The architects used their engineering competences to secure the building’s stability but without compromising the architecture – in fact they determined a new architectural style to reflect the engineering needs. It was the engineering solutions, such as Brunelleschi’s dome, that created the evolution of architecture. In his book, Palladio described bridges designed with a more technological approach. To some extent he must have possessed a good idea of how the forces in his bridges were distributed. Architects of the time secured their building’s stability by applying Vitruvius’ rules of proportion. Time had proven these rules to be sufficient, but for the ever slender structures of the Gothic style a more detailed verification of a building’s stability and a more scientific approach was needed.

SCIENCE AND ENGINEERING PAST

Chapter two

Science and Engineering

It was in the process of understanding the thrust in an arch structure that the first numerical calculations were carried out. The ancient and medieval way of constructing everything from arch bridges, vaults or even window lintels was based on the rules of proportion – if an arch window lintel was proven stable the structural dimensions could be scaled a hundred times to be used for a masonry arch bridge. With the Gothic style new shapes and thinner structures were being designed, differing from the Vitruvius’ rules of proportion, and new rules were necessary. The engineers had to find a way to ensure that a structure was stable. It became evident to the scientists of the time that the proportional rules did not cover the needs of the new Gothic style, reaching ever higher with ever slender columns:

“nor could nature make trees immeasurable size, because their branches would eventually fail of their own weight” 1

as Galileo Galilei (1564 – 1642) expresses it through his character Salviati in his “Discussions” published 1638.

HANGING CHAIN AND ARCH STRUCTURES In 1670 Robert Hooke (1635 – 1703) explains the principles of how an arch structure could be calculated to the British Royal Society. Hooke was trying to find the “true” form of the arch. In 1675 he published an anagram stating:

“as hangs the flexible line, so but inverted will stand the rigid arch.”

Robert Hooke was born on Isle of Wight, Great Britain. Since his childhood Hooke found interest in observations and mechanical work, for example he dismantled a brass clock to build a wooden replica. When he later studied Euclid’s “Elements” his lifelong interest in the study of mechanics began. Robert Hooke went to study at Oxford University where he was employed to assist the natural philosopher Robert Boyle (1627 – 1691). Hooke constructed and operated Boyle’s air pump machine called “machina Boyleana”. While at Oxford, Hooke made friends with scientist and architect Christopher Wren (1632 – 1723) for whom he designed several building after the great London fire in 1666. In 1660, Hooke discovered the law of elasticity, Hooke’s Law. He developed his law for practical purposes, to design a spring for a more accurately working watch. Hooke published his discovery as an anagram, something used at the time to establish an understanding of the discovery but to keep the method secret, something also used by Galileo. In the published anagram from 1678, Hooke states “as the extension, so the force”. Hooke held a lecture on gravity to the Royal Society. He had disputes about the gravitational theory with Isaac Newton (1643 – 1727), who also was a member of the Royal Society. In 1686, when Newton’s first book “Principia” was presented to the Royal Society, Hooke claimed that Newton’s notion of “the rules of the decrease of Gravity, being reciprocally as the squares of the distances from the Center” was his idea. Newton believed that Hooke only made a guess, and that Newton himself was the one who prove it mathematically. Robert Hooke was more the practical scientist, his diaries testify that he had a lot of ideas that he never followed. Newton was on the other hand more the theoretical scientist. “Principia” was first published in 1687, where Newton stated his natural laws of motion: First law: If there is no force on an object it will either have a constant velocity or be at rest.

Second law: Force is equal to the mass times acceleration. Third law: Static equilibrium, when a first body exerts a force on a second body, the second body simultaneously exerts an equal, opposite force on the first body.

17: Sketch of the inverted chain principle: (a) the chain is loaded, (b) the form is inverted and (c) the line of trust should lie wihin the cross section of the masonry. Image courtesy of Limitstate.com

SCIENCE AND ENGINEERING PAST 42

In fact, several people were working on how to calculate the stability and geometry of an arch structure (Leibnitz (1646 – 1716), Huygens (1629 – 1695), John Bernoulli (1667 – 1748)) but all scientists were at this time very secretive about their work, just like Brunelleschi did not want to explain the full principles of his concept for the Florence Cathedral dome structure. David Gregory (1659 – 1708), in 1697, asserted that if the shape of an inverted hanging chain lies within the masonry (thickness) of arch, then the arch would stand. He calls the inverted chain for the line of thrust – the line that represents the way the compression forces are carried to the supports. Gregory’s treatment of the problem was not perfect, but it suited the engineers’ need to tell if a structure would stand or fall. The theory of the thrust line was, like most of the other mayor science topics concerned with engineering, later widened and expanded by others. In 1846 W. H. Barlow (1812 – 1902) demonstrated that the thrust line can take infinite many shapes dependent on how the mortar between the masonry cracks. But Gregory’s safe theory – that if the line of the hanging change is within the shape of the masonry arch the structure is stable – is not far from our present plastic theories: if the engineer finds a way for the forces can be transported to the supports, then the structure also will find a way - but for masonry arches these are most likely not the same. 2

GALILEO AND THE IDEA OF STRESSES In Galileo’s “Discussion” from 1638 he described the stresses in a cantilevered beam. He acknowledged how a wood beam would crack at its top over the support when loaded, and he developed the idea of stresses that are distributed within the cross section of the beam – although Galileo did not find a correct stress distribution. He described how stresses are equal to the force distributed over an area – how single small fibres have little strength but when twisted together they made a strong robe. Galileo was concerned with the breaking strength of the material in order to determine the overall resistance of his beam, a similar approach used by present structural engineers. Galileo assumed a uniform, rectangular stress distribution and stated that the resistance of a wooden beam is equal to the absolute strength times the width times the depth squared, which is actually correct – but the way he derived his equation was wrong: Galileo assumed that when the cantilevered beam fractures it would turn about its lowest point, ergo all fibres in the cross section would be in tension. The resultant tension force must then be placed at half the height and Galileo’s mathematic and exterior physics are correct but the interior physics in the cross section wrong. The overall system Galileo set up is not in equilibrium, as the sum of transverse forces is not zero and leads to a horizontal pull at the support. Galileo’s approach was, however, not fully wrong either; he looked at how the beam actually failed and derived his theories from this. It shows similarities to the plastic theories to be perfected only

18: Galileo’s cantilever model as he imagined it in his “Discorsi e dimostrationi matematiche, Intorno à due nuove scienze, attenenti alla mecanica, & i movimenti locali”, 1638

SCIENCE AND ENGINEERING PAST

three centuries later. James Bernoulli (1654 – 1705) prove the linear relationship between the bending moment and the deformation with the newly developed calculus in 1691. Galileo never cared for the deformed shape of an element, he only cared if the structure was stable or not – deformations of a cathedral or a masonry bridge were too small to be noticed. In 1686 in France, Mariotte (1620 – 1684) was testing the theory of Galileo but could not achieve corresponding results. Mariotte suggests a linear, triangular stress distribution instead of Galileo’s uniform, which corresponds to Mariotte’s test results. Instead of Galileo’s factor of 1/2 Mariotte assumed 1/3. Parent (1666 – 1716), in 1713, realises that both Galileo and Mariotte have both assumed a horizontal pull at the support, something he could not find present. Parent dismissed the idea of the beam turning about the lower point and introduced a new theory; fibres at the top are extended and compressed at the bottom. This implies a new bending strength coefficient of 1/6. This theory satisfies the mechanical requirements of no pulling forces, but the theory did not apply to test results – for wood Mariotte’s factor of 1/3 seemed to fit best and Galileo’s factor of 1/2 suited stone test results. Bèlidor (1698 – 1761) published the first “building code” in 1729, in a form somewhat similar to Vitruvius with six books presenting and collecting the engineering knowledge available at

that time. Bèlidor offers no new work but collects the knowledge of, among others, Parent on beam strength. The books were the basis of practical design for almost 50 years until Coulomb published his scientific papers in 1773.

BERNOULLI, EULER AND DEFORMATION In 1741, 50 years after his aunt’s proof of linearity between bending and deformation, Daniel Bernoulli (1700 – 1782) simplified the mathematically exact formulas: by assuming only small deformations he could cross out many of the awkward terms in the equations. Young Bernoulli is best known for his work on fluid mechanics, but he also published his thoughts on the elastic beam: when an elastic beam is bent from its initial position energy is stored. This energy is recovered when the beam is unloaded and the beam returns to its initial position. In fact, nature will always respond by storing the least amount of energy. Finding the minimum energy a beam can store is a mathematical problem, it is the calculus of variations. Leonhard Euler (1707 – 1783) developed this new branch of calculus, and when Daniel Bernoulli in 1744 challenged Euler to solve the problem of the “elastica” he rose to the challenge. In Bernoulli’s elastica problem an elastic strip with a fixed length is thought to be held at both ends, having the start and end directions known as well. Bernoulli wants to know the shape of the strip having as little energy stored as possible. Euler derived the results for all possible shapes the strip could take, but he was particular interested in the

Galileo (1638) Mariotte (1686)

Resistance: 1/2∙b∙h∙σmax

Parent (1713 )

Resistance: 1/3∙b∙h∙σmax

Resistance: 1/6∙b∙h∙σmax

19: Sketch showing Galileo’s (top), Mariotte’s (middle) and Parent’s (bottom) theories for stress distribution in a cantilevered beam. Drawing by author

SCIENCE AND ENGINEERING PAST 20

solution with minimum energy stored – the half sine wave shape. This solution could only be obtained by the presence of a (compression) force, and Euler was able to calculate this load. A limiting value for the compression force in for instance a column was derived, and Euler proved the empiric formula for buckling loads found experimentally by Musschenbroek (1692 – 1761) a few years earlier. Charles Coulomb (1736 – 1806) was sent to Martinique by the French army shortly after his graduation from Mézières. After nine years of isolation from the scientific scene on the island, he returned to Paris in 1773. Coulomb had used Bèlidors book as a textbook at university, but was little aware of the theory of Parent. In Martinique Coulomb rediscovered Parents solution to the bending problem with a

coefficient of 1/6, but just as Parent, Coulomb could not match his theoretical results with his test results; Coulomb also finds the factor 1/2 best suited for stone and 1/3 best for wood. He published his work, which included great scientific contributions on other topics than the bending problem, on his return to Paris in 1773. Navier takes up Bèlidor’s book for revision in 1813 and continues to publish scientific papers afterwards. Claude Louis Marie Henri Navier (1785 – 1836) first attended the Ècole Polytechnique and afterwards the Ècole des Ponts et Chaussèes where he became a teacher after his graduation. One of his most influential works was published as lecture notes in 1826. The Parent/Coulomb theory was interpreted physically by Navier, referencing it to Hooke’s

20: Euler’s sketch of the possible solutions to Bernoulli’s “elastica” problem. The bottom left found his interest From Euler’s “Additamentum” published 1744

law; elastic deformations are recoverable and a structure kept within the elastic stress range will have no permanent set after unloading. Navier’s equations are linear, meaning that doubling the load results in doubling the deformations. The engineer approach of Navier was to calculate if the structure was safe. He was not specifically interested in the breaking strength of a beam, as Galileo was – Navier thought that an engineer should ensure safety of a structure under specific loads. Further, Navier made proof of hyperstatic structures using superposition and developed calculations methods for trusses, containing elements exposed to normal forces only. The theories of Navier were greatly expanded by Saint-Venant (1797 – 1886) in 1864.

PLASTICITY AND THE ULTIMATE STATE The elastic theory of Navier was used by many engineers in 1914, and still is today. At the time the world was booming with new steel structures. In Hungary, G. V. Kazinczy was performing tests on a steel beam fixed at both ends. When loading the beam, it behaved as predicted and the stresses reached yield level first at its ends. According to Navier’s theory, the

bending moment at the fixed supports should be twice that of the middle of the beam. Kazinczy noted that the beam still had spare capacity of resistance left after yielding had occurred at its ends. When further loaded, the beam soon started to yield at its middle as well. Great deformations were undergone before the beam collapsed in the end. The plastic properties of steel were perhaps not invented by Kazinczy, but his tests showed that the full capacity of the steel was far from exploit with Navier’s theory. Also in 1936 the Russian scientist A. A. Gvozdev (1887 – 1939) presented a paper that was not published outside the Soviet Republic until 1949. Gvozdev’s theory brings together three theories: • statical conditions – the internal forces must be in equilibrium, • yield conditions must be satisfied – the structural elements shall stay within the elastic stress range • geometric conditions – the structure must have a deformation mechanism at the state of collapse.

In 1864 James Clerk Maxwell (1831 – 1879) developed his Reciprocal Theorem, stating that if a known force is applied at point A of a linear elastic structure, as a consequence, a resulting deflection occurs at point B. If the known force is then applied in point B, the same deflection will occur at point A. In 1872 Betti expanded the theory that leads to an experimental way of solving statically hyperstatic structures. By the use of “indirect” model tests – smaller scale tests – solutions to the mathematical equations was provided and the behaviour of the real structure could be understood. A model that reflects the elastic properties of each cross section is made, but instead of loading the model, it is deformed and the consequent displacement elsewhere is measured. The entire heavy mathematic was avoided and the necessary structural analyses could be done on a model. 47

SCIENCE AND ENGINEERING PAST

The World War I held development of the plasticity theory back. In the 1920’s the theory won dissemination mainly in central Europe. In a conference for the Association for Bridge and Structural Engineering in Paris in 1932, eight papers were published on plasticity theory, mainly concerning the breaking strength of steel – bringing the concerns of Galileo on to the hyperstatic structures. In 1936 a second congress was held in Berlin against the background of a growing use of steel in industrial and large commercial buildings. The elastic theory of Navier ensured the engineer of a safe structure, but Kazincsy’s recent test results showed that the theory might be too safe – the plasticity theory could save material, weight and costs. The 1936 Berlin congress sought to establish the basic principles of plastic design of steel and reinforced concrete structures. One of the first steel building codes came in United Kingdom in the 1930’s and was in the beginning purely based on the elastic theories of Navier. After the 1936 Berlin conference, the British Steel Structures Research Committee started to include the plasticity theory. In 1948 the British Standard 449 (the use of structural steel in buildings) was revised permitting the plastic design.

The statical conditions arose from Galileo’s theories developed by investigating the cantilevered beam – the theory of how the stresses are distributed in a cross section. It is also founded in Newton’s thirds law. The theory was developed to include both internal equilibrium and the static boundary conditions by Parent and later Coulomb – Galileo wrongly assumed a pulling force at the support, something not present, and solved by Parent. The yield theory emerged from Navier and was further developed by St. Venant, von Mise and others. It seeks to ensure that the internal forces do not exceed a critical point. The geometrical conditions is concerned with the actual failure state of the structure, something first considered by by Galileo and later Coulomb who was concerned with the energy stored in a beam when deflected. The theory also included the work of

Kazinczy that showed that a steel beam critical loaded would create hinges at it’s yield points. From the three conditions Gvozdev proved three theorems, which still today is the basis of structural design: Uniqueness: if all three conditions are satisfied the exact resistance of the structure can be calculated. This is always the case for static determinable structures. Unsafe theorem (upper bound solution): by analysing only the geometric condition of a structure, it is possible to calculate the resistance for one breakage mechanism – but maybe not the most critical mechanism, therefore an unsafe approach. This is the basis of the crackline concrete theory; if the wrong crack lines are assumed the calculated bending moment

is underestimated. Safe theorem (lower bound solution): if the designing engineer can find one way the structure can carry a load, then the structure is safe. It is not certain that the most optimal distribution of stress is found, but if the designer can find a way for the applied load to go to the

foundation, then so can the structure. This relates back to the thrust line theorem of arch structures developed by Gregory 250 years before. 3

21: Graphic illustration of how all static problems can be solved. Drawing by Lars German Hagsten

SCIENCE AND ENGINEERING PAST

CODA CHAPTER TWO Even though not many of the people mentioned in chapter two actually build anything, their importance in the evolution of engineering is important. As described previously, engineering is not science â&#x20AC;&#x201C; it is applying and adapting science in a human world. Nevertheless it is important to know the history of the science engineers apply, and what the ideas sprung from. Gallileo was, for instance, interested in why a beam would fail. It was his determination to measure everything that led him to research how and when a cantilevered beam will fall down. Navier was on the other hand only interested in knowing whether or not a structure was stable, he did not care when or how it would fail. He lived in a time where the amount of material used did not matter that much. Only later, in the beginning of the twentieth century, scientists again started to investigate the failure of structures. They discovered that steel has an extra capacity after the point of yielding.

PAST 22: Drawing of Thomas Telfordâ&#x20AC;&#x2122;s proposal for a new London bridge, 1803 Image courtesy of ICE - Institution of Civil Engineers

STRUCTURAL ART PAST

Chapter three

Structural Art

Alberti and Palladios books made the rules of proportion available to laymen. Kings, Emperors or Bishops could now be their own architects. This might be the reason why the 19th century sometimes has been characterised as the age of the battle of the styles, where romantic Gothic was pitted against the classic tradition.

The bridge is shaped similar to typical stone masonry bridges, but the higher compression strength of iron makes the visuam translucent impression possible. A town developed with the bridge and it even took name after the bridge, today simply called Iron Bridge. The “Ironbridge” spans ca. 30 m with a series of semi-circular arches. 2

The industrial revolution in the first half of the 19th century was not in terms an ordinary revolution, it did not happen from one day to the other. With the industrial revolution came a new stronger and lightweight construction material: iron. Along with the “battle” between Gothic and Classicism, the structural Age of Iron had begun. 1 Already in 1779 the first cast-

Iron had previously existed as a construction material, but due to the relatively high price it was primarily used in smaller scale. During the industrial revolution the price of iron was decreasing as demand rose. Iron was mainly being used in the development of a new infrastructure for turnpike roads, canals and railways.

iron bridge was completed in Coalbrookdale, England, bridging the river Severn. Thomas Farnolls Pritchard (c.1723 – 1777) designed the characteristic arched shaped bridge and it was built by iron-founder Abraham Darby.

THOMAS TELFORD Thomas Telford (1757 – 1834) was born in an isolated community in southern Scotland. He was trained as a mason and while working in Edinburgh he started to study architecture books in his spare time. In 1778 he helped build a three-span masonry arch bridge near Langhold, Scotland. He left Scotland for London in 1782 but he kept a close relationship with the leaders of his home town community, which – after working his way up to become his own boss through several small scale projects – gave him the job as County Surveyor in Shrewsbury County in 1787. Beginning with Montford Bridge he was responsible for the design or reconstruction of over forty bridges in the county, along with being the architect of a number of churches and public buildings. In 1789 Telford was also appointed by the British Fisheries Society as a consultant to identify sites for harbours and piers, some of his early jobs involved construction of docks. Buildwas Bridge, opened in 1796, was the first iron bridge by Thomas Telford. The design showed Telfords determination to break with the conventional structural forms, the arch bridge, and a desire in Telford to make a more efficient and economical use of the material. Telford was in charge of the design of several cast iron bridges, including Bonar Bridge, which was considered an aesthetic triumph and was made in collaboration with ironmaster William Hazledine (1763 – 1840). Thomas Telford’s work for the Fisheries So-

ciety led to work for the Highland government. He carried out surveys recommending an improvement of the infrastructure, which, besides smaller upgrades on harbours and inland communication, included the construction of the Caledonian Canal. The construction of the canal was so extensive that it continued throughout his life, and it led to the invitation to become a “General Agent” at the Ellesmere Canal Company in 1793.

“Feeling in myself a stronger disposition for executing works of importance and magnitude than for details of house architecture I did not hesitate to accept their offer, and from that time directed my attention solely to Civil Engineering.” 3

He worked on several canal projects where he build aqueducts using hollow masonry pillars combined with a bridge structure in cast iron, most spectacularly at Pontcysyllte on the Ellesmere Canal completed in 1805. Though the Pontcysyllte aqueduct in total is over 300 m long, each bridge segment only spans 13,7 m. Telford moves away from the classic bridge design with semi-circular arches. Instead he lets each segment support each other horizontally with the spandrels, to avoid them from spreading. Telford’s ideas for long-span bridges were even more spectacular. For many years the government seriously considered his proposal for the replacement of the medieval London Bridge in

STRUCTURAL ART PAST 56

23: Elevation of the navigable aqueduct of Pont-y-Cyssylte for the Ellesmere Canal over the River Dee Image courtesy of ICE - Institution of Civil Engineers

STRUCTURAL ART PAST

1803, but in the end it was never constructed. Telford had proposed a 186 m cast iron arch bridge and though never realised the design brought him national attention. Despite that, it was his last proposal for a long spanning bridge constructed after the principle of an arch bridge. 4 In 1814 Telford began working on bridging the river Mersey at Runcorn Gap, where the river is 300 m wide. Most of the width of the river was used for navigation, so Telford only saw one option: a suspension bridge. This construction method was unknown in Britain at the time and was only experimentally used in the United States. Telford only knew of one suspension bridge, a footbridge called Wynch Bridge, crossing the river Tees near Middleton. The Wynch Bridge consisted of two common chains stretching across the river, upon which a footway of wooden boards was laid. Telford started to elaborate a series of experiments on the tenacity of wrought iron bars, with the purpose of using it for the Runcorn Gap project, but the project was stopped, as funding was not available. 5

Telford took forward his experiments and studies to his next large scale project: bridging a 180 m span over the Menai Straits in Wales. Here, Telford for the first time put the suspension bridge principle to use on a large scale. The Menai Bridge represents one of the first great triumphs in British civil engineering and all large-span bridges was built as suspension bridges for the following many years. When the bridge was completed in 1826 it was the longest spanning bridge in the world. Telford was not the only bridge builder in Britain at his time, but he was his opponents technically superior. For the Craigellachie Bridge, as for other shorter bridges, Telford used the circular form of masonry and stone bridges, combined with elements previously seen in wooden bridges (Palladio), but he stretched the design to appropriate the properties of cast iron. In a comparison of cast-iron bridges built between 1779 and 1871, Telford had eight bridges in top ten in their order of technical quality, with five of them still standing today. But Telford was not a great scientist; in fact he hardly ever performed any mathematical calculations of

24: Elevation of a steel bridge segment spanning 13,7 m for the aqueduct of Pont-y-Cyssylte Image courtesy of ICE - Institution of Civil Engineers

his design. He tested both elements and performance of his erected structures, and from that he developed a great intuition of how both the single element and how the structure functioned as one system. Still Telford was reputed to have limited or no knowledge of even geometry, and did not make use of calculus existing since the seventeenth century. The laws of nature developed by Newton and Galileo were neither used by Telford when designing, nor were his designs derived from mathematical formulations of geometry. Actually, Telford more or the less rejected the scientific discoveries and advances made at his time:

“[The scientific studies have] led to no one useful practical result, […] as to the thickness of archstones, side walls, and piers, the horizontal section or ground plan of the bridge, the manner of filling up its haunches, of forming the joints, of connecting it with the abutments, wing walls etc., we are still left in the dark. […] It was only [after] Newton had opened the path of true mechanical science, that […] arches attracted the attention of mathematicians. We are much inclined to doubt whether the greater part of their speculations have any value to the practical bridge builder.” 6

25: Pont-y-Cyssylte agueduct as it looked in 2009, the same year where the canal and the aqueduct were inscribed in UNESCO’s world heritage list. Image courtesy of UNESCO

STRUCTURAL ART PAST

A complete distrust in scientific was not the case though: for the Menai Bridge Telford took the advice of Davies Gilbert to increase the sag of the chain. Gilbert later published the mathematical theory he developed for the Menai Bridge, and Telford gave credit to Gilbert for influencing his design. 7 Meanwhile Telford did not forget his passion for architecture and beauty. Although declaring to leave architecture behind in 1793, when joining the Ellesmere Canal Company, he had an aesthetic intention in all his projects throughout his career. Telford’s intuition of beauty was ahead of his time and his writings did not describe old architectural subjects as proportion, symmetry and rhythm. The nineteenth century’s prediction of art and beauty was that it could only be pure if the form did not arise from natural laws or social necessity – needs. As philosophe Immanuel Kant (1724 – 1804) puts it:

“The necessary cannot be judged beautiful, but only right or consistent.” 8

than from images and formulations refined over centuries for masonry structures. 9

“He emphasized that the primary visual purpose of architecture was to express its loadcarrying function. For Telford, it was the laws of nature and the needs of society that gave stimulus to form, not preconceived aesthetic rules.” 10

The industrial revolution brought a new scale to architecture, a bigger and previously unseen scale. After the introduction of iron, structures would show how rational decisions based on natural laws communicate the artist’s aesthetic ideas and at the same time answer the needs of the society.

“[Telford was] the first modern engineer to show how a concern for aesthetics does not compromise technical quality but rather can improve it” 11

as it is described by David P. Billington.

In short, Kant believes a work of art is as useless as a tool is useful. Telford broke with this prediction. He wrote about the new engineering problems he faced, thinking about appearance and form of his structure in the landscape. Telford described how beauty should come internally from what is technical and economically possible, rather

26: (Previous page) Picture of the Menai Bridge as it looks in the twenty-first century Image courtesy of Anglesey Heritage

27: Painting by George Arnald of the Menai Bridge 1828 Image courtesy Jullian Simone Fine Arts Ltd.

STRUCTURAL ART PAST 28

ISAMBARD BRUNEL

flooded.

Isambard Kingdom Brunel (1806 – 1859) was the son of Mark Isambard Brunel (1769 – 1849), the creator of the worlds earliest mechanized production line for the British Royal Navy in Portsmouth. Isambard grew up with an inventive father and joined his father at The Thames Tunnel project soon after completing apprentice at a famous watchmaker. The tunnel project was a very daring idea, and Mark invented a new way of constructing a tunnel with a great iron framework housing the excavating workers, while bricklayers worked directly behind them. The shield was pushed forward using screw-jacks. The work was suspended after a series of accidents, the last one almost killing young Brunel, when the incomplete tunnel was

Isambard was sent to recover in Bristol by his parents. In 1829 Bristol held a competition to span a gorge at the river Avon. Brunel made four designs for a suspension bridge, and in total there wer twenty-two submissions. Thomas Telford, who was assigned to be the judge, rejected all twenty-two design. Telford then, at the time seventy-two years old, made his own design. He was afraid that his world record at Menai was to be broken, and Brunel objected sharply to the commision about Telford’s design.

28: Temple Meads Station in Bristol Image courtesy of Ben Brooksbank

The commision agreed with Brunel, and held a second competition which Brunel won. Brunel’s design was to be the longest span-

ning suspension bridge in the world and it was only completed after his death. Isambard Brunel was lucky to be born at the time in which Britainâ&#x20AC;&#x2122;s world dominance peaked, and he was living in the first industrial society. The contry was to construct their railway network and the merchants of Bristol were afraid to loose income to Liverpool, which has already been connected with a railroad line to Manchester. Plans to expand the LiverpoolManchester line further south to Birmingham and London were on the table. Isambard was appointed to survey the route for a new line between Bristol and London. For nine weeks Brunel worked for many hours a day on the back of a horse between London

and Bristol to find the ideal route. He thought the ideal route would be a direct and straight line, avoiding smaller towns. Slower branch lines, to be constructed later on, could then pick up the smaller towns. Brunel also deviated from the standard gauge, proposing to have the rails wider apart to allow for larger and heavier locomotives. A larger locomotive would have a lower centre of gravity and be more stable under high speeds. Brunel was chosen as the engineer to design the line, and he was in charge of all aspects of the railway. The new, broad gauge locomotive was built to his specifications, but the first ones were unreliable and underpowered. They were modified with help from a young superintendent, Daniel Gooch (1816 â&#x20AC;&#x201C; 1889), who 29

29: Paddington Station in London photographed in the middle of the nine-teenth century. Image courtesy of Ben Brooksbank

STRUCTURAL ART PAST

might have saved the Great Western Railway Company. Brunel was still in charge of designing all aspects of the railway, from bridges and tunnels to stations and other structures. An example of his skills are two impressive engineering accomplishments that he made on the Great Western Railway: the Box tunnel and London Paddington station. The Box tunnel took four years to complete and costed more than a hundred lives, but when completed it was the longest tunnel ever conceived. Paddington station was made with a glass roof supported by three sets of iron arches spanning 21 m, 31 m and 21 m, and all in all being 210 m long. Brunel mastered many variations in construction style; he made Egyptian-styled viaducts, castellated tunnel portals and Gothic bridges. For instance Temple Meads Station in Bristol was made from a gigantic wooden structure, Bath Station was designed on the basis of an Elizabethan country house and the Box Tunnel had a classic archway on a heroic scale. Brunel’s ingenious talent and vision did not only apply to land-based infrastructure; he went on to design steamships. While planning the Great Western Railway, during a casual dinner conversation he said:

“Why not make it longer, and have a steamboat go from Bristol to New York, and call it the Great Western?”

No steamship had been built big enough to carry the amount of coal needed to make the trip across the Atlantic, but Brunel realised that a bigger boat will have a more favourable power-to-weight ratio. The SS Great Western was completed to Brunel’s specifications in 1838, and on its maiden voyage it crossed the Atlantic in fourteen days. The great success led to the construction of an even greater steamboat, the SS Great Britain which in 1845 crossed the Atlantic in only seven days on its maiden voyage. The SS Great Western was the most revolutionary ship in history, as it was for the first time using a iron hull and with a propeller engine that was so well designed that modern propeller engines only are 5-10 % more efficient. Not everything Brunel touched became golden. Like the design of his first locomotives, several projects did not go as well as he hoped. He tried to realise a new idea with atmospheric traction, which suggested to use a vacuum that was created in a pipe running between the rails that would pull the trains of the South Devon Railway. The failure resulted in a massive loss for the shareholders. Also Brunel’s third great ship, the SS Great Eastern, caused big problems. His idea was to build a ship big enough to house enough coal to make the trip to the Far East and back without refilling, due to the lack of coal supply at the far end. The SS Great Eastern never made profit; it was too big and before it’s time, no one had ever tried to control a machine of that size before. The project encountered several problems and ended bankrupting the Eastern Steam Naviga-

30: (Previous page) Painting of the west entrance of the Box Tunnel 1846 Image courtesy of Brunel 200

31: isambard Brunel’s Saltash Bridge over the Tamar River near Plymoth during construction in 1858 Image from William Humbler’s “A Complete Treatise on Cast and Wrought Iron Bridge Construction”

STRUCTURAL ART PAST

tion Company, leaving Brunel to finance the project himself. Brunel had invested so much in the ship – morally, intellectually and financially – that an explosion on board of the maiden voyage that killed five men, probably hastened his dead a week later. 12 Brunel was the inventor of both machines and structures. Where most of his structures still stand and are still being used today, none of his machines are no longer in use. Together with Telford’s works, Brunels structures have become the symbols of their age – the last great era of British world dominance which climaxed in the 1850’s. The self confidence in the British structural pioneers Telford and Brunel characterise the British society of the time, leading the industrial revolution and dominating the world. Telford was, in contrast to Brunel, a servant of the state; he was paid a fee from the gouvernment to produce a new design almost weekly. Brunel was the opposite: a private entrepreneur, who designed entire networks. Brunel usually invested his own money in the ventures for which he was the designer. 13 Brunel’s biographer L.T.C. Rolt (1910 – 1974) described the generation of great British structural pioneers by stating:

“He and his generation bequeathed a sum of knowledge which, like his great ship, had become too large and too complicated to be mastered any longer by one mind.” 14

The first world exhibition was held in Londonâ&#x20AC;&#x2122;s Hyde Park in 1851. The whole event was housed under a great glass roof, which later gave name to the area: Crystal Palace. Landscape gardener Joseph Paxton, who earlier had designed smaller green house constructions in iron, designed the Great Exhibition Building. Although he did not enter the competition one year earlier, he realised that the design proposed by the jury could not be completed within the time frame. He therefore submitted his ridge-and-furrow roof system, which was particularly notable for the use of standard, mass-produced construction elements. The buildingâ&#x20AC;&#x2122;s footprint was enormous for the time, measuring 563 m in length and 124 m in width. The central nave was 19,2 m high and spanned 21,9 m. The interior structure had 3.300 columns supporting the cast-iron trusses. 15 Though the 1851 exhibition was the first World Exhibition, France had been holding a national Exhibition to promote the improvements in progressive agriculture and in technology since 1798. In 1889 when the world fair returned to Paris, a distinctive monument was suggested to commemorate the centenary celebration of the French Revolution. 32

32: Painting of the Crystal Palace 1851. Image courtesy unknown

STRUCTURAL ART PAST

GUSTAVE EIFFEL “The English engineers have almost entirely bypassed calculations and they fix dimensions of their members by trial and error and by experiments [...] and small scale models.”

The Crystal Palace construction was in Eiffel’s eyes a primitive structure with semi-circular wooden arches and standardised steel pieces bolted together. Eiffel continious:

“The English went ahead of us in their practice, but we have had the honor, in France, to surpass them by far in the theory and to create methods which opened up a sure path to progress, disengaged from all empiricism.”

Gustave Eiffel (1832 – 1923) was born in Dijon. He attended the École Centrale des Arts et Manufactures where he after his graduation expected to take over his uncle’s paint factory. But when graduating in 1855 - the year of the first World Exhibition in France - family relations had soured so Eiffel began a career in engineering. Eiffel took job for Charles Nepveu (1791 – 1871), an engineer and contractor who constructed mainly steel structures, machines and many smaller items. During his first years Eiffel got to work for a railroad company on a short steel bridge spanning 22 m. At the age of twenty-six Nepveu promoted Eiffel to become

the supervisor for the longest bridge in France at that time, a 500 m metal railway bridge over the river Garonne near Bordeaux. Here Eiffel learned of pier foundations constructed with the use of compressed air caissons, something he used frequently in his later projects. After a few jobs in the south of France, including a railway station in Toulouse and several bridges, Eiffel decided to start his own company in 1864. In the beginning he had limited financial resources, but he had good technical knowledge and contacts within the railway companies. After his first smaller projects, Eiffel was soon awarded the contract to construct two viaducts in central France. Here Eiffel launched a prefabricated steel deck from the riverbank, a method to become common later. His business grew steadily as he received work on several more bridges in France, and even began to do smaller works in South America, for instance a prefabricated church in Chile which is still standing today. In 1875 the company had two new big commissions in Europe: the central station in Budapest, Hungary, and a large viaduct near Porto, Portugal. For the competition of the Maria Pia viaduct Eiffel, together with his engineer Theophile Seyrig (1843 – 1923), developed the concept of a large spanning arch bridge of 160 m. This allowed Eiffel to avoid the expensive scaffolding across the river and therefore beating his competitors in both price and design. The competition was held by the Royal Portuguese Railway in 1875. Eigth different designs was submitted and they all had to be fixed in

price. The two simplest forms were also priced lowest, both arch bridges. The design of Eiffel’s company was 31 percent lower than the other, confirming the fact that Eiffel, at age forty-three, was the leading bridge designer in Europe. At the same time he was responsible for several other bridges in Portugal and Spain and he designed the City of Paris’ pavilion for the World Exhibition in 1878. The success of the Maria Pia viaduct brought Eiffel a contract to construct similar Garabit Viaduct over the Truyére gorge. The bridge, completed in 1884, had a visual impression of lightweight which was incomparable to any bridge of its time. It spanned 160 m and led way to the construction of the 300 m high tower in Paris, which would be designed on the same principles, calculations and construction methods, and by the same team of technicians.

Brunel by resting much more on theory than on empiricism. He was convinced that theory:

“permits exact calculations [from which come] structures which are much lighter and at the same time are stronger than those built earlier. [...] At the start [of modern structural engineering], designers multiplied the number of loadcarrying members and thus complicated their 33

For both the Pia Maria and the Garabit the arches were designed with hinges at the suports. Eiffel’s engineer, Seyrig, explained how the arch shape was:

“at the same time the most graceful and the best suited to the load-carrying necessities. [By chosing the hinged supports] the calculations were markedly simplified and all those who have once made complete arch calculations know that such an advantage is not to be disdained.” 16

Eiffel differed from his predecessors Telford and

33: Garabit Viaduct Image courtesy of Johns Hopkins University Department of Civil Engineering

STRUCTURAL ART PAST

structural systems; today, on the other hand, there is the tendency to simplify them as much as possible, because the more a system is simple, the more one is sure of how the loads will be carried.â&#x20AC;? 17

Eiffelâ&#x20AC;&#x2122;s team of engineers proved their talent when designing and constructing the structural framework for the Statue of Liberty in New York in 1884. Eiffel showed his ingenious talent when designing a rotating steel dome of the Nice observatory, making the dome float on a circular tank filled with salt water to prevent freezing. 34

34: Painting of the Ponte Maria Pia in Portugal, 1877 Image courtesy of Biblioteca Nacional de Portugal

For the World Exhibition in 1889 Eiffel wanted to go beyond his previous work as an engineer and he spent months promoting the idea of a tower among official circles, and a competition was launched in 1885. As known, Eiffel won the commission but he had to provide almost half of the financing for the construction himself. He made a deal with the City of Paris that he would receive the rights to operate the tower for 20 years in exchange for providing funds. The Parisian habitants were from the begging not very pleased with the aesthetics of the tower; the leading Parisian newspaper Le Temps wrote in 1887:

“This offence to French good taste, […] building this monstrous Eiffel Tower in the heart of our capital”.

“The principle of architectural beauty is that the essential lines of a construction be determined by a perfect appropiateness to its use.”

The article described the tower design as baroque, mercantile imaginations of a machine builder, a gigantic black factory chimney. Eiffel responded:

Construction of the Eiffel Tower only took twenty-two months and was completed in time for the 1889 World Exhibition. The work was a masterpiece of precision, efficiency and speed. Each prefabricated part had a precision of one-tenth of a millimetre and was produced at Eiffel’s factory located just outside Paris. The tower was a triumph for Eiffel himself, for engineering as it was standing as a masterpiece of several decades of daring experiments and finally for the French. It ensured Eiffel world fame.

“Can one think that because we are engineers, beauty does not preoccupy us or that we do not try to build beautiful, as well as solid and long lasting structures? Aren’t the genuine functions of strength always in keeping with unwritten conditions of harmony? Yes, I affirm that curves of the four corners of the monument, as mathematical calculations determine them […] will give a great impression of force and beauty, because they will visible reflect the strength of the overall conception.”

But the tower was in many ways provoking. It was unseen before; it was only needed to show the world France’s technical supremacy, and unlike Telfords and Brunels work it had no real use other than a great view from the top. Eiffel continued his argumentation for his tower:

Eiffel was less lucky with the construction of the Panama Canal, where he wrongly assumed it was possible to build a level canal through the Panama rainforest, as his company previously had done with the Suez Canal. Eiffel was later forced to change the plans and to build ten huge locks, including massive, sliding steel doors with a budget fifteen times higher than that of the tower. The climate was difficult to work in, and the medical skills of the time could not efficiently treat malaria and yellow fewer outburst. This made the work progress so slow that the Panama Canal Company in the end went bankrupt, involving Eiffel in one of the biggest financial scandals of his time. After the fiasco, Eiffel began a second career as a theoretical scientist in three major fields:

STRUCTURAL ART PAST 76

35: The construction of the Eiffel Tower from 1887 to 1889 Image courtesy gustaveeiffel.com

STRUCTURAL ART PAST

meteorology, radiotelegraphy and aerodynamics. He founded all his laboratories himself and was actively publishing his scientific results until his death at the age of ninety-one. His wind tunnel constructed at Rue Boileau in Paris is still in use today. 18 Even though Eiffel and the World Exhibitions brought iron structures to the architecture scene, architects was in the beginning very nervous about exposing the material in their buildings. Iron was seen as the dirty symbol of industrialisation - the material of the working class, the material used for machines. Architects of the time began to use iron framework structures, but it was in the beginning cladded with a traditional stone faรงade.

STRUCTURAL ART PAST

Towards the end of the 18th century Romanticism originated throughout Europa. Partly a reaction to the Industrial Revolution, it was also a revolt against aristocratic social and political norms of the Age of Enlightenment and a reaction against the scientific rationalization of nature. The French mob had started a revolution for democracy, with the storm of the Bastille, already in 1789. The path for democracy was set on pause when Napoleon I (1769 – 1821) in 1799 staged a coup and installed himself as the Emperor followed by a re-emplacement of the monarchy in 1814. The Romantic Movement received greater momentum after Napoleon’s invasion of central Europe, and gave rise to modern nationalism: the uniqueness of all people. The Romantic Movement became a European wide movement especially after the Industrial Revolution started, when many people revolted against the mechanisation of labour and society in their writings. Louis XVIII (1755 – 1824) reined France from 1814 to 1824 and was followed by his younger brother Charles X (1757 – 1836) who ruled until 1830. Romanticism continued to grow in reaction to the effects of the social transformation caused by the Revolution. The Romantic Movement, first starting as a negative reaction to

37: Panoramic view over Boulevard Haussmann, 2013 Image by author

the French export of Enlightenment theories as universal scientific rules governing all humans, emphasized the uniqueness of the human being and the mysterious nature of the human mind. The Romantic revolution swept across France in the 1830’s, and frustration among the labouring classes rose. It reached its climax in 1848 when workers broke out in an armed uprising against the anti-labour and anti-democratic policies. Napoleon III (1808 – 1873) was elected as president in December 1848. The rising industrialisation of France in the period had made workers seek towards the capital. At this time Paris had the same structure as in the Middle Ages, narrow, interweaving streets and cramped buildings that resulted in unhealthy conditions for the inhabitants. Napoleon III applied Baron Haussmann (1809 – 1891) to reform the city, and from 1853 to mainly 1870 the city was almost completely modernised. Napoléon III was, with good reason, afraid of rebellion from the Paris habitants. The medieval Paris was smelly and dirty and had narrow interweaving streets that were almost impossible to control for Napoléon’s large artillery. Haussmann demolished entire quarters and on the ashes he formed a network of broad avenues and boulevards crossing the whole city. The famous Haussmann houses became the city’s new skin, and with the also newly built sewage network it became a model for other cities for a long time.

STRUCTURAL ART PAST

VIOLLET-LE-DUC

the July Revolution in 1830.

Wereas Gustave Eiffel was the famous French engineer Eugène Emmanuel Viollet-le-Duc (1814 – 1879) was certainly the famous architect of the 19th century. Born in Paris in a prominent artistic and literary family, Viollet-leDuc received much of his education from his uncle, who was described as a painter in the morning, and a scholar in the evening by architectural historian Sir John Summerson (1904 – 1992). Viollet-le-Duc came of the generation living in the midst of the Romantic revolution in 1830’s France. He refused to attend the École des Beaux-Arts as he was afraid of being swallowed by its academic doctrine. He was trendy philosophically in his early years, being against the monarchy, opposing power to the Catholic Church and rebellious building barricades in

Viollet-le-Duc’s drawing talent was prodigious and with his fascination for Romanticism and French history, especially the Middle Ages, he spent endless hours capturing France’s architectural heritage. In 1834 he won a drawing competition and therefore he was able to travel two years in Italy on the price money. From his journey he brought back hundreds of drawings in a pictorial essay trying to bring history back to life. Instead of attending École des Beaux-Arts Viollet-le-Duc opted for a more direct practical experience in architectural offices, working for two Parisian offices before leaving for Italy. On his return he was appointed smaller commissions for restoration works on churches. His breakthrough came in 1840 when he was entrusted to restore the church of La madeleine at Vézelay. The work included both archaeological judgement as well as structural stability issues, as the complete vault in the narthex collapsed during the first years of restoration. The narthex vault would have taken the rest of the church with it, if Viollet-le-Duc had not already restored the other main parts of the vault. The completed job was a great success for Viollet-le-Duc, giving him both the insight on practical abilities as well as how structural issues of historical buildings could be solved. He had learned to put himself in the position of the original builder who had constructed the building in the first place. After the success of Vézelay, Viollet-le-Duc

STRUCTURAL ART PAST

won the prestigious commission to restore the Notre-Dame in Paris in 1844. Throughout the process not a single stone was left untouched, and Viollet-le-Duc even added the famous fiftysix demonic gargoyles circling the upper gallery. The restoration work sealed his reputation as Gothic Revivalist and it would provide the footing for an entire architectural revolution. With the revolution in 1848 and Napoleon III, Viollet-le-Duc widened his architectural influence as he was appointed in a position of the administration of historical monuments

service. With his new position he found himself head of a battalion of architects working all over France, from which his ideas could spread. He even expanded his control further by transforming the spiritual centre of architecture at the time, the Ă&#x2030;cole des Beaux-Arts. Around the same time Viollet-le-Duc wanted to change the public view of him as the Gothic Revivalist, finally turning his attention to modern iron. His famous series of visionary drawings of iron projects was designed in a short period between 1864 and 1868. His sketched

projects did not represent usual iron buildings, stations, market halls or exposition buildings his attention was instead focused on great assembly spaces for the public as concert halls, town halls etc. His architectural theory was largely based on finding the ideal form for specific materials, and using these forms to create buildings. His writings was targeted on the idea that materials should be used ‘honestly’. He believed that the outward appearance of a building should reflect the rational construction of the building. In Entretiens sur l’architecture, Viollet-le-Duc praised the Greek temple for its rational representation of its construction. For him

“one of only two supremely eminent theorist in the history of architecture”, 19 the other one being Alberti. Viollet-le-Duc’s greatest legacy was perhaps in his writings, and he is considere to be one of the first theorist of modern architecture. His writings and sketches of iron structures laid the basic foundation for the following Art Nouveau style.

“Greek architecture served as a model for the correspondence of structure and appearance”.

His writings were to become hugely influential to the many following generations of architects, Frank Lloyd Wright (1867 – 1959) described Viollet-le-Duc’s Dictionnaire raisonné de l’Architecture Francaise du XIe au XVIe siècle as

“the only really sensible book on architecture in the world”

and Sir John Summerson in the 1940’s claiming Viollet-le-Duc to be:

STRUCTURAL ART PAST

CODA CHAPTER THREE The industrial revolution separated the engineer from the architect. A whole new field of construction had arisen: infrastructure. Industrialisation meant railways, including bridges and stations, factories. Bridges, viaducts and train stations’ large spanning roof structures were not considered architecture. It was looked down upon from art critics and architects as not sophisticated – a dirty product of industrialisation. Exposed steel visually reflected factories and the working class. The engineers tried their best to implement aestheticism in their structures. They invented their own style; “Structural Art” as David P. Billington names it. Despite the, at the time, remote locations, engineers sought to achieve beauty in their structures. The big portal entrance to the Box Tunnel is a king worthy, and one cannot image such level of detail being employed on a tunnel project today. A great - probably the most important - part of the key to succeeding was the economy of a project. Together with the birth of the dedicated engineer came the need for optimisation. As many of the first engineers also functioned as contractors, their design had to fulfil the functional need by using the minimum amount of material. Great creativity was needed to be able to out win your competitors. New construction techniques were invented and new structural

methods applied. The early English engineers clearly showed that they possessed an idea of forces, despite their lack of scientific education. Telford’s “invention” of building a suspension bridge in large scale was truly revolutionary, and stands as one of the greatest leap forward in engineering history. Brunel’s Saltash Bridge shows how he knew of compression and tension forces and shaped the bridge to reflect the forces, with a big round tube that could resist the compressive forces without buckling. With Eiffel the engineer for the first time started to incorporate the scientific research on structures into large scale constructions. But with the calculating engineer everything became more complex; the whole thing became too big for one single mind to master. Both Eiffel, but also Brunel, had specialised engineers working for them. But none of the great engineers in the nineteenth century were recognised for their aesthetic capabilities. They were successful in their field, which mainly concerned infrastructural projects. Viollet-le-Duc was very different from the engineers, but I chose to include him as he had a significant influence on the generations that followed him. He did not built much on his own, mostly refurbishment projects, but through his writings he showed the architectural world of

the time, that incorporation of steel structures in architecture could create a whole new world for architects. He is in that sense, a founding father of architectural engineering. He leads us to the next big step that influenced the engineering history; the Modern movement.

THE MODERN CRAFTSMAN PAST

Chapter four

The modern craftsman

By the turn of the 20 century, the idea of one Master Builder for a project was extinct. A clear distinction between craftsman, architect and engineer has been slowly emerging since Mignot told the Italian workers to focus on the proportional principles of Gothic in the year 1400. Now, 500 years later, industrial revolution, colonialism and new materials had opened the world of construction. Specialised schools for craftsmen, architects and engineers were now the only way into the field of construction. Each branch was now being specialised within their own field.

To expose iron had begun in the precedent art nouveau style (c. 1890 – c. 1905). Modernism started to expose not only iron but also the newly invented reinforced concrete at the turn of the 20th century. Mixing aggregate, cement and water had already existed at the time of the Romans, but in the late 19th century concrete was enhanced with iron by the creation of Le Béton Armé.

Craftsmen were becoming a tool, being reduced to the client’s construction machine. The architects thought of themselves as more than other artists like painters or sculptors; they were more valuable in the eyes of society, as the architects considered themselves the designers of the useful. Engineers saw no limit to their work; everything was possible with the

François Hennebique (1842 – 1921) used concrete for the first time in 1879 for a small house project in Belgium. His job was to fireproof the structural metal frame, and his solution was to cover the beams with concrete. He realised that his invention could be used to create a composite beam, in which the concrete is mainly put under compression stresses and the iron

new materials and only the sky was the limit.

FRANÇOIS HENNEBIQUE

41: Drawings of Hennebique’s patented system of reinforced concrete beams and columns Images from Radford’s Cyclopedia of Construction by William A. Radford (1909)

THE MODERN CRAFTSMAN PAST

Auguste Perret (1874 – 1954) was a prominent French architect and builder, who together with his two brothers, carried on their father’s construction company. Perret was in full control of everything from concept to finished details due to the fact that their business was a construction company. He designed churches, theatres, artist’s studios, museums, industrial warehouses, and large-scale urban development projects. Perret was one of the leading architects in the beginning of the 20th century and was to bring concrete as a material to the eyes of the public. Giving full architectural expression to the reinforced concrete structure and by handling the details with the same care as fine woods or stone, he moved the material from mainly being used in industrial buildings to a highly refined solution for structural design. Perret had attended the École des Beaux-Arts, at that time still highly influenced by the structural rationalist theories of Viollet-le-Duc. Perret balanced his architectural style in-between the Art nouveau and classicism of Viollet-le-Duc and the modernist style rising in the beginning of the 20th century. Perret’s first reinforcedconcrete frame building was an apartment building constructed in 25 bis Rue Franklin in Paris in 1904. For the structural system Perret used the Hennebique system, clearly visible in the façade emphasised by the ceramic tile pattern. The grand apartment building became a demonstration of the expressive potential of “béton armé” – but still with the emerging modernism it was soon to be used in different kind of buildings. Reinforced concrete had become a stylistically formative construction material. 1

42: (Opposite page) Model of the Rue Franklin apartment exhibited at Cité de l’architecture & du patrimoine. Photo by author 43: (Above) The front façade of the Rue Franklin. Image courtesy of Charlie Brigante

THE MODERN CRAFTSMAN PAST

takes up tension forces. He further developed his concept so it contained continuous iron bars, which could take up the tension forces when laid in the right position. His ingenious invention proved to be both fire-resistant and favourable under seismic loads. Hennebique completed an apprenticeship as a stonemason and set up his own company at the age of twenty-five. In 1894 he took out a patent for reinforced concrete system that used iron stirrups. Reinforced concrete was not the invention of Hennebique alone, it had three main sources in the late 19th century. The French gardener Joseph Monier (1823 – 1906) strengthened thin concrete vessels (containers) with an iron wire mesh already in 1867. German bridge designers wanted to reduce labour costs by producing artificial stones on site. They reinvented Roman concrete, as to pore the concrete on site was cheaper than to carry natural stones to the site. The Hennebique system consisted of thin concrete slabs, with a mesh-reinforcement like Monier’s, and was supported by the concrete beams described above. Due to the relatively low weight of concrete it became possible to realise higher and cheaper constructions. Labour cost was at the same time reduced, as concrete could be fabricated directly on the building site. Simultaneously the automobile factories’ assembly line was invented – work was brought to the worker and the distance formerly walked to obtain raw material was, at least for the automobile factories, eliminated.

Hennebique’s background as a craftsman might have given him an advantage as he was developing the concept of the reinforced concrete beams; to maximise the utilisation of the iron bars, which are the most expensive part, they should be placed as close to the edge as possible, but to keep the structure fire resistant at the same time, a concrete cover around them had to be ensured. When poring the concrete he therefore needed to keep them in the right position. The Hennebique system prescribed a series of stirrups, which not only kept the iron bars in the right place but also connected the bars with the concrete part under compression. Structurally, the iron stirrups therefore solved the problem of shear stress between the compression and tension areas. It was a solution made from a stonemason’s trick, a builder’s skill combined with the – at that time – abstract technique of a composite material. Hennebique used his patent to set up a multinational business strategy, in which local licensers promoted his system worldwide. In 1905, only eleven years after he took out his patent, his network included several hundred construction businesses, while dominating the market for concrete constructions. The key to the great success was that these licensed contractors did not only have access to the Hennebique system, they were also in direct contact with the specialised concrete engineers in the head office in Paris. The head office guaranteed all projects using the original Hennebique system. As a result of this popularity it was easier for the contractors to convince

clients around the world to use this new material. Instructions for licenses and quality control helped Hennebique to spread his system and prevent damage and catastrophes. The in-house training for the engineers who made up the firm were another important factor that allowed Hennebique’s company to expand to an almost unprecedented extend within the building industry. The company acquired the specialist knowledge to work with reinforced concrete at the same time that the material itself was being invented. The majority of new engineers were recruited straight from the Ècole Centrale de Paris, which ironically was an institution teaching mainly metal structures at the time and not concrete. The engineers would start in the central office in Paris, for then to be sent out in the world to work in the local offices. The Hennebique system was a rigid system, consisting of beams and columns poured together. All elements were based on a catalogue-like system, like the steel tables known today, and Hennebique distributed the basic formulas, tables and graphs. Hennebique made the effort to promote this new material and to take on business from steel construction systems. Hennebique travelled to conferences and distributed images in both hand-drawn and photographic pictures, showing parts of the earliest contractors commission. He propagated his invention through his magazine Le Béton Armé. 2, 3

THE MODERN CRAFTSMAN PAST

CODA CHAPTER FOUR What did the Modern movement mean for the architect, the engineer or the craftsman? For the craftsman it included to move towards standardisation and a higher level of pre-fabrication. Reinforced concrete was invented almost simultaneous at different places; Hennebique saw the advantage of covering steel with concrete for fire protection and further developed the principle; Monier was searching for something for the concrete to both stick to and keep it in place while drying for his vessels; the German bridge builders tried to reduce labour cost by producing the material on site and started to reuse concrete that had been forgotten since the Roman era. Hennebique was also the creator of one of the first world wide construction corporations. Reinforced concrete was replacing steel as a cheaper alternative. But it was a direct replacement; a house would normally still consist of a framework with a grid of beams and a set of columns. Concrete had another advantage in addition to the lower cost; it was Modern. Architects loved the new material; it did not reflect the dirty fabrics. Seen from today the popularity can seem weird, but as a material reinforced concrete was the zeitgeist of the time. Reinforced concreteâ&#x20AC;&#x2122;s popularity has to a great extend been ruined by the extensive construction of apartment blocks in the 1960s and 70s.

THE MODERN CRAFTSMAN PAST

44: Frank Lloyd Wrightâ&#x20AC;&#x2122;s Fallingwater house designed in 1935 Image courtesy of Figuura

MODERN ARCHITECTS PAST

Chapter five

Modern Architects

The German poet Heinrich Heine (1797 – 1856) saw his chance to escape German censorship and settle in Paris with the July Revolution in 1830. In 1843 Heine describes his “real world experience” of the newly industrialised society he was living in:

“What changes there must now be in the way we view things, in our ideas! Even elementary terms such as time and space have become intermediate. Railways are destroying space, and all we are left with is time. In four and a half hours you can now get to Orléans, and in the same amount of time to Rouen as well. What is going to happen when the lines have been extended to Belgium and Germany and linked up with the railways there! To me it is as if the mountains and forests were getting closer to Paris. I can already smell the scent of German linden trees; the North Sea is breaking in front of my door.” 1

The world was being connected. Old traditions were broken. The new technological possibilities were used to develop a new modern style. Modernism was most of all a breakaway from the past. Form and shape should be nothing like the past and the architectural language completely different. A building’s shape should be as simple as possible, and with no symbolic meanings what so ever. Modernism broke with the traditional precedents styles as Classicism or Gothic and at the same time the intention of an international style was created – a common architectural idiom for all people. The key design aspect in modernism was function. A building had to be obligatory functional, and all unnecessary decoration and ornaments must be left out. Only clean and simple shapes were acceptable. It was a design language inspired by industrialisation, symbolised through clean forms and new materials. 2

45 45: Interior of Frank Lloyd Wrightâ&#x20AC;&#x2122;s Frederick C. Robie house in Chicago from 1908. Wright designed everything visible on this photograph. Image courtesy of The Library of Congress

100

MODERN ARCHITECTS PAST

FRANK LLOYD WRIGHT Frank Lloyd Wright (1867 – 1959) was perhaps most of all an architect of the interior. He was a functionalist and one of the most influential modernist architects. He grew up in Wisconsin and his education as an architect took place as an apprenticeship at Louis Sullivan’s (1856 – 1924) office. Sullivan believed in the development of an American architecture and rejected the Beaux-Arts classicism. Wright later came to realise Sullivan’s idea, and became the most significant American architect. In 1893, after completing his education, Wright’s first achievement towards the establishment of an American architecture came with designs for the archetypical American house: the single-family suburban house. In the beginning Wright’s work was focused on this symbol of the American dream, and his designs often included furnishing, lightning and even the design of stained-glass windows. One talent of Frank Lloyd Wright was his ability to design “the space within” as he called it. Starting around the turn of the 20th century, Wright was opening rooms to each other in a way that was uncommon previously. Rooms flowed from one to the next and he suppressed corners between them or dividers so that a wall could run continuous from one space to the next. The floor spaces were not wide open though, he often put a centerpiece or a spatial change in his designs – this way the interior flow went around. The dynamic of the interpenetrating spaces produced an open, mul-

tifunctional interior, which Wright connected with the exterior by horizontal window bands at eye-height – something unseen before in Western architecture and allegedly Wright was inspired by Japanese architecture. The integration of the surrounding nature was one of Wright’s concepts, linking context and interior order to human occupation and experience. Another of Wright’s principles was the honesty of materials. Wright engaged new materials with almost every design, yet reinforced concrete proved to be the most consistently challenging material for him. He was critical to what he described as concrete’s lack of order. It was a material with the ability to shape in any form, and Wright therefore found it difficult to find the “nature” of the material and to determine its appropriate use. His solution was to develop a concrete block system; something that he meant would give character to the previously formless material. The Edgar Kaufmann House in Mill Run, Pennsylvania, from 1938, called Fallingwater, exemplified Wright’s belief that architecture is born of its place, the context. He rejects the idea that architecture can be a product of an “International Style”. Fallingwater is believed to be one of Wright’s greatest “natural” houses, a place where people can truly be at home in nature. It is the vision of his former master Sullivan, who thought that the American architecture must depend on variations on the local climate, landscape, construction methods and materials available. 3

46: (Opposite page) The “Great Workroom” in Frank Lloyd Wright’s Johnson Wax headquarters, 1939 Image courtesy of Jack E. Boucher

101

MODERN ARCHITECTS PAST 47

102

47: The big atrium in Frank Lloyd Wrightâ&#x20AC;&#x2122;s Solomon R. Guggenheim Museum completed in 1959 Image courtesy of David Heald

48: The skylight at the Guggenheim Museum Image courtesy of David Heald

103

MODERN ARCHITECTS PAST

MIES VAN DER ROHE Less is more. Ludwig Mies van der Rohe’s (1886 – 1969) contribution to architecture was simplicity and extreme clarity. Soon after his death in 1969 Mies was branded as the father of all those “glass and steel” buildings, which for the critics stood as a symbol for the anonymity of the late 20th century urban landscape. Born in Aachen, Germany, Mies van der Rohe was the son of a stonemason. He never went to architecture school but was the apprentice to German architect and industrial designer Peter Behrens. After the First World War Mies established a small practice in Berlin, but immigrated to the United States, both for work and to help developing the new architectural school at the Illinois Institute of technology, in 1938. Mies was inspired to develop the new modern style, which he thought could represent the modern times just as classicism or Gothic did for their own eras. The forms created by Mies were the symbols of modernism. Using straight lines and with no ornamentation, he was characterised as a formalist by his critics. Mies was perhaps best known for his tall steel and glass buildings. One of his first known buildings was the 860-880 Lake Shore Drive Apartments in Chicago, being the first steel-frame apartment building in the world. He was known to be willing to explore all possible solutions and thinking about which to be the right one for a long time. For the Chicago

104

apartments, Mies explains the reason to put none-structural I-profiles as mullions on the outside of the façade skin in Architectural Forum’s November 1952 edition:

“First I am going to tell you the real reason for those mullions, and then I am going to tell you a good reason by itself. It was very important to preserve and extend the rhythm which the mullions set up on the rest of the building. We looked at it on the model without the steel section [I-profiles] attached to the corner columns and it did not look right. That is the real reason. Now the other reason is that the steel section was needed to stiffen the plate which covers the corner columns so this plate would not ripple, and also we needed it for the strength when the sections were hoisted into place. No, of course, that’s a very good reason – but the other one is the real reason.”

In another interview he also explained how the vertical steel profiles in the façade reflect the inner structure and therefore give truth to the aesthetics of the building. The idea of structural “truth” in a building aligns with the aesthetic principles of the international style promoted by Mies. The use of steel framework in his tall buildings gave him advantage of a minimal structure, and at the same time he strived for the freedom of the open space. Mies called his buildings “Skin and bones”, where he sought a rational structural approach. He had no problem in showing the structural steel elements in the façade.

49: (Opposite page) The exterior steel profiles on the 860-880 Lake Shore Drive Apartments in Chicago, 1951 Image courtesy unknown

105

MODERN ARCHITECTS PAST 106

50: Farnsworth house in Illinois, completed in 1951 Image courtesy Sandra Cohen-Rose and Colin Rose

107

MODERN ARCHITECTS PAST

While Mies was designing the Lake Shore Drive Apartments he worked on a much smaller-scale project, the Farnsworth House in Plano, Illinois, which was completed in 1951. The minimalistic house was planned as a weekend retreat, the site was located on a known floodplain of the Fox River. The pavilion, totally glazed, is raised 183 cm above the floodplain and surrounded by forest and smaller vegetation. In this project, Mies explored the relationship between people, shelter and nature. The house, with an asymetrically placed central core of bathrooms and linear kitchen, has become the standard by which all minimal hous51

108

51: Mies chose the same I-profiles everywhere Image courtesy Dallas Center of Architecture

ing is judged. Mies van der Rohe once told The New York Times that “God is in the detail”, implying that whatever one does should be done thoroughly and that good architecture is both the overall shape of the building as well as a nursing for detail. The Italian architectural historian Manfredo Tafuri (1935 – 1994) suggested that Mies’s work was the architecture of silence and order. Mies’s buildings could be interpreted as the silent contrast to the hectic chaos of urban life. 4

Peter Behrens (1868 – 1940) was a pioneer within industrial design. Born in Hamburg and educated as a painter, Behrens spent his first working years as painter, illustrator and bookbinder, frequenting the artistic bohemian circles of Munich in the late 19th century. In 1899 Behrens built and designed his own house and everything inside. He fully conceived furniture, towels, paintings, pottery, etc. The construction of this house is thought to be the turning point in his career, leaving behind the bohemian Jugendstil – the German name for Art Nouveau – and constitutes the move towards a sober and harsh style. Behrens went on to start his own architectural company, and he was one of the leaders of German modernism. He was a major designer of factories and office buildings in brick, steel and glass. In 1903 he became director of the school of applied arts in Düsseldorf, Kunstgewerbeshule – a school of arts and crafts. In 1907, together with ten other people, Behrens created the German Werkbund, an organisation basing their philosophy on the principles of the Arts and Crafts movement, but with a modern approach. The main goal of the organisation was to improve the design of everyday objects and products. Also in 1907 AEG retains Behrens as their artistic consultant. For AEG he invented corporate design and a corporate identity, creating a strategy for a similar and recognisable design throughout the company’s product line. For this invention he was considered the first industrial designer, starting an understanding of the power of good design, which only few companies have been able to repeat since – companies like Olivetti, Braun and Apple. Throughout the rest of his life he continued to work for AEG but also designed smaller houses and was appointed head of the Department of Fine Arts Vienna. He was later headhunted back to Berlin, allegedly by Hitler himself, to conduct a Master’s class in architecture at the Prussian Academy of Arts in 1936. In his architectural practice he took in students who spread his ideas around the world, among them Mies van der Rohe, Le Corbusier and Walter Gropius. 5

52: Peter Behrens, A. E. G. Turbine Factory in Berlin, 1909 Image courtesy of Boston University

109

110

MODERN ARCHITECTS PAST

111

MODERN ARCHITECTS PAST

LE CORBUSIER Chales-Édouard Jeanneret-Gris (1887 – 1965) renamed himself Le Corbusier in 1920. He was born in the northern Switzerland close to the French border, were he attended the art school La-Chaux-de-Fonds. At the age of 18 he started to design houses, the first for one of his teachers. Around 1907 he travelled to Paris where he found work at Auguste Perret’s office. During his time there he acknowledged the new possibilities that reinforced concrete opened up to. Between October 1910 and March 1911 he worked at Peter Behrens’ office where he met both Mies van der Rohe and Gropius. In Behrens practice he realised the importance of industrial design and of the possibilities in industrialising fabrication of construction materials. Unlike his former master Perret, Le Corbusier had no real interest in expressing the structure on the exterior. Reinforced concrete was used by Corbusier to highlight the modernist idea of clean and simple forms. His measure of value was formal and aesthetic in contrast to Perret’s structural rationalistic ideas of “honestly” expressing a building’s structure. Corbusier’s principle was to make the structure visible in a satisfying form, even if it meant twisting the structural logic to do so. Reinforced concrete was in the beginning of the 20th century the hottest material among architects. It was lightweight, strong and could be moulded into any shape. On top of that it was relatively cheap which made it possible

112

to erect better-constructed buildings for more people – one of Le Corbusier’s visions. In 1923 he proclaimed that architecture should not only be beautiful buildings, it should also play a role in society. Most importantly architecture should – according to Corbusier – create the best conditions possible for as many people as possible. The key to realise this vision, he thought, was found by looking towards the machine. Just as concrete buildings nowadays are sometimes looked down upon, today machines are also something people do not want to be associated with. This was, as well as concrete’s reputation, different in the begging of the 20th century, and for Corbusier the machine was an ideal. It was practical, had a clear purpose and function and it was efficient. He wanted to change the construction industry and architecture in that direction. He stated:

“The house is a machine for living in.”

This falls in line with one of the basic ideas of Modernism, that the house was not to symbolise social status or wealth but being functional, practical and emphasise equality between people. With the so-called Dom-Ino House, Le Corbusier expressed the fundamental principle of Modernism through structure. Between 1914 and 1915 he developed the Dom-Ino project in Switzerland, where he had returned as a cause of the break out of The First World War (1914

53: (Previous page) Farnsworth house in Illinois, completed in 1951. Image courtesy Sandra Cohen-Rose and Colin Rose

– 1918). It was a radical new approach to construction. Separating the structure from the enclosure – or separating the bones from the skin – meant that he could design his façade completely free of any respect to the structural rhythm and internal walls could be place freely on each floor.

“Until now: load-bearing walls; from the ground they are superimposed, forming the ground floor and the upper stories, up to the eaves. The layout is a slave to the supporting walls.

Reinforced concrete in the house provides a free plan! The floors are no longer superimposed by partition walls. They are free.” 6

Any part of the external wall could be made of glass and continuous window bands were possible. Corbusier defines the principles of the modern house and what he calls “structural rationalism” through his “Five Points of a New Architecture” published in 1926: • It should be raised on pillars (he calls thin reinforced concrete columns ”pilotis”)

54: Drawing by Le Corbusier for the patent of Dom-Ino house, 1915 Image courtesy of FLC - Foundation Le Corbusier

113

114

MODERN ARCHITECTS PAST

55: Front of the entrance side of Villa Savoye Image courtesy of Ville de Poissy

115

MODERN ARCHITECTS PAST

• It should have an open plan, benefitting from the freedom the pilotis brings and allowing each floor to be arranged differently

Vitruvian man, Alberti and others, in an attempt to relate the human proportions on mathematical formulas.

• The free façade, like the structure also allows for a freely composed façade

With the industrialisation and the newly invented the assembly line, Corbusier proposed the idea of smaller modules constructed off site and then being assembled on site like a giant game of LEGO. The buildings could be broken down into elements of concrete, steel and glass. The wall could be one module, a window a second and the roof a third. Corbusier used his Modulor measurement to determine the sizes of his modules.

• The Fenêtre en longueur – it should have window bands and windows stretched to the width of the building supported only by the floor slabs • Finally, it should have a functional roof

Geometry was the most important tool for Corbusier – it was his way to find the ideal. Like Alberti in the Renaissance, Corbusier admired the ancient Greeks’ use of “pure” geometric shapes: the circle and the square. His ideal was to bring the human proportions into his designs, and he made great use of the golden ratio. His scale of proportions, called “Le Modulor”, was developed as an extension of the works by Pythagoras, Vitruvius, da Vinci’s

A relatively large part of Le Corbusier’s designs was never realised. One completed project that manifests his five points is the Villa Savoye in Poisy, a suburban city in the outskirts of Paris. The villa, erected between 1929 and 1931, is a simple white box raised on pillars over a circular ground plan. Using the “pure” geometric shapes, Corbusier determines the overall form of the villa. The villa was built as a county retreat for the Savoye family. The circular ground plan houses the garage and rooms for the private chauffeur and maids. The first floor is built around a roof garden, which is enclosed within the box. Here he exemplifies his idea to compensate for the green area consumed by the building and relocating it on the roof. The kitchen, with build-in cabinets that easily slide open, is a good example of Corbusier’s “Machine for Living” ideal. The house should function as an efficient machine to make the daily life of the in-

116

56: Le Corbusier’s Chaise Longue - LC4 (“long chair”) photographed on display at Villa Savoye Image by author

habitants easier. The house therefore employs a number of mechanical devices - like the build in kitchen cabinets - for easing manual tasks. Le Corbusier had a long career, not only within the field of construction. He completely dedicates himself to painting in the years between 1918 and 1922, and from the begging of the 1930s he started to show his interest in city planning. He is also known for his furniture designs. Towards the end of his career, which ended with his death in 1965, his buildings

became more colourful and he rediscovered natural materials. There is a photograph of him and his two friends working on carving a pattern in the faรงade of his first work of architecture, the Villa Fallet in 1908. The photograph allegedly documents the last time Le Corbusier contributed directly to the construction of his projects. He was known for being frequently disappointed by the quality of craftsmanship in his buildings and he almost never visited his building sites. 7, 8, 9

57: The master bathroom in Villa Savoye, everything is build-in. On the right the rational structural system is seen, with the columns clearly offset from the faรงade. Image by author

117

MODERN ARCHITECTS PAST 118

58: Upper room and terrace at the Villa Savoye Image courtesy unknown

119

MODERN ARCHITECTS PAST

CODA CHAPTER FIVE The Modern movement brought one of the biggest changes to architecture ever. It also brought engineering into architecture; together with function, structural honesty was one of Modernism’s important characteristic. Most architecture of today still reflects Modernism in some way. Frank Lloyd Wright showed how architecture was no longer about decorating a façade with ornaments; it is just as much concerned with interior design. Together with the visual interpretation of a building, interior design is to this day one of the most important subjects for engineers and architects to collaborate on. Mies van der Rohe has been criticised for the cold style, a style without any ornamentation. Through their apprentices at Peter Behren’s office, Mies van der Rohe, Walter Gropius and Le Corbusier were the pioneers of bringing industrial design, including prefabrication, into architecture. Le Corbusier was perhaps the most influential architect ever. He became a legend when he was still alive, he even designed whole cities, but most importantly from an engineering point of view did he recognise that the role of the engineer was not to negligee. Only through the engineer could he fulfil his mantra of complete honesty in his buildings. It is with Le Corbusier’s visions and theories that engineering starts to make sense for an architect – and it creates a new approach from the architect towards the engineer.

120

The Modern movement introduced the beauty of structures to become accepted as architecture. It is not only because of the great infrastructural projects of the nineteenth century, but indeed also because of the arrival of a new kind of engineer in the beginning of the twentieth century: engineers who function also as architects.

121

ARCHITECTURAL ENGINEERS PAST

Chapter six

Architectural engineers

In direct competition to Hennebique’s system, the German firm Wayss and Freytag had adopted the ideas of Monier, the French gardener who reinforced thin concrete vessels with an iron wire mesh. G. A. Wayss (1851 – 1917) had a theoretical approach to the design of concrete structures his company was erecting, compared to the much more empirical approach of Hennebique. Both companies were highly successful at the beginning of the 20th century, but their approach to design was completely different. In contrast to Hennebique’s company, reinforced concrete became a material whose properties were well tested and could be mathematically calculated in the hands of Wayss and his colleagues. The two companies were in direct competition; it was in such a battle between French business and German science. Hennebique’s early dominance in the world market of concrete started to come to an end

122

when a Hennebique-guaranteed design collapsed in Basel in 1901. Swiss engineer Karl Wilhelm Ritter (1847 – 1906) warned in a serious of papers that the Hennebique system was not safe just because the Paris headquarter guaranteed it. The collapse eventually led to the first code of practice in Europe. The Wayss organisation still stands today, where Hennebique’s multinational company had fallen apart even before Hennebique’s death in 1921. The monolithic structures of the Hennebique system were based more on the success of previous structures rather than calculations. In sort of the same way, Wayss structure were based on calculations of successful forms – the designers at Wayss and Freytag were driven away from forms for which they had no calculations. Wayss changed his forms to fit with the formulas. The engineers at Hennebique were in that sense more open-minded; as forms proved successful they experimented

further, and faced with economic competition they tried to find ever lighter forms.1

MAILLART In Switzerland both French and German cultures meet without competing as nationalities. Maybe that is one of the reasons why Swiss engineers in the first half of the 20th century where among the most innovative in the world. Robert Maillart (1872 – 1940) attended the Federal Institute of Technology in Zurich, where he among other subjects was taught Graphic Statics by Ritter. Maillart was not the best in his class in academic theories, but he understood the necessary to make the right assumptions and he had a developed skill of analysing structures through visualisation. He was a symbol of the Swiss synthesis of French and German culture; he was on the one hand annoyed over the overuse of complex mathematics used by Germans and he used highly simplified calculations for many of his designs, and on the other hand he kept his intuition for forces up to date by full scale load tests.

mechanical and electrical services underneath the slab. The design also created a smooth visual transition between column and slab. The mushroom slab made new designs possible, and it was to great use of many modernist architects – one example being the Johnson Wax building in Wisconsin by Frank Lloyd Wright. Where the mushroom column stands close to an edge the slab can cantilever outwards and free the façade for any load-bearing function, something which, among other modernist architects, Le Corbusier found very useful. Maillart had started his own construction company Maillart & Cie in 1902. Unlike the big architectural names, like Frank Lloyd Wright, Maillart had to win his projects through competitions. This meant that he had to make his designs competitive, mainly on price. It meant a search for form that leads to minimum materials and minimum cost. Maillart had in the beginning of his career used the Hennebique system for two industrial buildings, but with his invention of the mushroom slab not only did he find a cheaper solution, he also made the concrete construction appear even lighter. His

In 1908, Maillart invented a new type of reinforced concrete system – the mushroom slab – where a flat floor slab rested only on capitalised columns. The Hennebique system relied on primary and secondary concrete beams, on which the concrete slab was supported. Maillart used columns with exposed capitals and eliminated the need for beams. His design increased the usable space and allowed for

59: Mushroom slab on the third floor of the Grain storage of the Swiss Confederation in Altdorf (1912) Image courtesy Chriusha

123

60: Salginatobel Bridge constructed across an alpine valley in Schiers, Switzerland between 1929 and 1930. Image courtesy of Alessandro Weiss von Trostprugg

ARCHITECTURAL ENGINEERS PAST 61

126

design of column capitals shows in their shape how they at the same time provide a smooth way for the forces and a simplicity that was inexpensive to build. Maillart had designed two concrete bridges in 1899 and 1901. His first project at Maillart & Cie was the Tavanasa Bridge over the Vorder Rhine River. Maillart had recognised that reinforced concrete made new designs possible, designs not previously possible when using stone or metal. An evolution of design is visible throughout his first bridge, even though the overall static principle is the same: a three hinged arch. His first bridge, from 1899, is located in Zurich and spans almost 40 m. The sidewalls were decorated with masonry designed by the City Architect, and the load-carrying reinforced concrete structure was thus hidden behind the walls. In his next bridge, the Zuoz bridge from 1901, Maillart turned

the decorative sidewalls into structural walls forming the first hollow box bridge ever built in concrete. It was the visual suggestion of form that stimulated him to recognise how decoration could be turned into utility. At the time, no mathematical theory existed for analysing a hollow concrete box. To Maillart’s luck, the client’s consultant was his own former teacher, Ritter. In many other countries the design would have been disapproved, but to Ritter the box design was more a challenge rather than an obstacle. Ritter carefully designed a fullscale load test to insure the bridge’s strength. The test did reveal minor cracks in the walls, but the overall result made Ritter approve the design. Maillart, who was there for the entire three-day test program, had learned an invaluable lesson. For his first design as an independent – the Tavanasa Bridge designed in 1904 – he removed that part of the wall

that had cracked. The result was a new form, ahead of its time and with quite the visual power. Even though Maillart had increased the material efficiency, the bridge was not a great aesthetic success at the time, and it was generally disregarded by the conventional taste of pre-war Europe. An avalanche destroyed the Tavanasa Bridge in 1927. Twenty-five years would pass before Maillart again had the opportunity to further develop his bridge design. In 1914, Maillart and his family were in Riga on summer holidays as the First World War broke out, cutting them off from home. The family decided to remain in Russia where Maillart built a number of large works until the Russian Revolution in 1917. He returned to Switzerland in 1919, where he set up a business as a designer only. In 1929, Maillart won the design-construction contract of the Salginatobel Bridge. The design of the 90 m spanning bridge was the lowest bid of the eighteen contestants, and the bridge was the longest spanning bridge in Maillart’s career. Upon completion the bridge received international attention, even though Maillart had used the three-hinged arch, concrete box design already twenty-five years earlier. Perhaps the dramatic scenery, with the bridge spanning a 90 m deep gorge with the mountains in the background, was the reason why the Salginatobel Bridge have been recognised as “Structural Art” – or perhaps the world was now ready for a design like this. However Mail-

lart himself was not completely satisfied with the Salginatobel Bridge, writing after its completion that the shape of the arch should have been pointed rather than pure curved, if it were properly to match his structural analysis:

Even the [Salginatobel Bridge] cannot lay claim to complete sincerity of form. Indeed, if both constant and shifting weights are taken into consideration, the extreme curves of exerted pressures form two lenticular surfaces whose lower contours meet at an acute angle. 2

Since returning from Russia, Maillart had been seeking ever thinner and more elegant forms throughout his works during the 1920’s. For smaller bridges he had tried to use a second new form, where he used the stiff road deck to support an ever-thinner concrete arch against buckling. With his Schwandbach Bridge, completed in 1933, Maillart arrives at his thinnest design. The slightly horizontally curved roadway is smoothly integrated with the vertically curved arch.3 David P. Billington is especially fond of the bridge:

“The bridge is strikingly thin, fully integrated, and contrasts vigorously with the setting. It is undeniably a work of man not of nature. It springs not from any organic, natural forms (forms found in nature) but from the imagination of an engineer. It expresses the ideal of minimum waste of materials and monies as well as the unique personality of Maillart.” 4

61: (Opposite page) Tavanasa bridge completed in 1906 Image courtesy unknown

127

62: Schwandbach bridge photographed after the completion in 1933 Image courtesy unknown

ARCHITECTURAL ENGINEERS PAST

FREYSSINET Eugène Freyssinet (1879 – 1962) was a pioneer within pre-stressed concrete and concrete shell structures. His first job was for the French highway department. As Maillart, he designed a series of small bridges, something that:

“[…] made me perfectly happy because the joy a work gives to its creator does not depend upon the size but upon the love which he brings to it.” 5

Freyssinet was born in the Corrèz plateau east of Bordeaux, an area characterised by harsh climate and bad soil. The people living there where though, poor and proud, they almost never felt the need to beg for assistance and in spite the rough conditions they took care of themselves. Freyssinet was, like his people, not an artist in the sense of art for art’s sake, but rather “universal artisans”. This view formed his feeling towards architecture and construction;

“I loved this art of building which I conceived in the same way as my artisan ancestors, as a means of reducing to the extreme, the human toil necessary to attain a useful goal.” 6

Freyssinet made his breakthrough as a great ingenious engineer in 1907. The highway department needed to replace three old suspension bridges over the river Allier, and they had

130

already made the design of a new bridge in stone at Le Veurdre. The price estimated for the stone bridge was very high which made it impossible to replace the other two bridges within the budget. Freyssinet proposed that if he would be allowed to design the bridge on his own, he could do them all for the price of one. He had never been in charge of anything this scale before, but he was given full design freedom as long as he could construct what he promised. Freyssinet designed his bridge at Le Veurdre as a three-span arch bridge in concrete. It was to be the largest spanning concrete bridge at the time, and despite the limited budget he made a full-scale test arch to study the behaviour. He designed his test arch with steel tie bars connecting the two supports and thereby avoiding them from sliding away when the arch was loaded. He could also use the steel bars to pull the supports together, thereby putting the arch into permanent compression. With his test model, Freyssinet made his first prestressed structure. The test finally showed Freyssinet that the French codes at the time did not give correct results, as the concrete kept contracting, even when the compression load from the pre-stress was kept the same. The contraction of the concrete taught Freyssinet the importance of “creep” in concrete, an effect that led to the arch having an increased movement. In 1910 the Veurdre bridge was completed, but Freyssinet kept observing the bridge’s movement carefully. He had constructed a concrete

bridge with pre-stressing methods for the first time: in the arch crown he had built in openings for jacks to push each half-arch apart, thus raising the arch. In early 1911 he realised that the bridge was moving downwards – because of the creep – in a rate he did not expect. Freyssinet felt he had to take action:

“Returning to Moulins in the night, I jumped onto my bicycle and rode to Veurdre to wake up Biguet and three reliable men. The five of us then re-inserted the decentering jacks – I had always kept the possibility in reserve – and as soon as there was enough daylight to use the level and staffs, we began to raise the three arches simultaneously. It was market day and every few minutes we had to interrupt the operation to allow a few vehicles to pass. However, all ended well and once more aligned, cured of the illness that had almost killed it, the Veurdre bridge behaved perfectly until its de-

struction in the war in 1940.” 7

Afterwards the holes for the jacks were filled with concrete and became a solid part of the arch. Freyssinet was able to keep the budget he had promised. Freyssinet eventually left the highway department in 1914 to join the building firm Claude Limousin where he stayed until 1929. While working for Claude Linousin, Freyssinet designed structures that gave him international fame, among them two hangars in Orly constructed between 1921 and 1923. The hangars span 86 meters with a height of 50 meters in the middle. They seem to rise directly out of the ground, but were in fact connected under the ground. The thin shell roof structures were made from a series of arches with a hollow cross section.

“The two Orly hangars became world famous

63: Freyssinet’s new Pont le Veurdre on the left and the old bridge it was replacing on the right Image courtesy of the Freyssinet association

131

ARCHITECTURAL ENGINEERS PAST

immediately. Perhaps for the first time artists and architects began to sense that something totally new was emerging in concrete.â&#x20AC;? 8

The hangar design was at the same time highly spectacular and simple. Each arch carried its own load directly to the ground. The arches was connected laterally with small thin slabs, given the structure a ribbed feeling. Freyssinet went on to design great bridges and industrial buildings, among them a railway repair shop in Bagneaux, near Paris. He patented his idea of pre-stressing in 1928. Although Freyssinet did much to develop prestressed concrete, he was not its inventor. Besides bringing focus to pre-stressed concrete

64 132

64: Orly hangars in Orly at the outskirts of Paris Image courtesy unknown

through his many bridges and shell structures, one of his contributions was to recognise that only high-strength pre-stressing wires were able to counteract on the concreteâ&#x20AC;&#x2122;s contracting behaviour. 9

65: Image during construction of the Orly hangars in 1923 Image courtesy of the Freyssinet association

133

ARCHITECTURAL ENGINEERS PAST

Heinz Isler (1926 – 2009) was another great concrete shell designer. Isler was born in a Zurich suburb, and in school he had a reputation of being independant and always go his own way. Graduating from the Zurich Federal Technical Institute in 1950, Isler was the only one from a class of more than one hundred students to study thin shells in his final-year design project. At the Federal Technical Institute Isler was influenced by his teacher Pierre Lardy (1903 – 1958). Lardy inspired many students, among them bridge designer Christian Menn (b. 1927), with a mixed passion for the beauty of structural form and mathematical analysis. Isler, at one point, brought an idea to Lardy, that first any given structure should be seen as a whole and only afterwards analysed in its parts, something Lardy completely agreed on.

134

Isler started his career as an assisting teacher of Lardy, while doing free-lance engineering for thin shells on the side. He seriously considered a career as a painter, but in 1953 chose to follow the path, which he was

66: Concrete dome roof covering the former company Kilcher’s office in Recherswil, completed in 1965 Image courtesy Chriusha

already trained. In 1955 he showed his designs on a congress in Amsterdam and quickly got numerous commissions for the design of thin shells. As something new, his design approach was with no reference to architecture but a more academic approach.

In 1959 Isler attended a conference on concrete shell structures in Madrid, organised by Eduardo Torroja. Isler’s presentation on his advances within the field led to a long discussion; Torroja wanted to know how he could ensure results from small-scale model tests could be applied to large scale, French engineer Nicholas Esquillian was concerned that the cost of formwork was underestimated and finally Ove Arup spoke about the need for collaboration with architects, explaining how he thought buildings should be: “[…] functionally right, architecturally right and aesthetically right, and it is not any great consolation that we get something else cheap”. Isler responded Torroja, that he could ensure accurate models by controlling them with precise Swiss measuring tools. To Esquillan he answered how a new formwork technique allowed him to keep the price low. Isler made no response to Arup and the question to collaborate with architects; for Isler the form finding process with thin-shell concrete structures was merely an engineering problem. A problem that involves imaginative shaping, but imaginative towards a structurally optimised form. 10

67: Two concrete shell roofs for a highway gas station near Deitingen, Switzerland, completed in 1968 Image courtesy Chriusha

135

ARCHITECTURAL ENGINEERS PAST 68

136

TORROJA Eduardo Torroja y Miret (1899 – 1961) was a Spanish structural engineer who designed mostly industrial buildings, stadiums and infrastructure. Torroja was a pioneer in Spain within the design of concrete shell structures, and he made great use of pre-stressed structures.

“Torroja is an outstanding example of the exceptional structural engineer. Mathematically minded, and aided by the extraordinary physical intuition with which his creative growth has endowed him, he designs structures on the sound bases of economy and strength. But the humanist in him has a flair for the beautiful, and a refined, delicate sense of beauty pervades his conceptions in space. So subtle is his esthetic feeling that you will find him spacing a set of columns in different intervals in or-

68: Initial and final design of the Tempul Aquaduct Image from The Structures of Eduardo Torroja

der to achieve a pleasing architectural rhythm, thus repeating in concrete what the Greeks did in marble.” 11

In 1925 Torroja designed his first large project, the Tempul Aqueduct near the ancient Spanish city Jerez de la Frontera. The aqueduct initially consisted of fourteen similar bridge segments, each 20 meters long. In one end the aqueduct crossed a river, and two of the concrete column supports were placed in the river in his initial design, but the local control office thought that foundation might be undermined for those columns. Torroja was pressed for time and did not want to change the whole design, but rather keep the same cross section of the bridge elements. He removed the two supports standing in the river, and used the material he gained from this to enforce and raise the two supports standing on the riverbanks.

Cables were spun from the higher supports to the point where the river supports used to be. To be sure that the bending moments in the middle bridge segments were the same as for all the other segments, they were connected to the adjacent segments with a hinged connection â&#x20AC;&#x201C; a so-called Gerberette system. 12 Throughout his career Torroja went to design several aqueducts. He was always aware of the internal forces, as the bending moment, during the development of his designs. For the Alloz Aqueduct, completed in 1939, Torroja used cables to pre-stress the U-shaped concrete cross section. He wanted to design

a structure that was always in compression, in that way the concrete could not crack and therefore being watertight. He used the prestressing cables to again change the curve of the bending moment to his advantage.

â&#x20AC;&#x153;The bending moments due to the weight of the water of the water channel are therefore negative throughout the whole length of each section except at the free ends and at the midpoint, where they are zero.â&#x20AC;? 14

Even though this resulted in higher stresses, 69

69: The Alloz Aqueduct in barren landscape. The pre-stressed U-shaped cross section is always under compression Image courtesy unknown

137

ARCHITECTURAL ENGINEERS PAST

The German engineer Heinrich Gottfried Gerber (1832 – 1912) obtained a patent for a hinged girder in 1866. The Hassfurt Brücke over the river Main, recognised as the first cantilever bridge, was completed in 1867. Gerber had studied civil engineering in Nuremberg and Munich before he was employed at the Bavarian State Company where he worked until 1858. He then took over a bridge company and went on to design and construct about 600 steel bridges, mainly for the railway. Besides from the research he made for the hinged bridge design, he also studied the limits of elasticity theory and the most unfavorable distribution for traffic loads. As the entrepreneur he had become, he was interested in finding the most efficient structure and thereby ways to save materials. 13 At the time Gerber designed the first hinged bridge, Navier’s elastic theories had been known for more than 50 years. The theory of plasticity had not yet been developed, but as Gerber did research for ultimate state, the limits of elasticity theory, he probably had some idea that the steel had spare capacity left when designed only on the bases of elasticity theory. The Hassfurt Brücke was designed as a three-span bridge, with two main supporting piers standing in the river. If the middle span was cut out and examined, it would normally function as a beam cantilevered from both sides – a hyperstatic structure. According to Navier’s elasticity theory the bending moments at the supports are the double of the bending moment in the middle, assuming that the load is equally distributed over the entire length. As Kazinczy explored in 1914, the bending moment at the cantilevered supports was rarely the double of the bending moment in the middle when testing his beam, as the theory assumes that the cantilevered supports are able to keep the beam from any rotation – they are infinitely stiff supports, something that does not exist in reality. The value of the calculated bending moment goes from negative at the cantilevered support to positive in the middle of the field and back to negative at the other support. Therefore the bending moment must be zero at two points. Gerber used hinges so that he himself can chose the point where the bending moment is zero. If he choses the same point as the point he would find according to elasticity theory, the bending moment in the middle would be half the one of the two supports. This is not an efficient structure. When Gerber moved the hinges towards the supports he could lower the bending moment at the supports, and eventually find the point where the value was the same for the supports as for the middle. The Gerberette system became known worldwide, and have been used in famous bridges, as The Forth 70 Bridge in Scotland that opened in 1890 but perhaps best examplified in the Centre Pompidou.

138

70: Elevation of the Hassfurt Brücke Drawing by E. Hermes

and more material, the overall cost was less as the concrete itself was watertight and no extra tightening layer was needed. The Alloz Aqueduct was mainly erected in barren nature, but even then Torroja designed the columns to be esthetically pleasing. In 1933 he had designed a viaduct in closer contact to civilisation, that being inside the Madrid University City. Here Torroja played with the layout of the columns to achieve a pleasing visual experience for the students.

“The spans between the columns are not all equal. Those between the shorter columns are slightly smaller to improve the optical effect of the proportions.” 15

Torroja was perhaps most known for his design of thin concrete shells, perfectly exemplified in the Madrid Racecourse, which was completed in 1935. The grand stands are covered with cantilevered, thin concrete shells. The general layout and cross section of the construction was found by examining the functions needed; betting hall, betting offices and staff gangway, a promenade and of course the covered stand itself. He needed to cover both the stands and the promenade, and he used the natural seperation line between the two areas to place the main columns. The stand being much larger than the promenade, the promenade roof was used as a counterweight for the stand roof but the weight was not enough to balance the system. Torroja there-

fore connected the upper promenade roof structure with the under-laying betting hall roof with thin tension columns. In this way the betting hall roof structure could be done without columns towards the entrance.

The shape of the stand roof structure was examined thoroughly. To analyse the stress in the concrete shells, Torroja had to use simplified methods as the theory of elasticity was not yet well enough developed to be used for structures of this type, being double curved surfaces.

“However, even without precise analysis, it is known that they possess good structural properties in space.” 16

Torroja simplified the curved shapes to a simple I-beam, breaking it down into three parts; compression, tension and the connection web.

“In each vault the haunches may be regarded as the web of a beam in which the crown of the vaults corresponds to the tension chord, and

71: The structural principle of the Madrid Racecourse’s grand stands as sketches by Torroja Image from The Structures of Eduardo Torroja

139

140

ARCHITECTURAL ENGINEERS PAST

72: Grand stand at the Madrid Racecourse Image by Ximo Michavila

141

ARCHITECTURAL ENGINEERS PAST

the edges align which the vaults meet each other to the compression chord.” 17

Of course Torroja also used other, more detailed methods for analysing the shell structures – for example he looked at the isostatic lines to design the reinforcement – and a full scale test was setup to ensure stability. Torroja tested several shapes for the roof. The initial option was a simple arcade-like structure, which was then further developed to be more conoid. The conoid option had the barrel shape, like the arcade, over the supports, but was at the free edge strait. Finally, at one o’clock in the morning, Torroja found what he thought to be the most preferable option: the hyperboloid. The final hyperboloid concrete structure was just 5 cm thick but still cantilevers 13 meters.

“Of the many basic types, the resulting surface could well have been a conid but for the objection that the conoid is not very attractive. It seemed preferable to choose some other form of curvature. Among the better known ones, none seemed more adaptable than the hyperboloid.” 18

Torroja was only very careful with the design of the intersection line between two vaults, what was previously described as the compression chord.

142

“The possibility of making the lines of intersection curve downwards had been contemplated, but finally this shape was discarded, and it was decided to have straight lines of intersection sloping upwards towards the free edge. It was felt that the lines of intersection curving downwards would be less agreeable to the eye, or it would at least be too startling to observers accustomed to a more classical expression.” 19

Among the many projects he did, some of the mayor ones was the Fronton Recoletos, an indoor stadium for the game called pelota, which unfortunately was bombed under the Spanish Civil War, soon after completion in 1935. The stadium roof, covering an area 34 meters wide and 55 meters long, was designed as two barrels with different radia, and within them large parts with skylights were incorporated. For the construction of a covered market at Algeciras in 1933, Torroja also incorporated large skylights in the dome design. The dome had a diameter of 50 m and was done as a prestressed concrete shell structure. Torroja was known for believing that the structure should follow the personality of its designer. He developed a new way of looking at structures, something founded in his Spanish cultural heritage. He lived in, and worked for, a country in which aesthetic values have always had big importance for the inhabitants. The Spanish economy was after The First World War and until the breakout of the Spanish Civil

War in 1936 in bad conditions, and therefore many of Torroja’s works had a limited budget. Torroja sought to improve the appearance of the structures through form rather than by decorative additions – simply because there was no money for it. 20 In 1958, in his autobiography, he sums up his view on the design process:

the basic principles that a long experience of technical creative work leaves in the unfathomable depths of our personality, so that these may later subconsciously condition our own intuitive thought. But those basic principles are not enough in themselves to create, critically and deductively, a new form. For this to emerge, a spark of imagination is required. Indeed, it often appears at the most unexpected moment, when we are least trying to create.” 21

“To me it seems clear that the imagination can operate successfully only in conjunction with

73: Image of the full scale test roof during construction Image from The Structures of Eduardo Torroja

143

ARCHITECTURAL ENGINEERS PAST

NERVI Pier Luigi Nervi (1891 – 1979) was born in a small village in northern Italy. In 1913 he graduated as an engineer from the University of Bologna, but he saw himself as an architect. He had established his construction company in 1923, and as a building contractor Nervi became the designer of reinforced concrete structures. In that sense, Nervi had the function of the engineer, the architect and as craftsman as his company was in charge of erecting his designs. Because of this, Nervi was recognised as one of the first technician-artists or engineer-architects in the world. After his graduation Nervi had worked for the Societa per Costruzioni Cementizie for nearly ten years, interrupted by First World War where he served the Italien engineering troops. Returning from the war, Nervi was sent to direct the Florence office, which gave him considerable autonomy. Already in 1920 Nervi had formed his own company, and in 1923 he moved to Rome joining his company with an already existing construction company of Rodolfo Nebbiosi. The company, and Nervi as a designer, acquired international fame for his design of the Minicipal Stadium in Florence, completed in 1932. Like almost all of Nervi’s works before the 1950s, the commission for the Florence stadium was won as a result of a competition. Nervi’s design was selected for it’s extraordinary economy. When designing his structures, Nervi described his primary concern as with

144

technical, economical and functional requirements. For the Florence stadium the roof was made as a reinforced concrete shell stiffened by cantilevered curved beams. In a scissor-like system, the main beams used for supporting the stands continue to support both the curved cantilevered beams and at the same time holds down the back of the roof. The forces in the roof structure are clearly readable, with elements shaped bigger where the forces concentrate. Even though the aesthetics were not his main concerns, it was still an important factor:

“The variation in section of the main ribs is indicated by the law of governing the variation of moments. Purely esthetic considerations inspired the slight curve of the canopy and the haunching of the main ribs.” 22

In the beginning of the 1930s both Italian an international architectural critics were undergoing a cultural shift towards an appreciation of works of engineering in the industrial age.

“When I began my construction activity, the technical problems connected with architecture were very modest: open spans of ten to fifteen metres were, in fact, exceptional, coverings of fifteen to twenty metres, truly audacious. […] Year by year I have seen the growth not only of the complexity of static problems, and of those deriving from improving technical equipment, but above all I have seen the

growth of the size and impressiveness of bearing structures to the point when they have become so remarkable that they can never again be lost under traditional […] decorations […] and trimmings. […] A real revolution was being confirmed: the great majority of those concerned with architecture, professionally and culturally, agreed that even a bare, sincere structure could give the full effect of beauty and could be true architecture, and that, on the other hand, forms and volumes, set by technical and functional necessity, treated with sensibility, could become eloquent means of architectonic expression.” 23

The companionship with Nebbiosi ended soon 74 after the completion of the Florence Stadium and Nervi then went to start another design and construction firm with his cousin, called Ingg. Nervi e Bartoli. In 1935 the company won a competition organised by the Italian air force authorities to design a series of hangars. The competition programme called for covering an area of 4100 square meters and with door openings of 50 meters.

“I designed the structure as a geodetic framework acting together as a whole, as I believed this would give the most economical solution

74: Drawing from the competition of the Stadio Artemio Franchi in Florence Image courtesy of Comune di Firenze Archivio storico

145

ARCHITECTURAL ENGINEERS PAST 146

75: Hangar in Orvieto, destroyed during The Second World War Image courtesy unknown

147

ARCHITECTURAL ENGINEERS PAST

and the one requiring the least steel. With this type of design the calculations were extremely complicated. […] I therefore decided to make a preliminary calculation and to make a detailed study of the stresses by means of experiments on a model.” 24

This was one of the first times the results of a model test were applied directly to a largescale structure. The results enabled Nervi to, in detail, analyse the stress in the concrete ribs. The difference between his simplified calculations and the results from the model test was so small that the design did not need bigger modifications. The construction of the hangars in Orvieto (1935 – 1938) used timber formwork, which proved to be less economically favourable for the curved concrete ribs. To overcome the disadvantages, Nervi developed a pre-fabricated system.

“At the time, the need for economy in materials and had become even more acute and this is why, on the basis of the experience acquired, I decided to simplify and lighten the structure by designing the ribs as a lattice, which would enable me to make use of prefabrication.” 25

In the middle of the 1940s Nervi developed Ferro-cemento. He had acknowledged how normal reinforced concrete could withstand large strains without cracking if the reinforcement was subdivided into more bars and therefore more distributed. This let to the in-

148

vention of the Ferro-cemento, which consisted of a very thin mesh of steel wires, from 0,5 mm in diameter, sprayed with cement mortar. With Ferro-cemento Nervi could produce reinforced concrete slabs much thinner than previously seen, a normal ferro-cemento slab is only 2–5 cm thick, but still keeps a very high strength. Nervi used it for all sorts of projects, even for thin boat shells, but in the beginning it was developed for the purpose of structural formwork. The steel mesh could be moulded in any shape wanted and the density of the mesh ensured that the cement mortar stuck to the steel when sprayed on. It had yet another advantage when poured with plaster moulds the resulting surface was clean and smooth, requiring no extra finishing. The possibilities for his new Ferro-cemento material was first put on display at the Turin Exposition Hall completed in 1948 and Salone C in 1949. The roof of Salone B was mainly constructed by the use of precast elements, and it has been described as one of the most impressive interior spaces created in the 20th century. The old exhibition halls had been destroyed during the war and a competition was held for the reconstruction. Nervi’s proposed design was both the cheapest and the fastest to erect.

“The problem was particularly interesting, not only because of the dimensions of the hall but also because of the very short time allowed for the execution of the work, which was to start in September and had to be finished by the end

of April. This very short time was a real problem in view of the difficult climate in Turin. The solution I immediately thought of was a structure in corrugated Ferro-cemento, which would attain the necessary stability by virtue of the corrugations and would enable us to use precasting, and to manufacture the roof units while the floors and supporting structure were being built. On this basis I designed the roof structure with corrugations of about 2,5 meter span, divided into units 4 meters long. The units were to be made of Ferro-cemento in order to be as

light as possible and would be rendered mon- 76 olithic by reinforced concrete ribs cast in place, and located at the peaks and troughs of the corrugations. In this way the Ferro-cemento units would act as junction units between the in situ ribs which in turn would take over the main structural work. [â&#x20AC;Ś] The casting of the ribs proceeded without any difficulty and without the need for double formwork, as would have been the case with ordinary reinforced concrete.â&#x20AC;? 26

76: Main hall of the Turin Exhibition Image courtesy of Mario Carrieri

149

ARCHITECTURAL ENGINEERS PAST

For the main hall of the Turin Exhibition, Nervi introduced another precast system. Precast diamond shaped units measuring 2 meters by 4 meters were temporarily supported by wood forms and reinforced concrete ribs then cast in between the units and the formwork. Wood was at the time extremely expensive in Italy, and Nervi was in this way able to drastically reduce the need for formwork. Until the 1950s Nervi’s biggest commissions were domes and vaults, exhibition halls, stadiums and industrial buildings. The Tobacco Factory at Bologna and the Gatti Wool Factory in Rome are of the more famous factory buildings designed by Nervi. Here Nervi solved the formwork problem in an even more drastic way, designing re-usable Ferro-cemento moulds, mounted on moveable scaffolding. Columns were erected first, then the Ferro-ce-

mento mould was put in place and the ceiling was cast. Afterwards the mould was lowered and moved to the next position. The same mould was used for all the floors.

“This new freedom also made it possible to design roofs with ribs located at the isostatic lines of the principle bending moments, a design which makes possible strict adherence to the laws of statics and, therefore, makes the most efficient use of the materials. […] The esthetically satisfying result of the interplay of ribs placed in this way is a clear reminder of the mysterious affinity to be found between physical laws and our own senses.” 27

In the 1950s Nervi began to receive more commissions solely as a structural designer

150

77: Entrance roof of the UNESCO building in Paris have the signature of Nervi Photograph by Lucien Hervé

or consultant. He had achieved international fame and status, and people around the world wanted to collaborate with him. Big clients, like UNESCO or Pirelli, applied a team of architects to work together with Nervi and with him as the structural engineer in the team. Nervi had his troubles with these jobs; he primarily searched the correct structural solution and the elegancy and honesty sprung from this. Being true to the natural laws had given him aesthetically pleasing results, which normally had no need for dressing up. Nervi never had the need to add a decorative façade to his buildings. The architects working with Nervi were rarely concerned with an elegant structural solution, but more focused on making an architectural “statement”. They admire, and desire, the elegant structural solution, but for some it becomes structure for art’s sake. Nervi also tried to embrace a more architectural approach

when designing, but realised the gap was too 78 big for him:

“Because of flattering comment […] I tried to bring esthetic theories to bear upon structural problems […] but soon convinced myself that to find an architectonic expression becomes more difficult the more one works with such an idea in mind. […] The prime condition of architectural expression [is] the inevitability of its structural design.” 28

Nervi’s best works were created when he was able to search the most efficient design from a technical and economic point of view. In his late career he showed how good it could be done, when designing the Palazzo dello Sport for the 1960’s Summer Olympics in Rome. The indoor arena was designed to host up to

78: The interior and exterior of the second building at the UNESCO building, known as the “accordion”, also carry the signature of Nervi. Images by author

151

ARCHITECTURAL ENGINEERS PAST

5000 spectators. In contrast to the UNESCO building, arenas and stadiums are engineering problems before they are anything else, something that permitted a more exposed structure and less architectural “dressing up”. As one of the few Architectural Engineers working in the first half of the 20th century, Nervi lived long after The Second World War where the necessity of collaboration between architects and engineers became more and more evident. 29, 30 In an unpublished manuscript Nervi explained his idea on the design process:

152

structional essence, a structure is considered solely on the basis of its external appearance – which people try to adapt to a variety of different problems. […] The result is always unfortunate […] structural acrobatics are a sign of a false structural conception. […] I have thought much about it, and I believe that I can affirm that an architectonic structure must be born and derived spontaneously from the “static sense” that progress in engineering has in part brought about, sustaining and integrating it with the formal reasoning of the science of construction.” 31

“The present [ca. 1960] moment in architecture is full of promise, but the dangers should not be overlooked. Alarming symptoms can already be seen. […] Too often, through a lack of understanding of its structural and con-

79: Concstruction of the Palazzo dello Sport. Image courtesy of Sistema Archivistico Nazionale 80: (Opposite page) Inside view of the Palazzo dello Sport roof structure

153

ARCHITECTURAL ENGINEERS PAST

CODA CHAPTER SIX Optimisation of structures, which is the key subject to the architectural engineers described in this chapter, can in itself result in beauty. For the first time, engineers made use of scientific skills to be creative. Through calculations natural shapes occurred. Engineers of the time, as today, were bound to find the most economic structure. Only through minimising the forces one could optimise the use of materials. But an economic structure went hand in hand with the construction method. The engineer had to possess an ability to think how his structures could be built. Nervi went as far as inventing a new type of concrete, ferro-cemento, to suppress the need for formwork. Still today, formwork is one of the biggest costs when constructing in reinforced concrete. To design a slab to represent the isostatic lines of the bending moments both creates an optimised use of the material and a visually pleasing shape. The architectural engineers took the principle of the hanging chain to a new dimension. As it will be shown later, there exist two kinds of architectural engineers: the first is the type described in chapter six; the second is the engineer who collaborates in harmony with architects. The architectural engineering education of today is aimed towards the second kind. The first type of architectural engineers surely created some very beautiful structures, but the

154

projects can typically be categorised as largescale structure with a clear function; stadia, marketplaces, hangars or factory halls. When it comes to smaller scale buildings they had a harder time succeeding. The example of Nervi is to me the clearest: he was not able to create the beautiful structures he is known for, when he was not completely in charge of a project. He failed at the collaborative process. Ove Arup once said about Nervi:

â&#x20AC;&#x153;He is still the archetypal builder, for whom the ends and the means fuse into one harmonious creation. [...] The starting point is the need for economy. Naturally, a great and beautiful structure is the end in view [... but] Designing in reinforced concrete it is essential for economy to lead the forces down to the ground by direct thrust.â&#x20AC;?

To Nervi, a beautiful building should only visually express an efficient structural or constructional reality. To me that is not the truth. It can be true for a stadium like the Palazzo dello Sport, but it is not the case on a project like the UNESCO building in Paris. However, it is still recognisable how an optimisation of a structure leads to such beauty as it is the case with the structures of for instance Maillart, Nervi or Torroja. This type of architectural engineer still exists today, for instance Luxembourgish Laurant Ney, founder of Ney & Partners. I have chosen

to leave out Ney from this thesis, as he falls out of my historical timeline, but his works are nonetheless interesting in this context.

81: Footbridge of Knokke. The shape was found through an optimisation of the self-weight as a function of deflection and Von Mises stresses. Structure, architecture and landscape architecture was done by Ney. Image courtesy of Ney & Partners

155

ENGINEERING ARCHITECTS PAST

Chapter seven

Engineering architects

In the first half of the 20th century architects were slowly recognising structures as architecture. The Swiss historian and architectural critic Sigfried Giedion (1888 – 1968) claimed that engineering shaped modern architecture. He noted:

“It is construction and not architecture which offers the best guideposts through the century.” 1

The first modern architects were aware of the need to incorporate the structure in their architecture. It was a part of the honesty and simplicity that Modernism represents. In 1923 Le Corbusier wrote in his essay Vers une architecture, translated into English as Towards a New Architecture:

“The engineer’s aesthetic, and architecture,

156

are two things that march together and follow one from the other: the one being now at its full height, the other in an unhappy state of retrogression. The engineer, inspired by the law of economy and governed by mathematical calculation, puts us in accord with universal law. He achieves harmony.” 2

As some of the most influential engineers sought towards architecture, some architects also sought towards engineering. By understanding the limits of structures, architects could exploit the best of both or even go in a new direction.

GAUDÍ Antoni Gaudí i Cornet (1852 – 1926) was born into a family of craftsmen, both his father and grandfather being coppersmiths. As a child he watched how they where able to mould the

82: (Opposite page) Organic shaped façade on the Casa Batlló, Barcelona, cladded with colored ceramic tiles Image courtesy of Shawn Lipowski

157

ENGINEERING ARCHITECTS PAST

copper sheets into delicate rounded shapes only using a hammer and skilful hands. He was perhaps more an architect-craftsman rather than an engineering architect. In a time when Modern architecture was blooming, Gaudí went his own way. As a child Gaudí was already a bit of an outsider, suffering from rheumatism he was not as mobile as his peers and was often left to play alone. He once explained how he as a child often studied nature and natural forms instead of playing with his friends. He studied at the Escuela de Arquitectura in Barcelona, from where he graduated in 1878. He excelled among the students; he was skilful with his hands, creative and with a good sense for mathematical calculations. His teachers was a bit suspicious because of the unconventional way he treated structural shapes. His first works were notable for the mixture of styles, consisting of natural shapes combined with colourful ceramic tiles. In 1883, at the age of thirty-one, Gaudí was given the task to design the Sagrada Familia, which would occupy him for the remainder of his life. Between 1883 and his death in 1926, Gaudí created four different designs for the temple. Gaudí did not embrace the flatness and linearity of the Modern architectural style, and his uniqueness made him recognisable from his earliest years. As the son of a craftsman, he used the background knowledge that he adopted from his father and grandfather to realise the complex geometric shapes he de-

158

signed. He mainly used known building materials with walls of stone or brick combined with iron pillars and beams, but the walls he designed were rarely straight and the columns almost always inclined. Only in 1925, soon before his death, he started to use reinforced concrete. As a theorist, Gaudí was probably best known for his analysis of form. He took Hooke’s inverted chain to a new level, making large-scale upside-down models to find a perfectly optimised form. He used graphic static analysis to derive the form and applied rule shapes like catenary arcs, hyperbolic paraboloids and hyperboloids of revolution. Gaudí’s hyperbolic shape was:

“[…] of a higher order and greater complexity than the [usual forms] of the middle ages which were being revived at the time.” 3

This helped him to relatively easily reproduce the shapes on site with simple tools like a measuring stick and line. The idea of using funicular polygons - polygons derived from the principle of the inverted chain - was not new, but Gaudí was the first to construct a three-dimensional arrangement of funicular polygons. He constructed his inverted models with tension wires to act as linear structural elements, chains and bags of pellets to control the loading of the model pieces of silk paper were used to simulate infill of walls and vaults.

“For Gaudí, form did not follow structure and

83 83: Inverted model of the Sagrada Familia church. Allegedly GaudĂ took an image of the shape an turned it around. Image courtesy of ABOL

159

ENGINEERING ARCHITECTS PAST

construction. It was identical with them.” 4

The hyperbolic structure had several advantages over the classical circular shaped barrel vaults. The form followed the natural form of the forces, allowing for a thinner cross section and a lighter structure. For the same reason it is much stiffer and has less tendency to buckle. The hyperbolic paraboloid also contains straight lines, something that made it much

162

easier to construct. For Gaudí this had a deep religious meaning; the shape confirmed the naturalness he was seeking. Gaudí described the form to be:

“[…] a miracle of mathematics. [The form] attributed holy properties to the trinity of straight lines which determine any such surface.” 5

84: (Previous page) Nave of the Sagrada Familia Church. Image courtesy of Flicker user SBA73 85: Church of Colònia Güell. Image courtesy of Samuel Ludwig

On 7 June 1926, Gaudí walked on to street and was hit by a tram on his way to construction site of Sagrada Familia. Assumed to be a beggar because of his lack of identity documents and shabby clothing, the unconscious Gaudí did not receive immediate aid. He died later in a hospital for the poor. He was buried in his unfinished Sagrada Familia. Gaudí worked on the church for 40 years, and completely dedicated his last 15 years focus-

ing on the project. After his death many architects have tried to complete it, but severeal problems have occurred; the original models of Gaudí were destroyed in 1936 but later reconstructed from the remaining and from pictures; the local government of Catalonia have had difficulties funding the grand scale project. Today the church is the biggest tourist attracting in Spain, even though it is still under construction and has been since 1882! 6, 7 86

86: Main entrance of the Sagrada Familia as it looked in the beginning of November 2012 Photographed by Jessica Wurm

163

ENGINEERING ARCHITECTS PAST

Spanish form finding designers in the first half of the 20th century differed from those from Germany, Switzerland, France and Italy. The Spanish sought an ideal of thinness and smooth surfaces without ribs. Felix Candela (1910 – 1997) was, together with Gaudí and Torroja, among the leaders within modern vault design. Candela was trained as an architect, but was skilled in mathematical analysis. In 1935 Candela graduated from the Escula Superior de Arquitectura in Madrid. During The Spanish Civil War he fought on the loosing side and was forced to flee from Spain with the Franco victory. In 1939 he landed in Mexico where he started working part time with his brother and the other half in an architectural office. He had always been interested in thin shell structures and read all articles and books he could find on the subject. “[…] since I never had a high opinion of myself as an artist, I was more interested in the technical part of the curriculum and began to read extensively about structures.” 8

164

By 1951 he had built seven shell and some funicular vaults, when he was appointed to design the Cosmic Rays Pavilion in Mexico City. He won the contract with a design of a shell only 1,5 cm thick spanning almost eleven meters, which gave him immediate attention in the building industry. He had read so much on the topic, that when he was invited to the first thin shell conference in the United States in 1954 he said: “Having arrived with a complex about the prestige of the place and the numerous experts gathered there, I suddenly found that I was somewhat ahead of the experts myself.” 9 Candela now had a great confidence in his ability to design thin shell structures. His own business became greatly successful after 1954, as he was able to erect cheap umbrella shells for industrial buildings. The low cost was a mixture of low labour rates and reusable formwork. He was able to produce designs quickly: 10 “[It] was made in an afternoon, drawn up in a week, and calculated during construction.” 11

87: An example of the umbrella shell structure is the Cafe Los Manantiales in Mexico from 1958 Image courtesy of Universidad Politécnica de Madrid

165

ENGINEERING ARCHITECTS PAST

BUCKMINSTER FULLER His road to international fame was bumpy and long, and “drifter” is perhaps the best way to describe Richard Buckminster Fuller’s (1895 – 1983) personality until his bankruptcy and death of his daughter. He grew up in Maine, United States, and despite having troubles with geometry in school he had performed well enough to attend Harvard University. He was expelled twice from Harvard; first time for skipping exams and the second time for his “irresponsibility and lack of interest.” After his first dismissal, his family sent him to work in a textile mill in Canada. Here, his family thought, he would be taught a lesson and encourage him to succeed at Harvard. But rather the opposite happened: he started to take his interests in machines and his inventiveness more serious. After the second dismissal he had simple jobs, working as a cashier and he joined the US Naval Reserves during The First World War. After the war, Buckminster Fuller set up a company with his father-in-law where they experimented with prefabricated housing units. The ability to mass-produce houses was a driving force for Buckminster Fuller. His father-in-law had developed a light-weight concrete block that could be mass produced for a very cheap price. Together they developed several prototypes of a prefabricated house, but none of them made it into production. In 1922 Buckminster Fuller’s daughter died, something that led him to drink frequently. In 1927 he was forced out of the company with his father-inlaw. Being jobless, without money, and living

166

88: Prototype of Buckminster Fuller’s Dymaxion House Image courtesy of Lisa Raynes

in a low-income public housing in Chicago, Buckminster Fuller made a decision to make his life:

“an experiment to find what a single individual can contribute to changing the world and benefiting all humanity.” 12

In 1927 he designed and patented his first mass-produced housing unit, the Dymaxion Deployment Unit. In the beginning his sphere of influence was limited to Chicago, but even though many of his earliest projects never made it beyond prototype, they slowly gained him fame in the rest of the United States and later the world. Buckminster Fuller’s idea of a prefabricated house differed from other massproduced houses of the time, being completely equipped like a mobile home or a car. One of his other projects, a quite promising one, was the 3-wheeled car Dymaxion Transportation Unit, but an unfortunate car accident just outside the 1933 Chicago World Exhibition gave the project such bad press that the project in the end was cancelled.

The Dymaxion house was rather unconventional, hexagonal in ground plan, without loadbearing walls and only supported by a central mast. The whole house was elevated a full story from the ground providing parking space underneath it. Buckminster Fuller is best known for the development of the geodesic dome. Being one of the most important architectural innovations of the 20th century, the geodesic dome is a covered, self-supporting structure made of tetrahedronal elements that can be massproduced. Buckminster Fuller began working on the geodesic dome in 1948 and received a patent for his design in 1954, but already in 1923 Walter Bauersfeld had showed the efficiency of the design in the Zeiss I Planetarium.

A geodesic dome is a highly optimised structure; using the shell-effect to minimise bending stresses and efficiently cover a large area. At the same time the self-weight is limited by converting the shell into a grid of similar elements that can be mass-produced. By creating a triangular or hexagonal grid of structural elements, the membrane in between the elements could be done as planar elements, which is much easier to mass-produce. The geodesic dome geometry is based on a network of circles placed on a sphere. The intersection points of the circles are then connected to form a triangular mesh. Buckminster Fuller was teaching at a college in North Caroline during the summers of 1948 and 1949. Here, with help from students and professors, he reinvented the geodesic dome 89

89: The original American Pavilion for the 1967 World Exhibition in Montreal. In 1976 a fire burned away the buildingâ&#x20AC;&#x2122;s transparent acrylic bubble, but the steel truss structure remained. Image from a â&#x20AC;&#x2DC;67 Expo post card

167

ENGINEERING ARCHITECTS PAST 168

and made it popular. It was a result of Buckminster Fullerâ&#x20AC;&#x2122;s exploration of natureâ&#x20AC;&#x2122;s constructing principles to find design solutions. Already in 1945 he had made an early model, constructed at another college in Vermont, but the first geodesic dome that could support is own weight without practical limitations was erected in 1949. It was made of aluminium aircraft tubing and vinyl-plastic skin. The dome was 4,3 meters in diameter, and to prove the design he suspended some students who had

helped him build it from the top. Buckminster Fuller had a good relationship with the US army. He had tried to set up production of a second version of his Dymaxon House in an aircraft factory, a production line to be run by soldiers returning from the war. He was able to convince the US government of the possibilities within his new company, Geodesics, Inc., and starting with small domes for the Marines to shelter radar crews in remote

90: Photo from 1972 of Richard Buckminster Fuller holding a tensegrity sphere Image courtesy of AP Photo/Bill Ingraham

location, within a few years he had built several thousand all around the world. Buckminster Fuller was appointed the architect of the US Pavilion for the 1967 World Exhibition in Montreal. The dome, 76 meters in diameter and 62 meters high, brought the final international breakthrough for Buckminster Fuller, who during the 1950s had started to be internationally known. The geodesic dome has several advantages, but the hemispherical shape makes it primarily useable in industry and entertainment applications. Few have found it suitable for living.

with limited resources. He had a clear social agenda to make the world a better place through better design. Buckminster Fuller disliked being considered an architect. To him, architects were exterior decorators tied to tradition. Despite that, he was pleased to accept the American Institute of Architecture’s gold medal in 1970, for his lifetime contribution to architecture. 14

Buckminster Fuller used the engineering principles of the geodesic dome to develop another structural concept; tensegrity structures. Tensegrity structures are structures that consist of members only exposed to either compression or tension. Examples of tensegrity structures are skeletal structures, who’s tension rods are held together solely by members in compression. Another example is the octet truss; a space frame where tension cables and compression rods interlocks each other in a very efficient system.

Buckminster Fuller was inspired by nature, and he sought to find the architecture of nature. To Buckminster Fuller:

“The opposite of nature is impossible” 13

He published more than fifteen books and through many of them he worked hard to convince people that the earth is a closed system

91: Buckminster Fuller holding a vertical cantilevered tensegrity beam Image courtesy of Buckminster Fuller Institute

169

170

ENGINEERING ARCHITECTS PAST

171

ENGINEERING ARCHITECTS PAST

FREI OTTO Frei Otto (b. 1925) is one of the leading designers of tensegrity structures. Allegedly his upbringing in Nazi Germany, with its heavy monumental buildings, inspired him to become a designer of lightweight and adaptable architecture influenced by nature. Otto’s career can be compared with Buckminster Fuller’s architectural experiments; they where both teachers at Washington University in St. Louis in the late 1950s, both experimented with inflatable structures and both were architects of pavilions for the 1967 World Exhibition in Montreal. But mostly, they were both designing space frames and concerned with structural efficiency. Otto had been granted a visit to the United States while studying at the Technische Universität in Berlin in 1950, where he saw a sports arena in North Caroline that caught his interest. The roof structure was made with hanging steel cables and this stimulated his interest in large-scale roofs and tent structures. Tent structures had been seen as instable, cheap and temporary structures until this point, and Otto spent the following decades to convince the world of the possibilities within tensile architecture. In 1954 he published his studies in a dissertation called Das hängende Dach – The suspended roof. In fact tensile structures are the opposite way of solving the same problems as the great concrete shell engineers like Gaudí, Candela, Torroja and Nervi had solved through a form

172

finding process trying to eliminate tension forces. Otto’s idea is to use the principles of the hanging chain directly and not turn the model upside down. It is of course not that simple; Otto’s designs are made in space’s three dimensions and with multiple boundary conditions. He explored that as an architect he was not able to freely sketch the shape of a tensile structure, but the structure should be allowed to find it’s own shape – it should find it’s equilibrium. Otto called this design process for form-finding, an expression widely used today. Otto’s process is perhaps the simplest way of form-finding as there is normally only one solution to be found, but it does not mean that the process is simple. Much of Otto’s career has been within the field of research. In 1958 he founded a development centre for lightweight structures in Berlin and in 1964 he was offered a position in Stuttgart as the director of the newly founded Institut für Leichte Flächentragwerke – Institute of lightweight structures. Already in 1958 Otto had published a paper on adaptable structures – these are moveable structures or structures that are easily dismantled. It was Otto’s belief that conventional buildings were very hard to adapt to new users or if the intended use was to change. Retractable roofs, which could react to, for instance, changes in weather conditions and create shelter, had been known since the ancient Greeks, and Otto sought to reawaken the interest in them. His first retractable roof was for an open-air theatre in Cannes, France, completed in 1965, and soon after for another theatre in a ruined church. Otto has completed

92: (Previous page) Frei Otto and Günter Behnisch tensed roof structure for the Munich 72 Olympic Stadium Image courtesy of Jorge Royan

several projects in France and Germany with retractable membrane roofs for swimming pools since. Through the institute in Stuttgart he could spread his ideas around the world; the Stuttgart institute of lightweight structures is a laboratory carrying out research that is linked directly to the basic principles of architecture and engineering. People within the building industry could read the many research papers published by the institute and attend the numerous conferences and experiments carried out, and the attendees then took Otto’s ideas back home with them. The first cable-net roof designed by Otto was for the Swiss National Exhibition in Lausanne, completed in 1964. Together with Rolf Gutbrod

he won the competition to design West Ger- 93 many’s pavilion for the 1967 World Exhibition in Montreal, a structure that inspired Günter Behnisch in his design for the Munich Olympic Park for the 1972’s Summer Olympics. Otto collaborated in the construction from 1968 to 1972. This is perhaps the most famous of Otto’s structures. In the 1960s Otto started to invert his structures, using the same principles to create shell structures. It was still his belief that the form should be directly taken from nature – from the natural way a surface would rest in equilibrium.

“I try to understand nature, even though I have realised that nature can never be understood by a living creature who is a part of nature him-

93: Tent structures in the Munich Olympic Park Image courtesy José Calijuri Hamra

173

ENGINEERING ARCHITECTS PAST

self. ”Less is more” is a saying that fascinates me: needing fewer houses, fewer materials, less concrete and less energy, but building humanly, using the elements that are available: earth, water, air. Building close to nature and making a lot out of little, observing things critically and thinking them through before putting pen to paper. It’s better to hardly build at all than to build too much!” 15

Otto also made research within inflatable buildings, an extreme version of lightweight structures. He was one of the first to call for more environmentally and ecologically better buildings, and he constructed Germany’s first passive solar house in 1967.

174

94: Form-finding model for the ‘Multihalle’, in scale 1/100 ca Image courtesy Instituts für Leichte Flächentragwerke

95: The Manheim Multihalle designed by Frei Otto and Arup, completed in 1975 Image courtesy unknown

175

ENGINEERING ARCHITECTS PAST

CODA CHAPTER SEVEN The field of optimised structures is not only dedicated to engineers. Architects saw the opportunities and advantages optimisation brought, and some architects made it their specialty. Architects used engineering as tool for evolving the building industry and to make new inventions as the geodesic dome, tensegrity structures, and cable and tent structures. Lightweight structures, with Frei Otto and the Institute for Lightweight Structures and Conceptual Design at the University of Stuttgart, have developed to become one of the most interesting research fields within the optimisation of structures at present day. Engineers can tend to keep within the boundaries of their knowledge; use well-known solutions and materials. Advances within building technology demand the engineer to step out of his comfort zone. Examples from present day include: use of new materials as fibre-reinforced concrete or fibreglass; new ways of fire prevention by limiting the amount of oxygen in the air; a structure that respond to the load it is exposed to. The engineer needs to be aware of advances from scientific research, mainly materials. Glass fibre reinforced concrete (GRC) is becoming a common material, but other alternatives as steel fibre reinforced concrete are also showing potential.

176

In Denmark I once worked on a competition project with an architect who had read about how a fire work â&#x20AC;&#x201C; it needs oxygen. He proposed to lower the oxygen percentage in the air inside the building, so a fire could not burn. The idea is interesting, albeit a little crazy. As the head of the Buildings department at Polytechnic University of Denmark (DTU BYG), Michael Havbro Faber, states:

Engineers must free themselves from the norms. [...] The engineers are really misled to fulfil all these strange regulations instead of thinking for themselves.

The research made in the University of Stuttgart is to me very interesting. In 1994 Werner Sobek assumed the chair of Frei Otto as principal of the Institute, and in 2001 added that of Prof. Dres. JĂśrg Schlaich. At the University of Stuttgart researchers have experimented with Ultra-lightweight structures that respond to the applied loads, for instance with hydraulic supports. Sobek, Schlaich and others as Santiago Calatrava, are all people that could easily have made it into this thesis, but, as with Ney, I have chosen not to include them here as they fall out of the context of time - and their influence on the engineering history is hard to determine.

96: The “Stuttgart Smartshell” is an adaptive structure that reacts to the load applied on it. The shell is a product of the research done at the University of Stuttgart. Image courtesy Universität Stuttgart

177

OVE ARUP PAST

Chapter eight

Ove Arup

There exist two kinds of architectural-engineers. The one, described in chapter six, is the architectural-engineer who works independently with structures, optimising the structures to the point where only one true solution is left. Normally this kind of architectural-engineer only acts and succeeds in the field of large-scale structures, like stadiums, bridges or towers. The other type of architectural-engineer is the engineer who works in cooperation with architects.

Ove Arup (1895 – 1988) was perhaps the first really successful architectural-engineer of the second kind. Collaboration between architect, engineer, and builder was not a new idea, and since the early nineteenth century architects and engineers had worked together to create holistic designs, which were greater than the sum of their individual parts. But examples of projects with collaboration between architects and engineers from the earliest part of the

178

nineteenth century mostly cover large scale buildings, where the architect was more or the less forced to collaborate with the engineer in order to carry out the project. Ove Arup took the collaboration to other scales of buildings. He invented a new way for a design process to work, one where it was normal for engineers to participate in the creative process just as much as architects. In 1960 Sir Alan Harris, an engineer specialised within pre-stressed concrete who had worked as Freyssinet’s chief engineer in London, wrote Ove Arup to thank him for attending his lecture at the Royal Institute of British Architects:

“To see you in the audience made a very piquant situation. It prompted me to write you a line to say something which has been on my mind for some time. Those engineers who work with good architects in a spirit of harmony and common aims are really a new race,

created by, and creating, a new sort of architecture and, thank god, a new sort of architect. The whole situation is almost the invention of one man, yourself, and both architects and engineers should be profoundly grateful for it. Your capital contribution has been little understood; rest assured that I at any rate appreciate it. While there are a lot of us elbowing one another to follow you, you blazed the trail.” 1

What is architecture? Despite modernists’ movement towards prefabrication and simplification, architecture has in general always sought uniqueness in every single project. Architecture could be described as art that has a function – and even if the functions are similar from project to project, the art is never the same. Architecture is contemporary. It is always moving, it is never the same – architecture is always new. When Wolf Prix, cofounder of Austrian architectural office Coop Himmelb(l)au, in an interview with an online blog was asked “What is architecture?” he answered: “YES!”. In the same interview he elaborated his view on his design method and how his office always seeks new options. He stated:

“Complex problems only can be solved by complex solutions. The advantage of the complex solution is, that complex solutions are always new – very hard to understand, but always new. Simple solutions are very understandable, but they are always old.” 2

Ove Arup was a pioneer within the architectengineer relationship. He had his ideas and philosophies on how the best solution was reached. He believed that complex solutions only could be solved by teamwork, as no individual possess all necessary expertise to create a highly complex product. Ove Arup’s ambition was to create perfect harmony of architecture and engineering, something that he meant demanded changes from both architects and engineers. Inspired by Walter Gropius’ Bauhaus, Ove Arup spoke in favour of “Total Design”, where the terms Architect, Engineer and Builder belonged to a bygone age. Architects had to be taught engineering, philosophy and self-critical communication skills, engineers had to be taught draughtsmanship, design and aesthetics, and they had to learn to work together. A big part of Ove Arup’s legacy is how he shaped his now world famous engineering firm. Much of the work that made his company so famous was not done by Ove Arup himself, but was a derivative of his great ability to find engineers with the same belief as himself.

THE OUTSIDER “When I was just five or six years old, I wanted to be a master builder.” 3

Ove Arup has been described as “Master builder of the twentieth century”, but his road to becoming that was not straight, and per-

179

OVE ARUP PAST

haps that is one of the reasons why he became so successful. Ove Arup was born in Newcastle upon Tyne in North East England to a Danish father and a Norwegian mother in 1895. His father was trained as a veterinary surgeon and had moved to England to work as a government consultant, supervising the health of beef cattle being shipped from Denmark. Soon after Ove Arup was born, Britain banned the import of live cattle, and the family was forced to move to Hamburg. German became Ove Arup’s first language, so when he was sent to a boarding school in Denmark in 1907, he felt handicapped by his imperfect Danish and became a bit introvert as a person. At the school, Sorø Academy, Ove Arup learned to play piano. He was a good scholar, and he thrived the conditions the school provided, encouraging the pupils to think for themselves and open their eyes to the world around them. For a lecture arranged by the student committee at Sorø Academy, fifteen-year-old Ove Arup gave a glimpse at his characteristic as he spoke “On hospitality and entertaining”:

“What then, is entertaining? […] Here, everything works according to set rules and regulations, and adhering to these is considered good manners. […] It is important that you gather as grand a party as possible so that anyone may see that you socialise with persons of rank. […] Of course you must attend wearing a certain kind of clothing, stick to cer-

180

tain formalities, dance with the daughters of the house if you are attending a ball, praise the food, etc. What you actually do is eating several dishes. Make conversation, eat again, perhaps listen to a boring pianist, smoke, drink wine, eat fruit and dessert until at last it is acceptable that you leave. Of course you have had a splendid time, and you are looking forward to receiving the host and the hostess in your home in another occasion. To put it briefly, then, it is all a matter of appearances rather than reality, you say a whole lot of things you really do not mean, etc.” 4

The text shows how Ove Arup personally had a detachment and ironic approach to the social etiquette of the day – at an early age he questioned the prudency of his surroundings, could things not be done differently? He had been raised in a wealthy family with social Christian values - values that were deeply founded in his personality. Throughout his life he helped many people financially with gifts or loans. His family was from the upper middle class and he did not have to worry much about financial issues. One friend described him as a “champagne socialist”.

“For most people, wealth is not a beneficial thing. […] Only when you spend your wealth on some great cause to the benefit of mankind, or to reach scientific results that you work hard to attain, or to help some of those thousands of people suffering in the world, only then do you benefit from wealth. Wealth will damage all

those people who can’t resist its dangers, and it won’t benefit the particularly.” 5

Ove Arup went to study philosophy at the University in Copenhagen. After his graduation he was disappointed not being offered a vacant lectureship, which instead was offered to his friend. During his philosophy studies he decided to leave the Lutheran church, into which he had be confirmed at school, as he had realised he did not believe in a god. His professor was teaching a philosophy class in religion at the time which might have encouraged him to leave the church, but Ove Arup also questioned fundamental philosophic topics regarding the conflict between thought and action which led to several discussions with his professor. He rejected dogmas and doctrines of the Lutheran church and the abstract universal prescriptions of for instance Kant. One of his favourite philosophers was Søren Kierkegaard. He had experienced a bad heartache at his

time in University and his friends, most of them artists, could easily distract him from the studies to attend a party or go sailing. Fifty years later he admitted that while studying he had:

“Got rather into a muddle – a lack of control over one’s personal problems, feelings and behaviour – altogether taking oneself far too seriously.” 6

After finishing his philosophy studies, Ove Arup did not know what to do with his life. He “abandoned philosophy”, as he wanted to escape a life where everything was theory and none of it was practice:

“Closer acquaintance had made me realise that [philosophy] consisted of a series of specialised disciplines, Theory of Knowledge, Ethics, Psychology and so on, all involving a lifetime immersed in books, and none of them answering the question I had set out to solve. 97

97: Portrait of Ove Arup, philosopher and engineer Image courtesy of Arup

181

OVE ARUP PAST

In fact in philosophy there seemed to be no answers, only more and more subtle questions. Science, or any logical reasoning, cannot solve the human predicament: What is the whole thing about? What are we? Where are we? What are we suppose to do? What is good or evil? Can we, or should we teach the tiger not to eat meat? The greatest happiness of the greatest number? And what is happiness, and Truth, and Beauty? I revolted against ideologies, philosophical systems, moral codes, on Kant’s insistence on rectitude in preference to kindness.” 7

Ove Arup had been fed up with theory and wanted to create something real. To become an engineer seemed like the good choice, performing practice through theory, but he was not sure this was what he wanted. He halfheartedly enrolled in the engineering course at the Polytechnic Institute in Copenhagen, where in the beginning he spent most of his time with music and social life. Many of his friends were artists, and he thought that by becoming an engineer he as well could “be an artist on small matters”. He sought the satisfaction of creating something.

“[I was not sure I had] enough artistic ability to become a really good architect. […] I cannot say that I felt wildly enthusiastic about [studying engineering], I disliked the idea of specialising, and felt that I was perhaps giving up too easily. […] I realised that human relations are important, that art is important; listening

182

to fantasia and fugue by Bach. I felt that here was something that was good in itself no matter whether the mysteries of this world were solved or not.” 8

CHRISTIANI & NIELSEN AND THE FIRST YEARS IN LONDON In 1902, as a newly graduate from the Polytechnic Institute in Copenhagen, Rudolf Christiani went to visit Hennebique’s Paris headquarter, and he ended up staying there for two years. On his return to Copenhagen, together with Aage Nielsen, he formed the company Christiani & Nielsen that specialised in reinforced concrete and construction. They constructed the forty-meter wide roof dome in the Circus building in Copenhagen, but the company had a reputation mostly for their underwater work. Christiani & Nielsen secured their expertise by ensuring that they always recruited the best graduates from the Polytechnic Institute, and when Ove Arup applied in 1922 he was immediately accepted. At this point Christiani & Nielsen had expanded their business into other European countries as France and Germany, and Ove Arup was posted in his childhood city of Hamburg. Christiani & Nielsen were building harbours, jetties and bridges. Ove Arup faced engineering problems only, and aesthetic considerations did not apply. He had to design a structure that resists the forces, and he had to understand how that structure performed in the various tid-

al and ground conditions that might arise. Ove Arup took nothing for granted, he was curious and wanted to completely understand what was going on inside and outside the structures he designed. From the beginning he sought to understand and to re-examine the nature of the loading that jetties must carry, and to find the most appropriate structural form to resist them. In 1935 he published his research, where he proposed to use full-length bracing and raking piles as more efficient and as a more correct engineering response than the conventional solutions then being used. In his research he examines these conventional solutions of the stiff deck on vertical piles, and compares the failure modes of the different structural forms.

to improve on the structures then being built.” 9

In Hamburg Ove Arup witnessed the extreme inflation Germany went through in beginning of the 1920s. On the weekly payday, Ove Arup had to go directly to the bakery because the prices rose the next day. Prices were regulated hourly, a gramophone could cost 5.000.000 marks at 10 a.m. but at 3 p.m. it was 12.000.000 marks. In November 1923 Ove Arup pressed for a transfer to Paris, which he thought was exotic, vibrant and surely full of opportunities. He had his cousin Arne, based in London, to help him draft a letter to say that: “any man who has even the slightest chance of getting out of the country will try his very best to make the most of such chance”. The

Peter Rice described the research as:

“It is not a complex analytical treatise, but a series of pragmatic arguments justifying a clear engineering choice made, I suspect, instinctively. It is the work of a true engineer, not that of an intellectual playing with engineering He was working in an environment dominated by the classic engineering values: practicality, simplicity, adequacy and cost. His proposals are interesting from another point of view. Every proposed solution is related to, maybe even derived from, the construction method. And the construction method that most interested him was building in reinforced concrete. The advantages in durability, and flexibility of form were properties that could be exploited

98: Ove Arup’s first building, the Labworth Cafe, shows how he was inspired by Modernism Image courtesy unknown

183

OVE ARUP PAST

result was, that he was employed at Christiani & Nielsen’s London office by the first week of December 1923. In London Ove Arup felt lonely. He did not speak the language very well, and he was once again heartbroken. He felt “rather lost” and in the beginning he socialised mostly with other foreigners where social, aesthetic or moral problems were not topics being discussed. In the beginning of 1924 he wrote in his diary:

“I’m lonely and have nowhere to put all that tenderness etc. that I would like to put somewhere. Any connection between [Elsebeth] and myself has now been completely broken off… I really have ended up as a square peg in a round hole, seeing that I’m actually not a hands-on engineer, damn it, and all in all I’m dissatisfied with this big city drudgery. […] However: from the point of view of an energetic man, I may be envied. I have been appointed to take care of the Southampton job – i.e. I have a specific field where I can show what I’m worth and which I can take an interest in.” 10

In Christiani & Nielsen’s London office, Ove Arup was working under Herluf Forchhammer. Ove Arup described Forchhammer as the “inspired engineer”, something he honoured. Forchhammer’s main mean for securing contracts was good design, and he had established a system of circulating technical reports on interesting jobs and new calculation methods, something Ove Arup later copied when starting Arup. When looking back, Ove Arup recalled:

“It has been my – I almost said bitter – but at any rate frustrating experience, that the paramount of importance of getting the right design is hardly understood by laymen, including clients, is rarely grasped by building authorities and the legal profession concerned with building, often not by architects who leave the costing of jobs to quantity surveyors and therefore lose touch with the method of building, or by consulting engineers, who concern themselves exclusively with structural stability, but leave matters of construction to the contractor.” 11

By mid 1924, in a conversation with a friend, Ove Arup expressed his considerations to continue as an engineer or switch to architecture._12 He half-heartedly decided that he possessed the mathematical skills necessary to do engineering, but he probably lacked the creative skills sufficient to be an architect. Instead he chose to do what a philosopher only could do: he redefined his engineering field, so

184

99: Ove Arup on site of at London Zoo’s Penguin Pool designed together with Lubetkin Image courtesy of Arup

that all manmade structures to him were classified as architecture. 13 In early 1925 Ove Arup met Ruth Sørensen at a tea-dance at the Anglo-Danish society in London. He was thirty and she was nearly twenty-three and they married on August 13th that same year. Soon after the marriage, Ove Arup published his first technical paper. In a reply to an article in Concrete and Constructional Engineering, Ove Arup challenged the generalisation that architects are planners of structure and engineers work more as “technical advisors”. Ove Arup argued of two scales of structures:

“[With] engineering works, such as arch bridges with big spans, silos, hydraulic structures etc. […] it is the engineer who of necessity is the planner of the structure […] whereas the architect acts as adviser in matters of detail. […] The best architecture in reinforced concrete is generally to be found among those big engineering structures. […] A well-designed engineering structure generally possesses a quality […] accepted as an essential feature of good architecture, namely “truth”, whereby it is implied that the purpose of the building […] is met in a simple, logical and economical way, and that this purpose is openly and frankly expressed in the building, without disguise of any kind, but with respect for, and knowledge of, the material employed. Knowledge of the material belongs to the sphere of the engineer. […] The purpose itself is simple and clearly stated. And naturally imposes itself as the main consideration.

In ordinary building […] the architectural problem is much more complicated. There are a hundred different considerations to conciliate. The possible solutions are much more numerous. […] The architect is naturally inclined to apply traditional forms and proportions, which were natural for brick and stone, but which do not fully utilise the extended possibilities of the new material.” 14

At the same time as Ove Arup was completing his engineering studies, Le Corbusier published a collection of his writings: Vers une architecture. Here Le Corbusier proclaimed that genuine architecture was to be found among the work of engineers, as only engineers could follow the laws of nature. Corbusier’s writings interested Ove Arup, firstly because of the celebration of engineering, and secondly because of Corbusier’s homage to monumental architecture of the past and how he described that one can only build upon the past by honouring one’s debt to it. The Modernist architecture of Le Corbusier, which appealed to Ove Arup, was being justified by referring to a familiar past. 15 Another person that appealed to Ove Arup was the founder of the Bauhaus School, Walter Gropius. Through his school, Gropius aimed to fuse the fields of art and technology, theory and practice. The idea of unity and Corbusier’s celebration of engineering as the field that could unite provided the foundation for Ove Arup’s architectural engineering philosophy.

185

OVE ARUP PAST

Walter Gropius (1883 – 1969) was a German architect who founded the Bauhaus school. Ironically Gropius could not draw so he hired an assistant to complete his homework in architecture school. In 1908 he was employed at the firm of Peter Behrens, where he worked together with Mies van der Rohe and Le Corbusier. Together with a friend he established his own architectural practice in Berlin in 1910. In 1919 Gropius was appointed as the new master of the Grand-Ducal Saxon School of Arts and Crafts. Gropius merged the school with Weimar Academy of Fine Art to form the “Bauhaus” school, with the idea of creating a school where “total design” was being taught. The school was inspired by the early Modernism and Behrens’ integration of art and mass production at AEG. The education sought to unify art, craft and technology, where the students learnt the basic elements and principles of design, colour theory, and experimented with a range of materials and processes in their first years. The school aimed the students to create products that could be mass produced. In the beginning the school included everything from bookbinding over pottery and painting to sculptor, but architecture classes were not offered until 1927. The school was name for architectural projects before 1927, as all of Gropius’ own work in the period was also the work of his school. Gropius proclaimed his goal as being “to create a new guild of craftsmen, without the class distinctions which raise an arrogant barrier between craftsman and artist.” Gropius thought that art could not be taught, but craft could. Students were accepted at the school as apprentices, and gradually increased in rank, first to journeyman and finally master. The Bauhaus school had a big influence on design educations, and its approach to design education became a common feature of architectural and design school in many countries. In 1928 Hannes Meyer took over the school from Gropius and in 1930 Mies van der Rohe became its director, but it was closed by the Nazi regime in 1933. Many of the artist connected to the school fled out of Germany and where they spread the ideas which had a major impact on the art and architecture trends in Western Europe and United States in the following decades.

186

100

J. L. KIER & CO. AND CONNECTION WITH AVANT-GARDE ARCHITECTS During his first years with Christiani & Nielsen, Ove Arup worked mainly with harbours and industrial buildings, and the first building with which Ove Arup was closely involved was a small café in 1932-1933 – in fact Ove Arup acted as both the architect and engineer of the café. He had for some time been telling his family about growing differences with his boss at Christiani & Nielsen, so when he was offered a position at another Danish company in England, J. L. Kier & Co. as director responsible for designs and tenders in 1934, he was soon to accept the opportunity. 16 In the beginning of 1933, when Hitler became Chancellor of Germany, the political situation in Germany led many intellectuals and artist to emigrate to England, among them architects Berthold Lubetkin and Walter Gropius. Lubetkin was educated in Moscow, Berlin, Warsaw and Paris, and had worked with Perret in Paris where he had learned the possibilities of reinforced concrete constructions. Lubetkin had contacted the Paris office of Christiani & Nielsen for advice, and they referred him to their London office and Ove Arup. At Kier, Ove Arup had the chance to develop his new association with Lubetkin. In 1932 Lubetkin had formed the architectural partnership “Tecton”, and together with Ove Arup they developed projects as the building blocks “Highpoint” I and II and the Penguin Pool at the London Zoo. At Tecton Ove Arup met architects who were

interested in new ideas, and professed enthusiasm for engineering for the functional use of structural materials. It was the ideal of Bauhaus, but to Ove Arup looking back, it was not a genuine enthusiasm for engineering but rather narrow-minded obsession with an architectural style:

“[At Tecton I] met a number of young people who really were interested in new ideas, who in fact had plenty themselves, and were very fond of discussing them. It was stimulating, amusing, and also puzzling. […] They were in love with an architectural style, with the aesthetic feel of the kind of building they admired; and so they were prepared and indeed determined to design their buildings in reinforced concrete – a material they knew next to nothing about – even if it meant using concrete to do things that could be done better and more cheaply in another material.” 17

From the end of 1933 Ove Arup started to work closely with Lubetkin. He was later always keen to acknowledge Lubetkin as his first real teacher of architecture:

“He taught me how good architecture was produced, and what a serious business it was. [It] involves taking infinite care over every detail.” 18

Together they designed the award winning Penguin Pool ramp at the London Zoo. Ove Arup performed complex calculation for the

100: (Opposite page) The Bauhaus School in Dessau Image courtesy unknown

187

OVE ARUP PAST

special geometry of the ramps. For the calculations the ramps were conceived as beams. The inner edges had a slightly increased thickness to enable the cross section to withstand torsional forces that arose because of the geometry. For the Highpoint project, designed with the Tecton group, Ove Arup contributed to both the construction and to the design. In the early 1930s he had developed a system of climbing formwork, which could be raised by jacks, pouring one meter at the time, which was the maximum permitted at the time.

101

188

â&#x20AC;&#x153;At that time [1934], tall buildings were supported by an orthogonal frame - work of structural steel - the normal steel skeleton which is

101: The completed Penguin Pool at London Zoo Image courtesy of Arup

still in use [1974]. The calculations were crude â&#x20AC;&#x201C; no account taken of the strength of a frame, simply a matter of columns and beams, and the more steel the better, since that meant bigger profits for the suppliers. The steel suppliers generally threw in the simple calculations required, and also the actual erection of the steel framework, a job they could perform quickly and efficiently, and without excessive demand for building space in a built-up area. When reinforced concrete was introduced, the steel framework was simply replaced with a reinforced concrete framework. If the wall panels too were of concrete, the regulations required that they should be carried by the reinforced concrete frame: they did not contribute to the strength of the structure, only to the load upon it. In other words, plain facts were denied. When Lubetkin first set out plans for

Highpoint, he also had provided for columns and beams all over the place, and of course the whole building was also to be carried on stilts – that was a part of being modern. It also looked very well.

a lot of transverse reinforcement. Something 102 that would weaken the walls, Ove Arup argued, since it meant that the concrete had to be poured in a half-liquid state.

I at the time had long since, when building coal silos and the like, dispensed with columns in walls – and particularly at the corners, where they are totally unnecessary.” 19

“The structure of the Highpoint flats was in some ways the clearest example of a marriage of architecture, engineering and construction that [Ove] Arup achieved. The concept is simple. Walls are load-bearing, and when openings are required underneath to facilitate the architectural planning they act as beams. There are no columns or beams as such, just walls and slabs. The architecture demanded some engineering compromise, but the concept works, and works well. The construction

Ove Arup’s well foundation in the theory of concrete structures also came to show on the Highpoint project. To conform to building regulations he had to agree to use the double reinforcement of what was needed, including

102: Highpoint I apartment block Image courtesy unknown

189

OVE ARUP PAST

method was developed from silo construction with sliding shutters for the vertical walls.” 20

Ove Arup himself also took pride in the project, which promoted both Lubetkin and his own career, as he described how the significance of the Highpoint for him:

“lay in the almost perfect integration of architecture, structure and building method.” 21

The Highpoint project won an architectural price, but in Ove Arup’s old days he had become more critical; it did not fulfil the expectations he had of the project before the start. It should have resulted in cheap apartments for the working class, but construction cost made it too expensive for a middle class worker to afford. Ove Arup later concluded:

“It’s a muddy kind of structure, one doesn’t know exactly how the stresses are distributed.” 22

Lubetkin was a member of the communist party in England, and Ove Arup and Lubetkin strongly disagreed on ideological topics, but they worked together in relative harmony for several years. But when looking back on the Highpoint project Ove Arup claimed that the architects:

190

“simply got what they wanted [by consulting the client.] - What does the client know about architecture? [The engineer was not asked] – his job is to keep mum. 23 Architects don’t mind cheating. Lubetkin would tell any lie to the client. A wall like the one at Highpoint would have been much cheaper to build with bricks, but he claimed it was functional and economic. It wasn’t functional at all: it had to be ‘Modern’. Functionalism really became a farce. What is wrong with a sloping roof? They can’t afford to pay what it costs to make a flat roof really waterproof. Lubetkin didn’t care. He just cared for the picture in the architectural magazines. There is such a lot of humbug in architects, but there is such a lot of stodginess in engineers. I am almost in favour of humbug, temperamentally. 24 [Lubetkin] often told me himself, that he is not interested in Truth as such; for him any statement, spoken or written, is just propaganda to further an aim which is considered to be of overriding importance for mankind. This is of course the normal communist attitude, which I know only too well having clashed with it on numerous occasions. 25 I could only work with Lubetkin because as an engineer with what was at that time an unusual experience in the design and construction of reinforced concrete, I had the last say which Lubetkin had to respect; otherwise I would have been swallowed up by Lubetkin’s forceful personality. [...] We more or less agreed to put aside our basic conflict of ideology.” 26

103: (Opposite page) Corner detail of Highpoint I, showing balcony profiles Image courtesy of Fin Fahey

191

OVE ARUP PAST

Despite Ove Arup’s later displeasure on the project, it became widely renown as the finest example of this form of construction for residential purposes when the building was completed. The Highpoint project gave Ove Arup the reputation among architects of being able to do “tricks” with reinforced concrete. This was clearly not the intention of Ove Arup, as his own goal was simplicity. He believed that simplicity is a necessity in “total architecture”.

ARUP & ARUP LTD. AND THE SECOND WORLD WAR In Ove Arup’s first years in London he observed how the building industry of the day had a sharp distinction between consulting engineers who designed the projects and served the client’s wishes, and the contractors who only cared for profit. His first office, Christiani & Nielsen, was in fact mostly functioning as a contractor, but it preferred to also design their own jobs, something that influenced Ove Arup’s thoughts on practice. He had been transferred from the theoretical world of the university to the practical, real world of construction. He saw that concrete was made by a couple of buckets of sand and cement, by people on site who mostly did not understand. From those early years on, he knew that to design you had to know how to build. For Ove Arup the two went hand-in-hand. English clients of the day normally refused to adopt new ideas, and Ove Arup became frustrated by contractors’ emphasis on cost reduction. He argued that:

192

“Something should be done to make it more pleasing, more satisfactory aesthetically. […] Good architecture need not necessarily be more expensive. […] You cannot tack architectural advice on to an engineering scheme. […] All manmade features are also architecture, and must be judged as such.” 27

He found it difficult to collaborate with construction companies to build what he wanted – especially when dealing with new forwardlooking methods, materials and technologies._28 What Ove Arup sought was a company with an individual in overall charge of a project, who possessed an ability to judge architectural and structural quality, ensure functional efficiency, seek perfection and to keep all that within a budget. He himself could, if any, be that individual, a “designing contractor”. So in April 1938 he became independent, by starting the design and construction company Arup & Arup Ltd. with his cousin Arne A. Arup. Arne Arup had always been living in London, running his own company dealing construction materials. Half a year after the company had been established, Ove Arup had to focus on different activities as a result of the Munich crisis in 1938 that brought Europe on the verge of war. Already in 1937 Ove Arup had started to work for the British Air Ministry, constructing large reinforced concrete aircraft sheds. An act from the British governments required local authorities to undertake protection of life and prop-

erty from aerial attack. Ove Arup was commissioned by the London Borough of Finsbury to design community shelters.

“A complete scheme for the civil defence of the Borough […] was not an ordinary engineering problem where the objects to be achieved, the method of achieving them, and the stresses to be worked to are defined within narrow limits. It was rather a question of designing structures giving an undefined degree of protection against destructive forces about which very little was known. Not even the amount of money at our disposal was known.” 29

His proposal was forward thinking; designed as a helix spiralling six stories into the ground, one shelter could house 7600 people and after

the war the shelters could be converted into an underground car parking. Ove Arup also proposed alternative designs for shelters of different sizes, but none of his shelter designs were built. He did contribute to the war in other ways; in 1942, Arup & Arup started work on the secret underground headquarter for RAF Coastal Command located at Northwood, not far from Arup’s home in Harrow, and in 1944 he designed and built the floating jetties, a set of hollow reinforced concrete boxes that were used to create a temporary harbour needed just after the D-Day. During the war Ove Arup was supporting the Danish underground financially and he made radio broadcasts to Denmark through the BBC Broadcasting House in London. He opened his home to many Norwegian fighters, among them several ministers of the Norwegian exile government, and ‘Free Danish’ airmen. 104

104: Ove Arup designed the concrete caisons used for the temporary “Mulberry Harbour” on D-Day Image courtesy unknown

193

OVE ARUP PAST

OVE N. ARUP CONSULTING ENGINEER – AN ENGINEER WORKING WITH ARCHITECTS Arup & Arup had plenty of war time jobs, but although Ove Arup found himself working in close collaboration with architects in his position as construction entrepreneur and designer he did not find what he sought with this constellation. In his role as a contractor, he would try to keep the price down, sometimes to the point where it would conflict with his own innovative ideas. After the war Arup & Arup was financially not successful, and in 1946 the company was dissolved. During the war, Ove Arup had written a paper on “Science and world planning”. In this paper, he described his ultimate motivation to start what was to become one of the largest engineering consultancies in the world. In the paper he pointed out how modern buildings of the day were:

“Often badly planned, badly ventilated, badly heated, etc. In other words, only limited use is made of all the existing technical knowledge. [No one person could] hope to be familiar with the complete range of modern technical possibilities. […] New knowledge, new materials, new processes [have] so widened the field of possibilities, that it cannot be adequately surveyed by a single mind. [Because] our needs increase with the means [the problem arises of] how to create the organisation, the “composite mind” so to speak, which can achieve a wellbalanced synthesis from the wealth of available

194

detail. [This is] one of the central problems of our time.” 30

Having realised that in order to control design one had to be a consulting engineer, he started the new firm Ove N. Arup Consulting Engineer at the age of fifty-one. Even though he had spent twenty years in London by this time, the success of his new firm would depend on making new contacts. Through Lubetkin Ove Arup had joined MARS – Modern Architectural ReSearch – before the war. Well Coates, Maxwell Fry, and Lubetkin himself started the group and it was affiliated to CIAM (Congrès Internationaux d’Architecture Moderne). Ove Arup was the only engineer in the group, but more significantly, he was almost the only engineer with a genuine interest in architecture at the time. Nevertheless, the firm’s economy was tight the first ten years, with overdraft in the bank and Ove Arup had to take loans from members of Tecton and others to pay his staff at payday. Ove Arup himself was financially secured from investments in Denmark, which served as a psychological cushion for him. In 1949 the first partners were appointed; Geoffrey Wood, Ronald Jenkins and Andrew Young. From their time together in Christiani & Nielsen Ove Arup knew Wood, and Jenkins had originally been taken on by Ove Arup at Kier & Co. In 1949 the company employed twenty people. Ove Arup’s procedure when hiring, was to decide whether he could get along with someone and he cared less about the persons formal qualifications or accom-

plishments. He had the ability to put together an effective team: Jenkins was able to help out on any mathematical problem; Wood could get things done, both in the office and on building sites.

“The idea that an engineer should devote himself to working with architects was at least odd, if not faintly ridiculous. It was considered a marginal activity, one which you did in your spare time. It set Arup apart from the real engineers, a foreigner on two counts. It was an enormous affirmation of his beliefs.” 31

Ove Arup created a firm collecting a group of bright-headed specialist. As Peter Rice described it:

“I had joined Ove Arup & Partners because I had heard that it was a place where an oddball could fit. Engineering then was a very serious profession. Perhaps it still is. Engineers were expected to know what they were about, to have a natural feel for their profession. I was an engineer by accident, tentatively feeling my way to a career, without any natural instinct for engineering. The atmosphere of Arups helped me survive. Where then did this atmosphere come from? - clearly “the Old Man” was the fountain, but how? Why?” […] He was curious, anything and everything could interest him, so that he would always respond positively to any idea, any proposal, and

then see the other side. But the most striking thing about him was his humanity. He wanted always to see the context of any proposal, and to check its effect on the people concerned. In engineering he wanted to know how the structure was built; in architecture he wanted to know how the people would respond, how it would affect them. It was this humanity, this concern for the effect of our actions, which made him so influential on those he knew. He was difficult, exasperating, even-handed to the point of indecision. “On the other hand...” was probably his favourite phrase. He was also tough. When action was really needed he could take it. Usually this action was taken in the name of honesty. He was as honest and tough on himself as he was with others.” 32

Ove Arup was fully aware that he himself could not know everything there was to know about engineering or the construction industry. He assembled a group of specialised people and he knew how to use them. Ove Arup held that creative design was severely handicapped if not founded on a thorough theoretical background. But – “on the other hand” – he also saw problems with the specialised society:

“In some ways [specialisation is] the curse of modern civilisation. […] Specialisation is an evil because it makes a person narrow-minded and makes communication with others difficult. […] Imagination and invention – essential for a creative engineer – flourish on cross fertilisation from other fields of knowledge.” 33

195

OVE ARUP PAST 196

105: The Sydney Opera House Image courtesy of author

197

OVE ARUP PAST 106

198

The Arup firm lived mostly in the shadow in it’s first years. Early projects included large concrete shells for bus garage in Dublin, as well as domestic apartments constructed in reinforced concrete. The project that brought Arup onto the global stage was an international competition for an Opera House in Sydney in 1956. The competition was won with beautiful pencil sketches by Jørn Utzon, at the day an unknown quantity in the architectural world. Utzon was Danish, and Ove Arup thought that here was a Danish architect about to undertake a large commission on the other side of

the world; surely he needed Danish help. Ove Arup wrote Utzon, what is later described as his “cheeky” letter, proposing Utzon to use Arup as the engineering consultant on the project. 34 It is sure to say, that Utzon’s design for the Sydney Opera House would never have been built without the engineering team of Arup. In his letter to Utzon, Ove Arup pointed out that the shape of the shells made them very difficult to construct in pre-stressed concrete. The collaboration between Utzon and the Arup engineering team showed to be essential in the

106: Kingsgate Bridge at Durham University was one of the last projects where Ove Arup was personaly involved Image courtesy of Gareth Gardner

case of the roof shells, as Utzon, the sculpturer, was not able to make the shells stand up, and the engineers could not stand up what ceased to be mere sculpture. They had to work together on the design: preceded by mutual education. 35 Utzon was at the time an inexperienced architect, who never cared much for any budget or deadlines. The project was a big scandal, ending up with Utzon resigning from the project in 1966 and it was discovered that Utzon had produced almost no architectural drawings of the interior of the building. Utzon mostly cared about the exterior perception of his building, and even after fifty years of modification of the Sydney Opera House, its internal functions remains inadequate and the acoustics mediocre. Arup showed their innovative ability to use new methods by applying early computer methods when developing the geometry. The design work on the shells involved one of the earliest uses of computers in structural analysis, in order to understand the complex forces to which the shells would be subjected to. Computers were also used for drawing three-dimensional shapes, and in setting on site. 36

OVE ARUP’S LEGACY

107

As the firm he had started to grow and prospered, Ove Arup’s role as engineering leader changed. He became a figurehead, defining the standards, but did not have a finger on each project anymore. Two projects, where he had a deep personal involvement, exemplify his contribution, his way of working. One was the Sydney Opera House, the other the footbridge over the River Wear in Durham.

“The Kingsgate Footbridge, Durham, was the engineering jewel of his later life. He received the commission in 1961 and worked on it for about a year before going to tender.” 37

The bridge was to link the University and Dunelm House across the River Wear. Ove Arup functioned as both architect and engineer on the project, and because of a small budget, Ove Arup was given as much time as he needed by the client to develop the right design.

“From the beginning Ove asked himself, how would he build it? That became the motor of his ideas. He produced a whole series of ide-

107: Images showing the phases of construction of the Kingsgate Bridge Image courtesy of Arup

199

200

OVE ARUP PAST

as, which he finally reduced to six (the originals have unfortunately been lost). These six ideas were then developed to assess their value as architectural solutions for the space.” 38

In the end Ove Arup choose a solution where the bridge was constructed in two parts on each riverbank, and then swung into the final position. The concept was two panels, each erected parallel to the riverbank where scaffolding could be placed out of the water. Throughout the eight months erection process he kept improving the design;

“In reality he never stopped designing right to the time the concrete was poured. The design could always be improved. He behaved like the most fastidious architect. […] The precise form of each leg, the nature of the hand rail, the relationship of the bridge to the embankments, were painstakingly researched and improved. The whole endeavour was treated with enormous attention to detail, a perfectionist seeking perfection, but within the framework of an engineering concept drawn from the most basic of principles.” 39

Spanish engineer-architect Felix Candela described his feelings about the project:

“When I saw the pictures of Ove’s tiny footbridge at Durham University, not only I said “of course”, but felt a pang of jealousy, because nothing make me more jealous than seeing a

true work of art.” 40

In 1960 Ove Arup was awarded the honorary permanent age of sixty-four to enable him to stay as the Founding Partner of Arup, as he would normally have had to retire when turning sixty-five. Ove Arup’s personal contribution to mathematical calculations, or engineering detail, was not what secured his reputation in the annals of engineering and architecture, although they helped him on the way. What did secure his place are the people he chose to hire and who he inspired to develop and carry out his ideas. Until 1970 Arup consisted of “Arup Associates” and “Ove Arup & Partners”, but after internal discussion on how the firm should be run, the firm was reformed into just “Arup”. That same year Ove Arup spoke to partners from the practices around the world, in what is later known in Arup as “the key speech”. In the speech Ove Arup talked about his philosophy for the firm: the importance of working noncompetitively with colleagues, of engaging in interesting, useful, and morally responsible work, and of pursuing “total architecture”. The ideas of total design or total architecture sprung from the ideals of the Modern Movement and Walter Gropius’ ideas for his Bauhaus school:

“The term ‘Total Architecture’ implies that all relevant design decisions have been considered together and have been integrated into a whole by a well organised team empowered to

108: (Opposite page) Charnier detail at the intersection between the two bride-segments on the Kingsgate Bridge Image courtesy of Gareth Gardner

201

OVE ARUP PAST

fix priorities. This is an ideal which can never or only very rarely - be fully realised in practice, but which is well worth striving for, for artistic wholeness or excellence depends on it, and for our own sake we need the stimulation produced by excellence. […] And then we come up against the fact that a structure is generally a part of a larger unit, and we are frustrated because to strive for quality in only a part is almost useless if the whole is undistinguished, unless the structure is large enough to make an impact on its own. We are led to seek overall quality, fitness for purpose, as well as satisfying, or significant, forms and economy of construction. To this must be added harmony with the surroundings and the overall plan. We are then led to the ideal of ‘Total Architecture’, in collaboration with other like-minded firms or, better still, on our own. This means expanding our field of activity into adjoining fields: architecture, planning, ground engineering, environmental engineering, computer programming, etc. and the planning and organisation of the work on site.” 41

Ove Arup describes how he sees his firm as a place where the employees should work not because of money but because of the interesting projects. Yes, money is important to make the employees financially secure and happy, but money should not be the driving force.

202

“There are two ways of looking at the work you do to earn a living. One is the way propounded by the late Henry Ford: work is a necessary evil, but modern technology will reduce it to a minimum. Your life is your leisure lived in your ‘free’ time. The other is: to make your work interesting and rewarding. You enjoy both your work and your leisure. We opt uncompromisingly for the second way. There are also two ways of looking at the pursuit of happiness: One is to go straight for the things you fancy without restraints, that is, without considering anybody else besides yourself. The other is to recognise that no man is an island, that our lives are inextricably mixed up with those of our fellow human beings, and that there can be no real happiness in isolation. Which leads to an attitude which would accord to others the rights claimed for oneself, which would accept certain moral or humanitarian restraints. We, again, opt for the second way.” 42

He goes on to describe the quality that the Arup firm should stand for:

“Only a job done well, as well as we can do it – and as well as it can be done – is that. We must therefore strive for quality in what we do, and never be satisfied with the second-rate.” 43

“The key speech” is today a required reading for any new employees in Arup. Ove Arup kept

working in the office in London until his death in 1988 at the age of ninety-two. He was less and less involved in the projects, however his ever-expanding company undertook numerous prestigious projects, which only turned out successful because of the aim for “total architecture”. Among the projects where Arup has been engineering consultant is the Centre Pompidou in Paris (1971-1977), the new headquarters of the Hongkong & Shanghai Banking Corporation in Hong Kong (1978-1985), Lloyds of London (1978-1985), Millennium Bridge in London (2000), the Channel Tunnel Rail Link (1994), the “Birds Nest” Olympic Stadium in Beijing (2008), the “Water Cube” Olympic Water Stadium in Beijing (2008), the CCTV

headquarter in Beijing (2012) and King’s Cross Station (2012).

109

Peter Rice, who worked for Arup on ceveral projects, including the Centre Pompidou in Paris, (previously known as Beaubourg), described Ove Arup:

“He was addicted to his French berets. Every five or six years he would go to France to get a new beret. I remember well the problems he created when he came to visit Beaubourg, looking for the right type of Breton beret. I think that was really why he came.” 44

109: Collage of Arup projects: King’s Cross Station in London, Water Cube, Bird’s Nest and CCTV Headquarters in Beijing. Image courtesies of Arup

203

OVE ARUP PAST

CODA CHAPTER EIGHT Ove Arup did not only found the world famous engineering company bearing his name, he also lay the foundation for its raison d’être; the reason for architectural engineer’s existence. Of course this kind of architectural engineering (in oppose to the engineer who also acts as the architect as described in chapter six) would have existed if it was not for Ove Arup, but he surely was a pioneer within the field. Ove Arup was not the typical engineer, not of the time but neither of today. For a long time he was not sure if engineering was the right thing for him, something that helped him keep a distance to the very serious world of engineering. He was perhaps able to still have fun with what he was doing, meant it that way that he did not lose himself in long mathematic equations. He did not take his himself as an engineer too seriously. As an engineer he possessed a drive for making good design. He was theoretically well founded, something that gave him the overview to think outside the norms of his time. As a philosopher student he had a good background when it came to discussions, and as an engineer he was not afraid of going into discussions with architects, or other engineers. He was not afraid of saying his opinion on a design; he did not limit himself discussing topics only concerning engineering. He possessed the ability to see things in a larger perspective. He had a good overview of his

204

projects and he was confident in his own engineering skill. His background in contractor’s business made him well orientated on what was possible to do on a construction site and what he should be careful with. He learned the hard way that he had to design something that could convince architects, clients and construction workers at the same time. He had an honest passion for architecture, good design and beauty. His main goal in life was to save the world from all its horrors. As he grew older he accepted that he himself could not save all human beings, and he became satisfied with doing as much good as he was able to. He believed that one was obliged to use the skills one had, and for him this meant to build good designs for people. He was known for always looking at how the users of his design would feel. Ove Arup differed from other engineers of his time by his passion for architecture and design. He was well read, not only within the field of construction, and he knew the tendencies and trends of the day. This, perhaps, also helped him not feel diminished by architects; he believed that his opinion was just as important as theirs. And he felt an urge to change the view architects of the time had on themselves as raised above the rest of the construction industry. As the founder of perhaps the most influential engineering office in terms of inventions and progress in the construction industry, he was aware of the importance of good design – total

architecture. His philosophical education gave him a background that supported a company like this. A company with clear values of not doing it for the sake of earning money â&#x20AC;&#x201C; although it was important for Ove Arup to pay his employees well enough for them to live the life they wanted so they could be free to work their best â&#x20AC;&#x201C; always doing the best they could, and not to be afraid of new challenges within fields where they were not experts.

Zunz, Edmund Happold, Peter Rice, Cecil Balmond, and more recently Tristram Carfrae.

Ove Arup was very social, attended lots of meetings and opened up his home to many people. I believe this helped him get many of the jobs he got, especially his membership of MARS gave him contacts with some of the best architects at the time. Yet another good quality he must have possessed, but one hard to prove, was his ability to hire the right people. He was able to select a variety of individuals that fitted almost perfectly together in a team. He hired specialists who he could use for the different tasks; specialist who perhaps knew more on a certain topic than him. He had the overview and as the leader of the company he knew enough of every field to say whether or not something was possible, even though he might not himself was able to perform detailed calculations. As the Arup firm grew to substantial size, Ove Arup withdrew himself from the front and let younger powers take over. The company has given birth to some of the worldâ&#x20AC;&#x2122;s best creative engineers of the second half of the nineteenth century; engineers like Peter Dunican, Jack

205

PETER RICE PAST

Chapter nine

Peter Rice “By using only pragmatism, by using only rational thought, by always demanding and wanting to know whether what we want to do is right, we destroy the very basis upon which the good or noble things in our life exists.” 1

Peter Rice was a very bright and intelligent engineer, who only came to realise many years into his career, that he was actually a designer. In the beginning of his engineering career he could be described as the archetypical engineer: confident in his mathematical and structural analytic skills, but with no insight or interest in visual culture. In his mid-thirties, through the collaboration with the architects Renzo Piano and Richard Rogers, he gained confidence in his own personal taste of design. He was not afraid to bring some of his personality into his projects, making his projects both more interesting for himself but also for everyone else.

206

He was caught by the spell to find and design something good or noble, and through that search he became one of the most talented designers of engineering structures. Peter Rice grew up in a small town in Ireland, close to the Northern Ireland border. He was a bit of an outsider in school; he didn’t do any sports and was more a dreamer. Rice was especially interested in math. From early childhood he found numbers more interesting than letters, and he used to imagine how every number had its own special quality. At the time he grew, Ireland was up a place with not much room for eccentric personalities, and freedom was more a physical thing rather than intellectual.

““Free thinker” had the same malevolent sound as “racist” today. […] Freedom, if it existed, was in mathematics where great ideas could be explored without boundary.” 2

Perhaps he would have become something else than an engineer if he grew up in a different society, one that supported the creativity in a dreamer like Rice. He was put in a box by his teacher and was called “the mathematician”. Rice acknowledged that:

“[…] if you were good at math you could not write. […] You could not expect to be good at everything. The possibility of writing was postponed for many long years.” 3

So Rice went to become an Engineer. His father told him that engineering was the only job within the field of math where you could make money. He started to study aeronautical engineering in Belfast, Northern Ireland. He quit the study because he didn’t like the place where it was taught and changed to civil engineering.

ney Opera House roof structure. In the office he was a part of a large team working on the Opera house, and he did both calculations on the roofs final state and during construction. After working three years on the project, the scheme for the roof changed from a concrete shell concept to precast ribs. Rice must have had an exceptional spatial understanding and he wrote the computer programmes for the geometric definition of the precast concrete elements.

“[…] I found myself one of the few people with an understanding of how to solve and codify the geometry of the shells.” 5

“I didn’t understand the engineering but I could do the sums. Understanding came later when I started to work.” 4

Rice completed his primary degree in Belfast and after another year of studies at Imperial College in London he joined Ove Arup & Partners in 1956.

ARUP AND SYDNEY OPERA HOUSE Rice’s first job in Arup took 10 years of his life. Starting out in the London office, Rice performed the principal analysis of the Syd-

110 110: Peter Rice on site at the Sydney Opera House Image courtesy Arup

207

PETER RICE PAST

From the beginning of the project, the engineers at Arup had looked at the roof as twodimensional forms. They tried to design the structure exactly like the architect Jørn Utzon drew them in his competition sketches. When turning to the rib concept, Arup engineers with Rice on-board started to see the shells as three-dimensional objects. And with the change came the story of every single rib being a part of the same sphere. Utzon recognised this as the way forward. After three years in London, Rice moved to Sydney to work on site as a part of his agreement with Jack Zunz (b. 1923), the leader of

208

the design group at Arup. In Sydney he became responsible for the survey and positioning of the shell and tile elements. He discovered that:

“[…] the work was really rather simple, no more than applied common sense.” 6

Zunz had noticed Rice’s special structural analytic talent when working together in London. Zunz wanted to move Rice on to the construction site to have a guy there with detailed knowledge of the predicted behaviour of the

111: Conceptual drafting of the Sydney Opera House, with shells cut from the same sphere Drawing by Jørn Utzon

111

roof structure.

“Peter was also keen to play his part in its realisation. […] On the site at Bennelong Point, Peter’s contribution proved immense. He used his formidable intellect to help solve seemingly intractable problems in rectifying unpredicted deformations during erection. He wrote computer programmes that enabled the contractor to predict accurate positioning of the structural elements. He also experienced the real problems that can only be learnt on site, actually being a part of the construction process. He learnt to develop the art of the possible, the manner in which one can open up boundaries of accepted practice.” 7

On site in Sydney, Rice followed Utzon around the building site. It was like a long slow apprenticeship in the art of architecture for him.

“Before Sydney I had a very primitive appreciation of architecture. Life in rural Ireland in the 1950s had given few clues to what it was all about, so I came to the experience innocently, like blotting paper ready to absorb any information which came my way. 8 That the scale of the building works at every level, providing interest and articulation from wherever you are, is a key lesson I was to remember from my days in Sydney. […] the awareness of the importance of the integrity of the building’s construction and of the need to

provide interest and articulation in the structure at all levels. […] It is the details which control the reaction of the public and hence their perception of the scale and warmth of the building. ” 9

After three years in Sydney, Rice wrote Zunz asking to have a year away from practice:

“I would like to study the application of pure mathematics to engineering problems. I think that a more thorough understanding of the nature of the equations used to solve structural problems in design could lead to a better conditioned solution and ultimately a better choice of structural components.” 10

Returning to London, Rice worked in the Structures 3 group in Arup. Ted Happold ran the group, which worked with a mixture of cable and fabric structures, mostly with Frei Otto. Otto introduced architect sir Richard Rogers to the group, with whom they did a competition for a new stand for Chelsea football ground, which they didn’t win.

“As we sat thinking we realised that a good reason for entering competitions is not to win them but to explore relationships and design.” 11

209

PETER RICE PAST 210

112: Sydney Opera House during construction in 1967, the precast concrete ribs are at this point still visible Image courtesy of Life Magazine

211

PETER RICE PAST

CENTRE POMPIDOU Rogers soon after formed a partnership with Renzo Piano and together they were doing another competition. The Arup group asked them to enter the open competition for a new museum for modern art in Beaubourg, Paris, later renamed Centre Pompidou. Piano & Rogers agreed, but they where more focused on winning their other competition.

“Doing the competition was fun. It was all done quickly near the end, so there wasn’t any time for the fun to get lost. This is an important point about competitions, especially open ones. The entry must not become to deliberate or too detailed, or too closely argued a response to the brief, because the jury will only have the briefest of time with each entry. It is the idea they will see and the spirit of the drawings. […] We wanted to fire the imagination. After all, we did not expect to win.” 12

The team was declared winners of the competition in 1971. Winning the competition changed Rice’s career completely. From one day to another the team became stars within the building industry and as an associate in Structures 3 he was an accepted part of the design team’s head members. As a part of the competition premises, a mutual office for whole design team was set up in Paris. The French government was not particular happy with the fact that a foreign engineering company was taking jobs from the locals. “We have

212

some great engineers in France you know …!” At the time Rice, and many from the team, didn’t speake French, so all the local critic and scepticism to the radical and drastic structural design were not heard by the design team in the beginning.

“We just smiled at whatever people said and got on with the job of fulfilling our destiny.” 13

Shortly after winning the competition, Rice was in Japan to deliver a paper on tension structures. As a part of the conference, he visited the remains of the 1970 World Exhibition. One of the structures was a space frame in steel, where the nodes where made of cast steel.

“An idea was born. I had been wondering for some time what it was that gave the large engineering structures of the 19th century their special appeal. That is present in many of today’s great structural achievements, but they lack the warmth, the individuality and personality of their 19th-century counterparts. […] Like Gothic cathedrals, they exude craft and individual choice. […] I decided Beaubourg would be designed using cast steel as its core material.” 14

Richard Rogers described Rice’s contribution to the design development of Centre Pompidou:

“Peter transformed the competition entry for Pompidou from a design that was in some ways too mechanistic into one that was humanistic. He had a natural sense of scale and proportion. He was an artist and a fine mathematician. He softened the whole look of the building through the way he reconfigured the structure. There’s a lot of handcraft in the building, which might surprise many people, and that’s one of Peter’s great contributions.” 15

The first problem to be solved for the engineers in the Beaubourg design team was to find a suitable construction for the 44,8 meter clear span the architecture called for. Several boundary conditions had to be taken into account; a heavy library that could be placed everywhere for future flexibility, a movement zone for vertical transport of the public placed outside the front façade towards the plaza and a vertical movement zone for ventilation ducts and other “supply” for the “machine” on the back - the

structure had to enclose all three zones. As the structure penetrated the façade on both the front and the back, it seemed logic to create a natural point within the structure where the façade would go.

After trying several different concepts, some 113 not so favourable in the architect’s eyes, one of the team members came up with the idea of a so-called gerberette solution. Named after Heinrich Gerber, who invented it for bridges in the 19th century, a gerberette is in principle a cantilevered structure that supports another structure, dislocating the two almost completely from each other. The gerberette pieces are placed on the outside of the building, supporting the mega trusses spanning cleanly the 114

113: Plan and elevation of the cast steel gerberettes used in the Centre Pompidou. Image courtesy of Bruce M. Coleman 114: Arial view of Centre Pompidou in the heart of Paris. Image courtesy Richard Rogers

213

PETER RICE PAST 115

214

inside. The gerberettes at the same time supports the movement zones, and are hold back by cables at their edges. The ingenious solution solves several problems at once, making a natural transition between outside and inside. Yet another advantage of the gerberette, is that it transforms the static system of the three zones from a hyperstatic structure – with its calculation theory first solved by Navier – to a static determinable structure. Kazinczy explored the problem of the hyberstatic structural

systems, and his tests on steel beams showed that the full capacity was not reached with Navier’s theory. Gvozdev later showed that the designer of a hyperstatic structure can only be sure that he found the right way the structure carries the load in the case of uniqueness. Uniqueness is given in a static determinable structure, as the gerberette solutions are. Rice was very aware of the fact, that to design the optimal structure one must know as much as possible of how the structure works.

115: Gerberettes on the top floor with Paris in the background Image by author

Not only is the designer completely aware of the maximum stresses the structure will be exposed to, the performance of the structure is also much more predictable. Peter Rice described it as:

“The more you know, the better you can read and the more you can see.” 16

Even in the connections between the gerberette and the column supports Rice designed the joint so he was confident of the forces inside:

“In Beaubourg the joints are articulated to ensure that we can be confident of the way in which the loads are being transferred. The gerberette is assembled with its machined surfaces in contact with other parts of the structure in a way that guarantees how the load is transferred between the main floor trusses and the column. It is very important to ensure that the load on the column is as close to its central axis as possible, such as to avoid any excess bending moment or twisting being transferred into the column. An uncentred load is referred to as an eccentric load. The column diameter and all the details which are dependent on it are designed to eliminate any eccentricity in

116: The façade of the Centre Pompidou is as a structural diagram, it is possible to read the way forces run through the system. Image by author

116

215

PETER RICE PAST

the application of the load. Large spherical bearings guarantee this and ensure that we can predict the path that the load will take between the gerberette and the column.” 17

The gerberettes should of course be made of cast steel. They were shaped almost like a structural diagram, with wall thicknesses taking into account the flaws of cast steel and openings determined by the erection sequence.

“The forces and loads in the piece – I like the word piece, it makes me feel like an artist when I use it – were the principal determinants of its shape: slender at the tension tie end where the load is applied, deep and strong over the column where the load and moment reach a maximum, and slender again at the point of pick up of the beam.” 18

A design philosophy had begun to emerge: nodes in the mega trusses were designed in cast steel; compression members were made of tubes and tension members of rods. To allow more light to pass through the upper, compressed members of the trusses were split into two horizontally parallel members. In that way they don’t visually become so massive, as light can pass through them. The use of cast steel was not without trouble for Rice and the Arup team. When the tendering results arrived, all except two French contractors withdrew their offers, and the two combined theirs to one that was 50% above

216

the budget. The client declared the tender null. None of the contractors were willing to take the risk of the cast steel, as they had no experience with it. The team contacted the Japan steel company who had produced the cast steel for the 1970 World Exhibition construction and a German. Both said it was possible and that they could do it well within budget. Rice knew from his trip to Japan that it indeed was possible to use cast steel, but to convince the local industry was difficult. Industry always knows better:

“Never believe what industry says is possible or impossible. It is all said with some other motive in mind.” 19

The project had through the design phases general opposition in the French engineering establishment and industry. The real client, President Pompidou, always had faith in the team and the design, and without the support from the French client the project had not turned out as it did. One of the reasons why the project made Arup and Peter Rice famous is of course the visual structure. Architecturally, to put a giant machine-like building in the middle of beautiful Paris, with its classic stones facades and grand boulevards, is daring and people may or may not like it. But the project starts to become personal once the viewer gets closer. The gerberettes expresses personality, warmth and individuality, something that is normally provided

by the architectural finish.

“One day shortly after the opening I saw an old lady, dressed in black like the Irish mothers of my youth, sitting perusing the people and looking in wide-eyed astonishment. I watched her for a while, just sitting quietly, stroking the side of the gerberette, she was not afraid, not intimidated and she was on the fourth floor.” 20

Rice criticises how architecture is often from by a single photograph taken from a specific angle. A photograph can only show the building in two dimensions at a specific time of the day. Rice argues how the public, and

many architects, have a prejudiced feeling of a building from published photos.

“For an engineer this is particularly significant. Much of what we do, and the special quality that thoughtful engineering can bring to a project, are not photographable.” 21

Ironically, this is, like so many other things, opposite for the Centre Pompidou. The structure, including the gerberettes, is highly photographable and this might be one other reason that Arup and Peter Rice became known for this project. 117

117: The gerberettes at the manufacture’s plant Image from the exposition Traces of Peter Rice

217

PETER RICE PAST

Jean Prouvé (1901 – 1984) was an engineer who came into Rice’s life during the design phases of the Centre Pompidou. He had been a member of the competition jury, and was among the few in the French engineering establishment who supported the design. To Rice, the support of Prouvé was an important factor in the belief that the cast steel gerberettes could be done. Prouvé was actually hardly a member of the engineering establishment. Raised out of an artistic family, his father being an Art Nouveau painter and his mother a pianist, Prouvé grew up wanting to become an engineer. The outburst of The First World War broke his dream of attending engineering school and the war left the Prouvé family with almost no money. He took an apprentice at a blacksmith, instead of engineering school, in 1916. The apprentice gave him a passionate concern for both how materials were made as well as how they appeared.

118

He began his work in buildings designing and making architectural ironwork for staircases, doors, elevators, lights and furniture. In Nancy in 1923 he opened what would be the first in a string of his own workshops. In 1931 he opened “Ateliers Jean Prouvé” that in the beginning produced furniture in collaboration with architects. During The Second World War his Ateliers produced bicycles and other needs for the war. In 1947 he built the Maxéville factory where he produced furniture and undertook extensive architectural research on the uses of aluminium. Here he also developed a prefabricated aluminium house concept, Maison Tropicale, to be constructed in Africa. The idea of the prefabricated house was that all elements are flat and easy to pack. The aluminium ensured lightweight for easy transportation. Prouvé’s engineering work was characterised by a great attention to detail, manufacturing methods and constructability. On top of that he was a great inventor, always searching for new solutions. Like the automobile and aircraft industries at the same time was exploring the use of new lightweight materials to make safer and lighter cars and aircrafts, Prouvé did the same for the building industry. This is exemplified

218

118: Façade of the CNIT building in La Défense, the financial district in Paris Image by author

in the Maison Tropicale, where he uses the properties explored by other industries to achieve maximum perfor- 119 mance of the thin aluminium sheets. The house was also designed with natural cooling and ventilation, but in the end only three were built. Later he started “Constructions Jean Prouvé”, designing smaller buildings and pavilions. He designed a aluminium pavilion for abbey Pierre (1912 – 2007), who in 1953 had Prouvé design a shelter for homeless people. The 57-m2 two-bedroom houses, called “Maison des Jours Meilleurs” (A house for better days) was so simply designed that it only took two men seven hours to erect. Prouvé’s later turned his attention to façade design: windows and building envelope. He used his knowledge of metalwork to design slender mullions and glazing bars that was easy to manufacture and install on site. Two façade projects exemplify the ingenious creativity of Prouvé: Maison du Peuple du Clichy from 1938 and CNIT in La Défense from 1959. At Clichy the external vertical panels are made of lightweight steel sheets, which, to avoid buckling and distortion, are given a slight curvature by a spring placed inside. At the large façade of the CNIT-building, a concrete shell structure where Nervi was consultant, performance was optimised in several ways; the façade was hung from the top to avoid compression, and thereby buckling, of the steel mullions and braced horizontally by walkways. 22, 23

119: Drawing of Maison Tropicale

Image courtesy of Arch. Dép. Meurthe-et-Moselle

219

220

PETER RICE PAST

MATERIALS

quality of an engineering design.” 25

Peter Rice is also known for his willingness to use new materials. After the completion of Centre Pompidou in 1977, Rice was approached by architect Adrien Fainsilber to work on a project in La Villette in Paris. The project was to design large glass conservatory structures and central reception areas in the museum Cité des Sciences et de l’Industrie. Rice invited Martin Francis, an industrial designer, who invited along architect Ian Ritchie to join the project, and the three started the engineering firm RFR based in Paris. The glass boxes, called serres, was intended to be as translucent as possible, they were a visual transition zone between the museum space of the building and the park that surrounded the museum.

The articulated joints secured a vertical connection of the glass. To brace the panes horizontally against wind loads Rice used prestressed cable trusses. The steel cables were placed horizontally so they wouldn’t obscure the panoramic inside view of the park. The glass was used to restrain the cables, and in an smooth way the two, the glass panes and the steel cables, only functioned because of each other. 121

“Transparency was the quality the design was expressing”. 24

The boxes were eight meters high. The limit for the design team was the maximum size of manufacture, two metres by two metres. Each box consists of four glass panes, each pane suspended from the pane above, with articulated connections done in the corners.

“[…] by choosing a highly articulated connection we could test its behaviour and be able to guarantee its performance in us. […] The freedom to move guaranteed that we had a predictable combination of load and strength of the assembly. Predictability is a very important 120: (Opposite page) Peter Rice stress testing the cable truss in the glass boxes. Image from An Engineer Imagines 121: Glass boxes in the Cité des Sciences et de l’Industrie at Villette, Paris. Image by author

221

PETER RICE PAST

“This was possible because if the glass were not there to restrain the truss, then there would be no need for the mullion since there would be nothing loading it.” 26

The design also incorporated spring supports, to prevent disaster in the event that one or two glass panes would break. The glass panes are at the top of the box hold by springs, so the force will distribute equally through the remaining supports. Rice was in the same time that he was designing the glass façades at Cité des Sciences et de l’Industrie at his RFR office also working at Arup’s on a project in Houston, Texas with Renzo Piano. The Menil Collection Gallery was a new building to house Mrs De Menil’s private art collection, which she had decided to leave to the city of Houston. Together with Piano, De

Menil decided from the beginning of the project that the collection should be seen in natural conditions, and they agreed on that natural light transmitted through the roof would give the most direct contact between the gallery spaces and the outside. For preservation reasons, this is not the way art is normally stored. A high amount of natural light is not good for preservation, but in Menil the plan was that each painting had a cycle where it was only on display for one month out of six or six months out of thirty-six. The rest of the time they would be stored under ideal conditions in a specially constructed storage. The roof was designed as louvers, shaped after certain parameters as lightning and direction towards the sun, structural requirements and architectural shaping. Renzo and Rice decided to use ferro-cement for the louvers, a material invented by Nervi in the 1950’s. The final dimensions of the louvers were in the end

122

222

122: Connection detail and the cable truss sketches by Peter Rice Image from An Engineer Imagines

decided from data from a computer model simulating the light transmittance based on several years’ weather data for Houston.

“The essence of ferro-cement is that it has a very high toughness and uses the minimum amount of materials. It was invented for a situation where materials are expensive and labour costs are low. Low labour costs usually indicate that good quality craft labour will also be available. […] But it is also a material which, because of its fineness, low thickness and plastered finish, is capable of very elegant shapes and surface quality.” 27

It became very important to find a way to manufacture the louvers to a high level of detail in order for the calculations to be correct. Rice searched for companies all over the United States who could manufacture this ferro-cement, but ended up finding a company back in England who could do the manufacturing with spray concrete technique. Rice had to take big part in the manufacturing process to achieve the product he wanted, and for this project he in many ways acted as a modern master builder.

“We made a fibreglass mould of a leaf and then made a back-up mould to act as a surface to spray on to; we developed a way to manufacture the multi-layers of mesh to place into the mould. The key problem was how to guarantee that the mortar matrix was dense.

Obviously, if it came through the steel layers in 123 a dense form and looked solid and sound on the formed surface, that was a good indication. Early checks convinced us, however, that large flat water pockets could develop parallel to the mesh layers as the force of the spraying caused vibration of the mesh. The process was modified and some extra steps were introduced. With these extra steps and with careful monitoring of the process, we were able to guarantee the quality of the final product. It is interesting to think about what this meant. In most of industry the quality of product is guaranteed by testing after manufacture. Here was an element which we could not test after it was made. The guarantee of its integrity lay in the process of making it. So we were relying on the process and monitoring the process to ensure we had the quality we needed.” 28

Rice acting as a modern master builder also happened for the Pavilion of the Future project

123: Sketches section of The Menil Collection from the initial draft design Drawing by Renzo Piano

223

124: The Menil Collection, Houston. On top the ferro-cemento louvers Image courtesy of Renzo Piano Building Workshop

PETER RICE PAST

125

glass from La Villette. Stone and glass have the same brittle behaviour, the same physical characteristic.

“If we could protect the stone from tension forces and from sudden loads then we could perhaps build the screen using stonework as a primary structural material, but in a more sophisticated way.” 29

Rice again had to search for someone who could manufacture the product he wanted. Stone was mainly being used in façades, and found manufacturers that could deliver the stone precise enough. The contractor was in the beginning not very convinced of the feasibility of the project. As the contractor had to take responsibility for the final product, he of course had to be convinced. To do this, all calculations where made available for both contractor and client. Rice even produced a detailed erection method for the stone arch system. The contractor of course thought he knew better, the method proposed was far too complicated. They started construction, but soon saw cracking in one of the large stone pieces. for the 1992 World Exhibition in Seville, Spain. The client, the Expo ’92 Committee, had asked for something spectacular for the three hundred meter long façade of the pavilion. Rice proposed to use stone for the construction, a construction that looked like a façade but with no building behind it. To design the stone structure, Rice used his experience with

226

“After enormous initial scepticism about the feasibility of the design, the contractor generally relaxes and then makes all the mistakes which he feared in the first place.” 30

Luckily Rice had made sure that his team

125: Model from the exhibition Traces of Peter Rice showing the static principle of the stone façade. Image by author 126: (Opposite page) The stone “façade” of the Pabellon del Futuro, Seville. Image courtesy of Asociación Legado Expo Sevilla

227

228

PETER RICE PAST

where present on site, following the product until completion. Rice believes this is a very important task of the engineer, as the contractor should not find himself all alone if something starts to go wrong. If left alone with full responsibility, the contractor may loose faith in the design. In the end all problems that did arise during the construction of the Seville project were caught before the contractor started to lose confidence.

Paris, Rice used a combination of fabric and glass. Rice had a feeling that fabric, being a translucent material, where hard to perceive as a viewer – the form is always hard to read.

“We, the designers, must never forget that we do not build our designs ourselves. It would be much easier if we did. Nervi, the great Italian engineer, or Prouve, were able as contractors to correct any misapprehensions when they encountered them. […] This may not be easy, as the contractual conditions in different countries do not always allow for the presence of the designers throughout the construction process. The key thing is that people never lose confidence that the work can be completed correctly. And being there to help and resolve the problems before doubt can take hold is vital.” 31

He had previously used the seaming joints within the fabric, which makes a small shadow band, to reflect the form. This works when seen from the inside.

Rice was a pioneer within structural glass, and he had the confidence to use materials in new structural ways. He could even find new ways to use common structural materials as steel (cast steel in Centre Pompidou) and concrete (Lloyd’s in London). He continued throughout his career to work with fabrics, which was the key element in his first group in Arup. In the “Cloud” hanging within the Grande Arche,

“For some time I have felt that, as an object in space, fabric has very little physical presence and I’ve been looking for ways in which one can give physical context to a fabric surface.”32

“The joining together of those pieces to create the perfect surface means that the curvature and shape of the fabric itself are difficult to see, particularly from the outside and particularly with PTFE fabric, which is pure white once it has bleached to its equilibrium condition. It’s only really the doubling up of the jointing, the shadows of the seams which you get at the points where the cutting pattern panels meet, that enables you to perceive the surface form. This is a very important consideration. It means great care must be given to the selection of cutting patterns, so that the perception of the overall surface that you ultimately see is consistent with and enhanced by the detailing.” 33

“Spreckelsen’s [architect of the Grande Arche] original fabric intention was one large sheet-

127: (Opposite page) The “cloud” seen from the inside with the Grande Arche seen behind Image by author

229

PETER RICE PAST

like structure inside the space. I felt that while this might provide a certain amount of shelter in the space, it was very unlikely to provide any perception of scale, so I sought to think of the solution in a different way. I was worried initially about what I would call the presenceof-the-fabric solution, because I felt that a single-sheet solution would tend to disappear in certain circumstances to do with light and viewing angle and would not provide any of what might be called physical body with which to work. These things were called Nuages or clouds and in this context I felt that the thing had to have body, depth. It had to occupy a certain physical space.” 34

For the Grande Arche Rice placed a pattern of glass rings inside the fabric. The glass pattern clearly makes the shape of the fabric perceptible, and it also had the other advantages that the scale of the Grande Arche could be seen from underneath the “cloud”.

“I introduced the idea of glass into these sails so that somebody underneath could see through the glass to the arch above; therefore, although enclosed under the translucent fabric canopy, you would also be able to see through it in places to get a sense of the grandeur of the arch itself.” 35

That Rice took on a broader role than the usual engineer is also reflected in the way Richard Rogers described him:

230

“I see him as an artist, a poet, a sculptor engineer or engineer sculptor, a humanist, a Brunelleschi of recent times.” 36

Rice described his own idea of how to explore new materials within the highly regulated engineering world, where all norms and codes must be fulfilled, as:

“In the mid-nineteenth century materials were explored by building and waiting. Some of the structures built in this way, and even some of the great medieval glass windows, the rose window at York Minster for instance, would be difficult to justify today, with all our modern analysis techniques and our need to satisfy modern rules and norms on safety and soundness. Does that mean that we are forever condemned to forms and shapes that are self-evidently safe and sound? This is one of the great challenges for the structural engineer and one which I have always wanted to meet head on. Obviously we must satisfy all these modern rules and regulations, which, if intelligently interpreted, leave a lot of scope for invention and innovation. ‘Intelligently interpreted’ - two sweet little words - dropped into the argument from nowhere. Intelligent interpretation is required not only by the designer but also by the checking authority, those who will give permission for a structure and assembly to be built. How often have I heard disbelief when I have explained the La Villette glass solution to a control engineer. ‘That would not be

possible here ... !’ is the response, before the engineer has examined the argument, and with the visible evidence of a completed structure to go and see. As if gravity, or the thickness of the window, could somehow change as you cross the border from France into Germany, or Switzerland or wherever.” 37

128

128: The form of the “Nuage” or “cloud” inside the Grande Arche is clearly readable Image courtesy of Patricia Marino

231

PETER RICE PAST

THE ROLE OF THE ENGINEER Rice was truly a great engineer with a lot of knowledge. But can one single man function as the master builder in our highly specialised society? Rice him-self claims:

“I learned the minimum necessary. It is still the way.” 38

Jack Zunz, Rice’s team leader for the Sydney Opera house roof design group in Arup, describes Rice as an analyst first of all. In remembrance of assembling the Opera house design team in 1961, Zunz wrote:

232

“Even in hindsight it is difficult to imagine that this very bright, sensitive and thoughtful engineer would become one of the most talented designers of engineering structures of his generation. He did not appear to have any special concern for the design of structures. I cannot remember him attending any meeting with Jørn Utzon or his co-workers. His main interest was in the analysis of structures, and he was exceptionally good at it.” 39

So how did he become “the most talented designer of engineering structures”? According to himself, Rice was very good to use the people around him. He knew who knew something about every specific problem. The Arup com-

pany is full of highly specialised engineers, but Rice was perhaps one of the few who knew how to collect all the information in the right way.

“The Arup back-up had a critical role. […] Sydney had taught me how to use Arup and its gifted individuals. Beaubourg was the proof that it worked. The use of the talent of the many highly skilled Arup engineers has been central to everything I have done and my work would have been much poorer and much less adventurous without it.” 40

With his great ability to analyse structural system, he was able to think beyond normal

structural analysis and go on to start being the designer of them. But to do this, he was impacting architecture. He had the ability to approach the design process differently, it was to Rice not a question of convincing the architect to integrate his idea in their design – it was more a question of how he could contribute to the design. Zunz describe Rice’s role:

“[…] he had now established the confidence to use his exceptional talents to complement rather than to defer to those architects with whom he was working.” 41

Rice himself describes how he sees the different roles of architects and engineers:

129: The Passerelle Simone de Beauvoir was designed by RFR together with Dietmar Feichtinger Architectes. It combines function with structural necessity in a way that creates an incredible light shape. Image courtesy of David Boureau

129

233

PETER RICE PAST

in different ways.” 42 “I would distinguish the difference between the engineer and the architect by saying the architect’s response is primarily creative, whereas the engineer’s is essentially inventive. […] The architect, like the artist, is motivated by personal considerations whereas the engineer is essentially seeking to transform the problem into one where the essential properties of structure, material or some other impersonal element are being expressed. This distinction between creation and invention is the key to understanding the difference between the engineer and the architect, and how they both work on the same project but contribute

234

130: The Passerelle Simone de Beauvoir Image courtesy of David Boureau

This may sound like the archetypical relationship between architect and engineer, the relationship where both work on the project but essentially don’t talk to each other. But to Rice it is important to remember these roles and he didn’t like to be called architect-engineer or architect. As one example he describes his role for the La Villette project:

“As an engineer I worked essentially with the glass. It was the properties of the material which motivated the development of the de-

130

sign. Thus, although we can say that there was originality and aesthetic choice in the way that the design developed, this way forward was directed by the need to express the properties of the glass in full.” 43

New York based designer Frank Stella, who worked with Rice on a bridge in front of the new French National Library, recognised how Rice could bring the most conceptual ideas of an architect into a realised project:

“Once he had a handle, once he could grasp the image [of the architect], Peter just rolled on like a juggernaut, crushing the obstacles of practicality and cost, making it possible for us to built what we like.” 44

Rice called this the chameleon effect. He could be working with several architects at the same time, architects with completely different design philosophies. He was able to understand each of their architectural approaches, and didn’t try to diminish that but rather refine and develop the idea within each philosophy. The architectural intentions introduced by the glass boxes in La Villette was pre-defined to the engineer, and then it was up to the engineer to develop the story of the structure. Rice, perhaps differing from other engineers, had the need to express the properties of the glass in full. He goes on to describe what he sees as the engineer’s problem:

“I believe that it is this lack of a personality to identify with the work which is the fundamental weakness of the engineer’s position. Engineers need identity. Engineers need to be known as individuals responsible for the artefacts they have designed.” 45

But Rice still believed that engineers should not start to become the architect. He saw himself as an engineer first of all. He was still very aware of his lack of artistic creativity:

“[…] I can feel diminished, small, uncultured. As indeed I should. With great work about which I know nothing, so many great ideas, everything so clearly connected, ignorance is a kind of arrogance. But that’s how it is. And I think it is true for most engineers. They get lost in this world of visual culture, but not in music. Many are passionate musicians, not lovers of literature or history, and they fail to communicate in an intimidating world.” 46

Rice had not forgotten how it was to be the archetypical engineer, the one for which structural design was the design of reinforcement bars inside the concrete – the design of the invisible. Rice claims that many engineers today work incognito, the architects take the fame.

“Engineers work incognito. Unlike their predecessors, today’s engineers work behind a screen of other egos.” 47

235

PETER RICE PAST

Rice was so lucky in his career, that he over night, after winning the competition for Centre Pompidou, became famous. Everybody wanted to work with him, and this of course gave him the opportunity to always work on the projects he found interesting. Perhaps that is the reason why he was able to do so many inventive projects.

“I have done as an engineer, the scope for inventiveness and innovation that exists, and also to identify how the engineer’s contribution can enhance the architecture of many artefacts. […] Of course I do not wish to imply that all engineers are somehow unappreciated geniuses waiting for the opportunity to express their inventive skills. This is clearly not the case. Many engineers are themselves affected by the general expectation that society places upon them and they behave accordingly. Indeed they may in their pragmatic way encourage and foster the very atmosphere that inhibits them when they wish to be inventive. This I call the ‘Iago mentality’.” 48

Iago was a character in Shakespeare’s Othello. Through sound and sensible rational arguments, Iago undermines the fragile characteristic of love and loyalty in the Othello’s romantic idyll. Rice uses the Iago character to describe the engineer who destroys the design process, by always saying “no this is not possible, because…”. Iago is the prototype of a scientific man.

236

“In the dialogue of architecture and engineering, the engineer is the voice of rationality and reason. It is a role which is all too easy to play. After all, one is being sensible, reasonable, modern in questioning the more far-fetched flights of fancy of some architectural proposition. This then is the engineer’s principal destructive weakness, to play Iago to the architect, or indeed to another engineer, Othello.” 49

Rice believes that as an engineer actively participating in the design process, it is important to remember not to destroy creativity.

“I think I believe the argument [of Iago]. That’s my dilemma. That I believe this argument - that we [engineers] are in fact Iagos - and have got to escape.” 50

It is also important to remember that other players in the building industry might play the role of Iago, and only the engineer can then use his knowledge to refute the arguments:

“[…] This then is a noble role that the engineer can assume – the role of controlling and taming industry. The building industry has an enormous investment in the status quo and, like Iago, will use every argument to demonstrate that other choices are irrational and not very sensible. Only the engineer can withstand these arguments, demonstrate the wrongness of the position of industry and demolish its ar-

guments.” 51

good personal friend:

Rice’s personality had yet another important characteristic. He was able to talk to other people – architects, clients, and contractors – about something else than what happened in the construction industry. Zunz describes his personality:

“Peter was a delight to work with. He had an innate sense of design, but was never presumptuous. He kept his cool under pressure. He certainly didn’t seek fame or money. He lived modestly, but he lived to the full; he had a great sense of fun and loved wine, women and song! […] Peter was wonderful at connecting people.” 55

“His talents had developed with a strong intellectual spine allied to a warm personality. His relationships with collaborators, particularly some of the world’s great architects, were not only developed on the grounds of his professional talents, but also through his enjoyable personal encounters.” 52

Rice was diagnosed with a brain tumor in 1991 and died the following year aged 57.

“He had a very unconventional approach to working with people, and I think that’s why he was greatly loved as a person to work with he wasn’t conventional, he wasn’t linear, he wasn’t predictable.” 53 “We talked about art, we talked about life, we talked about feelings – and of course we talk about architecture and engineering – but it was always, before we got into the nitty-gritty, trying to understand what it was, what was this idea we were trying to convey with the project.” 54

Rogers is remembered of Rice’s ingenuity everyday, as in his home in London the main steel stair was designed by Rice. To Rogers Rice was not only a good engineer; he became a

237

PETER RICE PAST

CODA CHAPTER NINE As Ove Arup, Peter Rice was not sure whether engineering was what he wanted to do in his life. So, in general we can conclude that the successful creative engineer does not really know if he wants to be an engineer. That is of course not true, but maybe it is not completely false either. As Peter Rice says it himself, the engineering world is very serious. I think it helped both Peter Rice and Ove Arup on their way to success that they did not take themselves too serious. Chapter nine starts with a quote by Rice stating that engineers always want to know whether or not what they do is right. Through their education, engineers how become accustomed to only having one correct answer to the tasks given. Applying a mathematical approach when collaborative on a design can destroy the creativity, not only the engineer’s creativity but also the whole group’s creativity. Rice was aware of this “Iago” role that the engineer could take. Perhaps because he did not take himself too seriously, he was able to look beyond the norms and just do what he believed was possible. Throughout his, unfortunately short, career, Rice seemed to play with materials. He used stone, glass, foil like no one before him had done, mixing it with more common building materials as steel and concrete. One could mistakenly be led to think that he must be a specialist on materials, but as he says it him-

238

self: he was very good in using the specialists in Arup. I think he was very conceptual, and as Ove Arup he knew the necessary on a lot of subjects but was only really specialised in a few. He knew that stone, for instance, should only be exposed to compression for him to be able to control it. One could argue, that if he was a specialist in stone he would have used the stones limited tension capacity. Another example is the first set of cast steel Gerberettes that failed during the load test; if Rice really was the material-specialist he would have known this pitfall and reminded the German manufacturer of the correct temperature they should have used for the casting (apparently the casting process was not directly described in the documents Arup delivered the manufacturer, but noted as a reference to a new British steel standard for offshore structures). If I should guess on what really happened; Rice had a specialist in Arup do the detailed calculations for him and on the fly of handing over the calculations, Rice did not put enough notice on the calculation criteria that the specialist had assumed. Rice was a specialist in geometry. He first made the computer program that calculated the shape of the Sydney Opera House roof structure, and once returning from Sydney he worked at the Structures 3 group in Arup, who mainly worked membrane and cable structures. The geometry of such structures is linked to the equilibrium of the whole system, and a geometrical problem is in fact always a structural problem and vice versa.

Instead of being really specialised in materials, Rice was able to see things in a broader perspective. This made him become a great designer on a conceptual level. To take the example of the stones in Seville; he knew he needed to put the stones under constant compression to be able to control the stability of the system, and he did it by adding elements from his knowledge on cable structures, he could create a system where steel cables always kept the stones under compression. After the Centre Pompidou project, Rice was often involved in the very earliest stages of a design. He was approach by collaborators – artists, architects or clients – to help them, and they worked together on a mutual ground. He was not in a role where he was to judge if a design would stand or not; he was collaborating on the development of the design from the very beginning. What makes an engineer good in doing conceptual design is very hard to determine I think. It is interesting, and also the reason why I initially wanted to do my thesis on this subject. It is difficult - to me impossible - to generalise onwhat the good structural concept contain. It is always different from project to project, and as Wolfgang Prix says:

standable, but they are always old.” 56 Personally I believe it takes a lot of “common engineering sense” mixed with a good understanding of the architect’s design intensions and architecture in general. Peter Rice possessed the ability to bring some of his own personality into the projects he worked on. He was not afraid of exposing his personality, something I believe many nonesuccessful engineers might be afraid of. They hide themselves behind calculations. It was in fact a very intimate thing Rice did, something that must have taken great courage. But by doing so, he also took responsibility for the design. Not only the engineering responsibility, proving that the design is stable, but also for the shape and expression – total design as Ove Arup would have called it. Rice made his projects his own; he had an attachment to them that one normally only see architects have. It was just as much his baby as it was the architects, and he was just as proud of it as them. From his time on site in Sydney, he knew the importance of designing something that could be built. A competence already been established as a key to succeeding is the ability to not be afraid of doing something beyond the normal regulations. Rice had the confidence to do this, but to succeed in doing it in a more modern world; he needed to bring the actual builder on board as well. He did not only design the final shape of a structure, he also designed construction sequences and erection

239

PETER RICE PAST 240

methods. In the modern building industry, it is normal for a contractor to take the full responsibility for the final product. If the contractor thinks he is taken a huge risk, it will push up the price. If he is not confident on how to build it, it will also push up the price. It is not the easiest task to convince contractors that your design is constructible, but it is of much importance to do so. To make a good design that the construction company also believes in, does not only lower the price it also eliminates the chances of mistakes during erection.

241

BOLLINGER + GROHMANN PAST

Chapter ten

Bollinger + Grohmann “Whilst rarely pronouncing that an idea is impossible (or stupid, or inappropriate), they gently coerce you away from the inept towards the appropriate: but then to something more: towards the desirably appropriate. Then to something more still: the unexpectedly and desirably appropriate.” 1

Klaus Bollinger (b. 1952) and Manfred Grohmann (b. 1953) graduated from the same class at Technische Hochschule Darmstadt in 1979. Bollinger went on to work for two engineering companies before he started as an assistant professor to Stefan Polónyi (b. 1930) at University of Dortmund in 1981. Polónyi was a structural engineer born in Hungary but worked most of his life in Germany, where he started the “Dortmunder Modells” – an education combining architecture and engineering. Grohmann was after graduation employed at Wayss + Freytag for two years before he be-

242

came an independent. In 1983 they started their engineering company, Bollinger + Grohmann, at Bollinger’s kitchen table in Darmstadt. Their university education was, described by themselves, straightforward. They had acquired the necessary techniques to be structural engineers, but something was still missing to Bollinger:

“After graduating at TH Darmstadt I had the uneasy feeling to lack essential knowledge.” 2

Many graduates, both back then and now, are very able to perform detailed calculations, but have rarely learned how to draw and design. Bollinger used his role as assistant professor to learn from Polónyi. As a teacher at the faculty of architecture at Dortmund University, Polónyi became a role model for both Bollinger and Grohmann in the beginning of their career.

131

Bollinger learned about the creative aspects of engineering. This creativity was already recognisable in the first competition Bollinger + Grohmann did, a competition for the new ball sports hall in Frankfurt-Höchst. The entry drew the attention of Peter Cook (b. 1936), an English architect who at the time was teaching at the Städelschule Frankfurt. Though not winning the competition, Bollinger + Grohmann came in contact with Peter Cook who was looking for a structural engineering assistant for his architecture class. From the beginning Bollinger + Grohmann had competition consultancy as one of their core competences. Klaus Bollinger became the assistant lecturer to Cook, and in this function as structural engineering consultant for the postgraduate architecture students he would meet many of the young talented architects who

later became their first local clients. This was in fact the case with schneider+ schumacher. Both Till Schneider and Michael Schumacher had met Klaus Bollinger already in 1986, and Bollinger + Grohmann became the responsible structural engineers for schneider+schumacher’s first projects. Schumacher appreciates the unique relationship they have with Bollinger and Grohmann, by some described as the relationship of an old married couple. He describes the creative process as a “dancing process”: searching many options, investigating different branches of an idea for in the end to come back to the initial vague or fragment idea, which through the dancing process has become much clearer. Schumacher says he would leave every meeting with Bollinger + Grohmann with the positive feeling “to have made some progress together.” 3 schneider+schumacher’s architectural language is rather structural; not afraid of

131: First competition sports hall, Frankfurt/Main-Höchst, 1983 Image courtesy of Bremmer Lorenz Frielingshaus

243

BOLLINGER + GROHMANN PAST 132

244

exposing structural elements as a part of the visual understanding of their design. During the design process schneider+schumacher have showed willingness to turn their design in a new direction because of the structure. The engineers at Bollinger + Grohmann have in this way been privileged to actively influence the design, making it partly their own, and, moreover, they are entitled to criticise the design of the architects. Bollinger + Grohmann were the responsible engineers of schneider+schumacherâ&#x20AC;&#x2122;s first projects, and though both companies have worked with other partners it is always the wish of Michael Schumacher to discuss any conceptual idea to Bollinger or Grohmann

first. schneider+schumacher have been awarded several prices for their designs, one of them being the temporary Info Box Berlin construction. The red Info box stood as the first fix point in the giant construction site of Potsdamer Platz opening to the public in 1995. Just one year before, in 1994, the competition had been made to design a building that could at once function as the information centre of the, at that time, largest construction site in Europe and further more reflect the worldwide interest in the building activities in recently re-united Berlin. When demolished by the turn of 2002

it had had more than 10 million visitors. The hovering structure had become a symbol for the German transition to become united. 4

In the beginning of the 90s engineers from Bollinger + Grohmann would take their laptop to the architects office, and together they

“For architect Christoph Mäckler it was everytime an enrichment to work with engineers Klaus Bollinger and Manfred Grohmann. Both of them possess an idea of structural design that goes far beyond the basic calculation of the structure; both of them possess a creativity and sensitivity than many engineers – according to Mäckler – lack, but which is crucial to boost architectural ideas.” 5 It is perhaps not a temporary pavilion that will secure a company financially. Bollinger + Grohmann did execute more “normal” buildings, but have also never been afraid of doing these projects that do not have thousands of square meters. It is projects in all ranges that defines Bollinger + Grohmann, from the design of a single shop window, to the new headquarter of the European Central Bank.

COMPUTATIONAL DESIGN A key element in the success has been the use of computer analysis to examine a structure’s performance. As architects in the past decades slowly have adapted the use of computers to their advantage, Bollinger + Grohmann have since the 1990s used Finite Element Modelling programs to be able to perform detailed analysis of the architects models. Computational design gives both engineers and architects new methods to develop a design, and the amount of options to be examined can with computers be greatly enlarged. A wide term is parametric design; applied to architecture and structural engineering the key parameter is most often geometry.

could see the impact of design changes in real time. With Austrian architects Coop Himmeb(l) au, Bollinger + Grohmann have developed a digital workflow that have welded the two offices together. In 1994 Coop Himmelb(l)au had won several competitions in Germany and was looking for a German engineering firm that could help them realise the projects. Before 1994 they hadn’t really constructed anything, and Bollinger + Grohmann therefore played a big part in their first completed projects.

“Teams of Bollinger + Grohmann are often working in the architects’ office for weeks; in 2003 the engineers opened a branch of their practice in Vienna.” 6

Computational design is a big part of the Bollinger + Grohmann office in Vienna. In 1994 Klaus Bollinger became professor at The Institute for Structural Design at the University of Applied Arts Vienna. Together with the faculty Bollinger + Grohmann Vienna is developing the parametric finite element modelling program called karamba.

132: Infobox Berlin designed by scheider+schumacher in collaboration with Bollinger + Grohmann Image courtesy of Bollinger + Grohmann

245

BOLLINGER + GROHMANN PAST 246

“Albert Speer & Partner – Architects, Planners, are developing their buildings as a combination of engineering and architecture; the architectural design is one factor among others. Engineers Bollinger + Grohmann plan their structures with a view on functionality, but they are nevertheless sincerely inclined towards beauty and elegance. Both practices like to experiment with new materials and structures. To put it simply one might say: AS&P are not (merely) thinking like architects, but also like engineers; Bollinger + Grohmann are not (merely) thinking like engineers, but also like architects. They all see themselves as planners. Arguably, this forms the bottom line of this special ongoing cooperation much appreciated by both: an excellent foundation for further successful projects!” 7

133: Covering structure for the ZDF-Fernsehgarten, Mainz. Designed together with AS&P Image courtesies of Bollinger + Grohmann

“Bollinger + Grohmann are one of the few engineering practices that have employed architecture graduates as 3D designers.” 8

The development of karamba is done by a team of engineers in Bolinger + Grohmann with backgrounds within both architecture and computer coding. The software produced is directly integrated into the Rhinoceros3D / Grasshopper program, and the aim is to make the interface so easy to manoeuvre that architects, with only little structural insight, should be able to perform structural analysis of their early designs.

“karamba provides accurate analysis of spatial

trusses and frames, and is easy to use for nonexperts, tailored to the needs of architects, specifically, in the early design phase.” 9

With the ability to analyse structure in a parametric setup also comes the opportunity to investigate so-called parametric solutions. With the increasing possibility to 3D print, for example a connection, it is no longer necessary that the elements cross each other orthogonally or in 45 degrees angles. With karamba it is possible to search for the optimised structure, to put the elements in the place where they have the biggest effect. For the design of the BMW bubble Bollinger + Grohmann used computer-aided engineering in a different way than just for structural optimi-

247

BOLLINGER + GROHMANN PAST

Computational design in architectural offices is often related with the software Rhinoceros3D. The program is specialised in free form geometric shapes, making it perfect for architecture. Besides architecture, it is highly popular within the fields of industrial design, product design, jewellery design, and automotive design. Rhinoceros3D uses a mathematical approach for generating and representing curves and surfaces, a technique called NURBS (Non-Uniform Rational Basis Spline). The mathematical approach makes the program much more precise and it also makes it possible to generate the free-form “blob” architecture which has become the symbol of computational design within architecture. Ironically computational design seems to move away from the computer’s binary language of 0 or 1, black or white analogy, towards an imitation of nature. Built into Rhinoceros3D is a scripting language, making it possible to perform operations from an algorithm. Many have exploited this tool, and a big community within the architectural world has been built around this. If you as an architect don’t want to study programming before taking on computational design, tools have been developed where one can design an algorithm in a visual interface. The tool doing this with Rhinoceros3D is called Grasshopper. It has two mayor advantages compared to normal “manual” drafting on a computer: it allows for none-specialised people to exploit the NURBS machine in Rhinoceros3D to develop parametric geometries and – if everything is “scripted” correctly – no drawing mistakes can be made. Algorithmic design has many advantages; once the basics of the design have been created in the program one can easily generate new forms. Architects can explore new designs almost as quickly as they can think of them. To not take the stand of “Iago”, engineers need tools that can take these many design options and easily and fast convert it to something that can be calculated.

134

248

134: Visual illustration of how karamba can be used to rapidly check multiple options of column placement in order to find the most optimised solution. Images by Bollinger + Grohmann

An example of the possibilities karamba brings design process can be seen in the Infobox Vienna Central Station, a competition entry done together with Vienna architect Michael Wallraff. The Infobox Vienna had many similarities with the Berlin project: the function was the same – it was to be used as an information centre for the construction of a new underground station – and the idea of a hovering box was also the same. The ground conditions were quite difficult for the engineering team, as they had to take into account underground sewers, a parking lot and a bus station. The box was hovering 22 m above the ground and the steel columns supporting it could only be founded a few places on the site. As for the structure of the Berlin project, no walls were continuous to the ground, providing no bracing of the building. Inclined columns must be used. “Against the backdrop of these boundary conditions, a generative process was used to develop a complex three-dimensional structure that would be not only structurally effective, but also a fundamental element of the architectural design.” 10

135: Rendering of the Infobox Vienna, designed with Michael Wallraff Image courtesy of Bollinger + Grohmann

135

249

BOLLINGER + GROHMANN PAST

sation. Yes, the bubble should be acceptably safe, but it should also be constructible, within budget and satisfy different architectural and other demands. As the first project with architect Bernhard Franken, it was a relatively simple one. The form of the exhibition pavilion was made of two intersecting spheroids, symbolising two water drops merging. The initial structural idea was that the outer surface could be structural.

“This willingness to engage in fruitful compromise, here in effect trading thickness for flexibility, we could term “engineering without ideology”.” 11

136

250

The transparent double-curved surface was made of acrylic (also known as Plexiglass)

and was fabricated by producing polytheranian molds, milled by computer-numericallycontrolled (CNC) machinery, that was used to slump heated sheets of acrylic into their shape. The supporting structure was made as a framework of curved rib elements in aluminium. The shapes were produced directly from the engineers’ data extraction from the calculation program. In the Kunsthaus Graz project with Peter Cook even more parameters came into play. The design, in the competition called “The friendly alien”, is a giant bubble measuring 60 x 40 m in ground plan housing the local art gallery in the Austrian city of Graz. Bollinger + Grohmann had gathered experience with implementing plexiglass from the BMW project, but the larger scale called for a structural optimisation of the shape.

136: The exhibition pavilion “BMW Bubble” designed with Berhard Franken Image courtesy Bollinger + Grohmann

“Because of the close coordination of engineers and architects the shape could be optimised in terms of increased shell structure behaviour. The basic shape to start working with was a model of a sphere; the shape was then generated with the software Rhino [Rhinoceros3D] by dragging parametric gravitational points. Step by step this virtual sculpture was brought closer to the desired shape and as a result manufacture criteria such as lengths of bars, distances and structural edge conditions could be improved.

The main areas of research during the design 137 process were as follows: The restriction of the numbers of columns and finding a minimized access solution from below in order to not disturb the floating impression of a cloud; the development of material, shape and structure of the “skin” of the big bubble in order to create a translucent freeform volume, which at the same time provides protection from the elements and meets the requirements of a museum in terms of air-conditioning, lightning design and conservation.” 12

137: Picture during erection of the pavillion Image courtesy of Bollinger + Grohmann

251

138: Kunsthaus Graz designed by Peter Cook Image courtesy Bollinger + Grohmann

BOLLINGER + GROHMANN PAST

139

THE CHARACTER OF THE ENGINEER To Klaus Bollinger and Manfred Grohmann it is not important to have a “trade mark” in their structural design, as many architects are known to have in their architecture. Peter Rice described this as the chameleon effect, being able to see and understand different architects vision.

“We are sensitive partners not striving to advocate a particular architectural philosophy. We adopt the approach of the architect. We do not violate this approach, but refine it by means of structural design. We do not object to the architectural philosophy of any approach, but we do object to a lack of quality.” 13

Today Bollinger + Grohmann have offices in Frankfurt, Vienna, Paris, Oslo, Melbourne, and Berlin employing around 130 people. Even though the office started out with focus on consulting architects in competition designs, focus is to bring the projects to completion. For instance, Bollinger + Grohmann is currently involved in one of the biggest construction sites in Europe, the new European Central Bank (EZB) office towers. The project as been designed together with Coop Himmelb(l)au from competition to construction.

“We are involved in the conceptual design and see it through to the end.” 14

254

139: Structural rendering of the EZB. Image courtesy Bollinger + Grohmann 140: (Opposite page) Rendering of how the EZB will look when completed in 2014. Image courtesy of Coop Himmelb(l)au

255

BOLLINGER + GROHMANN PAST 141

256

141: Picture of the EZB during construction Image courtesy Bollinger + Grohmann

257

BOLLINGER + GROHMANN PAST

CODA CHAPTER TEN As already mentioned I have purposely left out interesting engineers as Ney, Calatrava, Happold, Schlaich as I feel they all fall into categories that in some way is covered by others. Bollinger + Grohmann is perhaps not the most forthright company to focus on in this final chapter of my historical review, but it is – with its forefront use of computational design methods – a smooth transition into present day. Bollinger + Grohmann have only recently – within the last five to ten years – achieved worldwide recognition. They did not win the Centre Pompidou competition and became famous from one day to the other; their success (if measured in fame or recognition) was growing slowly during the last thirty years. As collaborators Klaus Bollinger and Manfred Grohmann know how to reach the good design. But it is not enough that only the engineers have the intension of an integrated design process; no, the architects must feel the same way for it to be successful. The student-teacher relationship that Klaus Bollinger had with Schneider and Schumacher must have put them on a more mutual ground, and I believed this was the reason for the success on their first projects together. But it can otherwise be really difficult for the engineer to achieve this relationship with mutual understanding of each other’s challenges. When an architect approaches the engineer for collaboration on a project, it is today still up to

258

the architect whether he just wants to have a consultant or if he is willing to enter an integrated design process. If the engineer pushes the architect towards this collaborate process, the architect might just say no and find another engineer who is willing to be num. Architects have their reason for not throwing themselves headlong into this – you would not trust you baby to any stranger. It requires the architect to trust the engineer, not only on a level of calculation but rather on a personal and aesthetic level. In many competitions the time is limited, and the architects have enough to do with just finding their own intentions – understandable it can be frustrating to enter an integrated design process if this is not done within a safe and trustworthy environment between architects and engineers. This might be the reason why Bollinger + Grohmann only slowly have reached fame and worldwide recognition. With the advances within computational design, Bollinger + Grohmann are putting themselves in front of new design methods. Grashhopper, the geometric optimisation tool of tomorrow, is with the Karamba software developed in Bollinger + Grohmann reaching a new dimension where only a few other companies can follow. I believe only Arup has developed something similar, but oppose to Bollinger + Grohmann they keep their software in-house. Karamba can be used by anyone, and the structural optimisation tool is therefore not only limited to engineers of Bollinger + Grohmann.

The new design method often ends up in none-regular shapes. Strange angles between elements call for a greater collaboration with manufacturers. The importance of following a project from concept to completion is perhaps more important than ever and something Bollinger + Grohmann strive for.

259

260

261

PART TWO

PRESENT

Case 1: Mega-scale

KALININGRAD STADIUM PRESENT

Chapter eleven

Kaliningrad Stadium

Like bridges, stadiums are engineering problems before they are anything else. As seen with the bridges of Thomas Telford and the arenas and stadiums of Pier Luigi Nervi, the beauty of large-scale structures is often linked with an optimised and economic structure. The structure becomes the dominant part of the architecture and there is less need for architectural “dressing up”. It is what David P. Billington describes as “Structural Art”.

diums have been designed by architects, as newly built stadiums tend to play an urban role - the projects become icons. Today the focus is just as much on the visual exterior impression as it is on the layout of the stands, exemplified with the “Bird’s Nest”, the Beijing National Stadium designed for the 2008 Summer Olympics by Swizz architects Herzog & de Meuron and with Arup as engineer.

It is my belief, that beauty of large-scale structures made the majority of today’s structural engineers choose the engineering education. All engineering students should at least dream of building towers, bridges and stadiums, otherwise something is wrong.

“The greatest buildings in history have always re-

It seems that the architect-engineer relationship is notably different when it comes to largescale structures. For a long time architects did not design stadiums or bridges – they were purely engineering tasks. Only recently sta-

266

flected zeitgeist. And right now, the zeitgeist is sport. Sport is the global currency.” 1

The Summer Olympics and the FIFA World Cup are the two biggest sport events, both taking place every fourth year. The glory and prestige of becoming the host nation of one of these events are making competing countries spend a huge amount of money on iconic stadiums and related infrastructure to promote their bit.

142: (Previous page) Render of the view of the new Kaliningrad Stadium from the river Image courtesy of Wilmotte

“A stadium, more than any other building type in history, has the ability to shape a town or a city. A stadium is able to put a community on the map, establishing an identity and providing a focal point in the landscape. Stadia are the most “viewed” buildings in history and have the power to change people’s life: they represent a nation’s pride and aspirations. They can be massively expensive to build, but they can also generate huge amount of money. […] Consequently the stadium will become the most important building any community can own, and if it is used wisely, it will be the most useful urban planning tool a city can possess. In the last 150 years, since sport was codified

and professionalized, there has been a dramatic shift to urbanization, from the country to the city, and the meteoric rise in the popularity of sport has been the consequence…” 2

KALININGRAD, RUSSIA On December 2nd 2010 the twenty-two members of FIFA Executive Committee selected Russia to be the host nation of the 2018 FIFA World Cup. The Russian bid proposed thirteen host cities with sixteen stadiums. Three of the sixteen stadiums would be renovated existing stadiums and the remaining thirteen stadiums would be newly constructed. An internal competition in Russia was executed in order to select the host cities. Initially

143: Site plan of the urban development area, placed on an island close to the center of Kaliningrad Image courtesy of Wilmotte

267

KALININGRAD STADIUM PRESENT 268

thirteen cities applied, a number that needed to be reduced. Kaliningrad was, with less than half a million inhabitants, among the smallest cities in the competition, but it has the advantage of its geographical location which is closest to the West European countries, with their traditionally faithful and – compared to Russian standards – rich supporters. The current stadium facilities in Kaliningrad did not match the regulations for hosting an international event like the World Cup and a new, modern stadium was therefore needed. To encourage selection as host city, Kaliningrad held an international architectural competition for the design of a

new stadium. Paris based architects Wilmotte & Associés was selected as one of two finalists. The initial design was carried out by Wilmotte alone, without any engineers. Apart from the stadium their proprosal also included urban design of an island located close to the city centre. The footprint of the stadium building is approximately 50.000 m². Wilmotte’s design contains: • 49.000 m² roof structure, 9.000 m² being retractable, • temporary upper stand seating 20.000

144: Aerial view of the stadium and the urban development area Image courtesy of Wilmotte

COMPETITION

spectators, • 23.500 seats lower stand • approximately 60.000 m² facilities, VIPboxes, locker rooms, toilets, bars, restaurants, parking, etc. In Wilmotte’s brief they proposed a temporary upper stand that would be dismantled after the World Cup, and then the roof should be lowered so the stadium would not feel too big. The competition jury asked for the plausibility of the roof lowering, and Wilmotte contacted Bollinger-Grohmann for assistance. We therefore came into the project relatively late in the competition phase, in which the shape and layout of the roof structure was no longer up for discussion.

The engineering challenge in the competition was therefore to prove the feasibility of the architects’ design. After some of the latest big sport events, like the 2000 Summer Olympics in Sydney, stadiums had been left unused afterwards. The big peak in spectators during the events caused the capacity of the stadiums to be over dimensioned for the afterwards use. The same thing would happen in Kaliningrad, a city with approximately 430.000 inhabitants - and with no football team in the best national league - would never be able to fill the 45.000 seats. Wilmotte’s suggestion to create a temporary upper stand seemed to be the best solution. Another factor for the design was the climate.

World Cup configuration

Post World Cup configuration

145: Static principles of the World Cup configuration (left) and Post World Cup (Right)

269

146: 3D rendering of the future urban area around the Kaliningrad Stadium Image courtesy of Wilmotte

KALININGRAD STADIUM PRESENT

An open, outdoor arena would be suitable for the World Cup taking place during the summer, but the best solution for Kaliningrad in general would be a covered ground. The idea was therefore to lower the roof after the Wolrd Cup and make the roof retractable in order to create a completely covered space for sports or other events taking place in winter. The renderings Wilmotte showed in their competition entry had many “mega-columns” and showed partly a structure within rectangular boxes. At the stadiums short side, behind the goals, the roof boxes were placed at a higher level than at the long side. This was done to create dynamic in the roof and highlight the linearity and the straight lines of the boxes. Our boundary condition was therefore to find a way to keep all structures within these boxes. First the structure was analysed and a strategy of how to lower the roof was developed. A set of four primary mega-trusses was defined that would span along the interior periphery of the roof boxes. Secondary trusses would then span from columns in the façade to the primary trusses. In “World Cup configuration” the mega-trusses would be supported by temporary mega-columns in the façade to not block the view from the temporary upper stand. In “Post World Cup configuration” new support would be constructed further inwards, as the upper stand then would be dismantled. This would lower the length of the longest span from 223 meters to 125 meters and thus give spare capacity to carry the loads of a retractable roof.

272

147: Initial sketches of the mega-trusses and roof lowering principle

Two primary mega-trusses would span the long side of the stadium, a span of 221 m in the layout of the competition design, and the other two spanning the shorter side behind the goals, a span of 189 m. The aim for us was to keep the height of the mega-trusses the same, despite the different span lenghts, in order to make a symmetric impression. To achieve this, I came up with a solution where the longer spanning trusses would only be loaded by the secondary trusses in the middle part. The mega-trusses spanning the shorter side would then support secondary trusses over their whole length. The governing internal force in the mega-trusses, the bending mo-

ment, was almost the same despite the different span lengths. This was all working very well for the World Cup configuration, but in the Post World Cup configuration we now had the extra load of the retractable roof. Going from World Cup configuration to Post World Cup configuration the loads from a retractable roof was added. We sought to keep the retractable roof as ligth as possible, as it is a movable structure. This implied that it should span the shortest possible which meant spanning between the two longest spanning mega-trusses. The Post World Cup configuration lowered the internal forces

148: 3D rendering of the World Cup configuration. Several mega-columns shown and partly visible structure within the roof boxes. Image courtesy of Wilmotte

273

149: 3D rendering of the external view of the Kaliningrad Stadium Image courtesy of Wilmotte

KALININGRAD STADIUM PRESENT

World Cup configuration

Post World Cup configuration

276

(Option A)

(Option B)

150: Hand calculations of the World Cup configuration (top) and the two Post World Cup configuration options (bottom). A solution in between the two post world cup options was believed to be plausible

in the shortest spanning mega-trusses by more than 50 %, but due to the extra weight of the retractable roof, the internal forces in the longest spanning went up by 35 %. In the Post World Cup configuration too much load went in to the longest spanning trusses and the shortest spanning trusses was only used 50 % of their initial capacity. A solution could be to let the longest spanning trusses support by the shorter spanning

trusses, thus lowering the span length of the longest spanning trusses even further and using the spare capacity of the shorter spanning trusses. The problem with this scheme was that too much load were put on the shorter spanning trusses, but a solution in between the two would be eventually possible. To us it was still a competition, and we were developing concepts. Until this point all my calculations had been made by hand, using principle loads which actually means that I only com-

151: Top left: Post World Cup configuration; Top right: World Cup configuration; Bottom left: Post World Cup configuration with roof closed; Bottom rigth: Post World Cup configuration. Image courtesy of Wilmotte

277

KALININGRAD STADIUM PRESENT 278

152: 3D rendering of the roof structure for the competition stage, showing the primary mega-trusses and and secondary trusses.

279

KALININGRAD STADIUM PRESENT

pared the impact of different span lengths. The height of the mega-trusses was initially roughly estimated as the span divided by 18. Lowering of the roof was proposed to be done by hoisting mechanisms placed inside U-shaped temporary columns. The megatrusses could be assembled on the ground and when all four were completed they could be hoisted into position for the World Cup configuration. Afterwards the stands and secondary trusses could be erected. To go to the Post World Cup configuration, new columns where to be erected after dismantling the upper stand and the roof would the be hoisted down on the new columns. In the end the temporary columns could be taken down. Our design ended up winning the competition, and a couple of months later Kaliningrad was also selected to become host city.

280

SCHEMATIC DESIGN Approximately one year after the initial competition was announced, and almost six months after Kaliningrad had been selected as host city, we started the Schematic design phase. The design team was expanded with an energy engineer and a scene consultant. The architects had in the mean time changed the boundary conditions of the roof design: from being four separate boxes they now preferred to have one box – with the retractable roof then placed on top. Most important for the architects regarding the visual exterior impression were two, five meter high “fat lines” running all around the façade: the upper line being the roof and the lower being the VIP-box level. The change from a structure placed within four separated boxes to now one continuous rec-

153: Wilmotte changed the shape of the roof from the competition to the schematic design phase. The roof now consisted of one continuous box, with the retractable roof parts as cut out from the middle and placed on top. Render by Wilmotte

tangular box gave us some advantages. We could use the continuity to make each side push against each other and create a sort of a shell effect – a continuous element that could support itself. The unusual approach was already executed by Bollinger-Grohmann for the PGE-Arena in Gdansk, used for the 2012 UEFA European Football Championship. I changed the structure of the roof from a system of simple spanning trusses to be a three dimensional structural grid; a space frame. I proposed the idea to the architects, and it was well received. The initial competition design called for eight mega-columns placed on the edge of the upper stand. Being able to suppress these mega-columns and instead place thin, slender columns distributed evenly along the back of the upper stand helped us achieve a more harmoniously perceptible structure. I wanted to search for an optimised 3D-grid for the space frame structure, but in order to change the structural concept we had proposed in the competition phase I felt we had to have a convincing argument. We knew that everything we delivered would be cross checked by a Russian team of engineers, so I set out to research both options in order to compare them thoroughly. To determine live loads, snow loads, seismic loads and wind loads the Russian SNIP norms had to be used. Luckily a colleague from the Frankfurt office had an old copy of the norm from 1985, which could serve as a good guideline. In the scope of work we had described in

our contract that we would design everything according to the Russian norms, but we were allowed to use the Eurocodes as a supplement in these early stages. In the contract we asked the client to hire a local engineer who could support us on such matters. It is the intention that the local engineer will take over more work later in the process for in the end to be responsible for the execution phase, but the client did not hire a local engineer to support us yet. The roof substructure was the first thing I investigated. On the top of the roof space frame the outer skin was done with trapezoidal steel sheets, with a water barrier on top creating a smooth surface, and some areas with translucent ETFE. This was chosen in collaboration with the architects, but it was difficult to get them to chose something as they did not know how much light they wanted to pass through the upper skin. On one side they wanted to create a box that could be completely blacked-out for concerts in the Post Wolrd Cup configuration, but on the other hand they were mostly interested in designing something beautiful for the temporary World Cup configuration. The trapezoidal steel plates span three meters between a substructure. The substructure then span between the primary structure, with the span width differing in each options. The bottom membrane was to be made of a translucent, light weight material like Velum or another textile product. The architects really like the idea of a three dimensional structure and they wanted to show it through the membrane. It should still feel as it was inside a roof “box”, but the outline could been seen

281

142

KALININGRAD STADIUM PRESENT

from below and might even be lit for an evening game. As the bottom membrane was not exposed to snow load, it could span longer than the top membrane and the substructure could be less dense. In between the primary structure, a technical walkway with a minimum height of 5 m had to be placed, from which any maintenance could be done. The size and amount of substructure needed varied with the different options developed as a part of the optimisation process and I therefore made a comparison table of different sizes. The weight of the substructure should be taken into account when comparing the different options, as the need for substructure in for example a 4x4 m grid space truss is much less than for the 20 m span between the secondary trusses in the mega-truss option. I created a calculation sheet in Excel where I could easily change the span and I programmed it to dimension the profile needed. With the table I could compare the weight of the structure for different spans, which showed an almost linear

relationship between span and weight. Each option also had to function with the retractable roof, so the next step was to predimension its structure. The geometry of the retractable part could be seen as being cut out of the main roof and then put on top of it. It was designed to have the same height of 5 m as the main roof and of course to be able to cover the hole in the main roof â&#x20AC;&#x201C; with some spare margin. The two retractable roof parts measured 60 x 75 m each. The retractable roof was designed to follow the concept of the chosen option, whether this is single spanning trusses for the mega-truss option or a space truss for the other option. The loads from the membrane were applied to the space truss at the intersection nodes and the model was supported along two sides, where the supporting tracks are placed. The tracks and the movable mechanism were a difficult part to design, as I did not have any experience within this field. I found an old stadium project that Bollinger-Grohmann had made, but was never

60 Top

Wheight [kN/mÂ˛]

Bottom

40 30 20 10 0 0

15 Span [m]

284

154: (Previous page) Rendering of the view from within a VIP box. Courtesy of Wilmotte 155: (Above) Comparison of the substructure weight for different span widths

executed, in where a specialised company had made a description of a moveable mechanism. I adapted their design to fit with ours, but no calculations were made for it. We agreed to bring a specialised company into the project in the next phase to approve the design. For the initial competition design with megatrusses I tested two options; one as previously described with four mega-trusses and one with only the two longest spanning ones. In the competition concept the secondary trusses

were five meters high and kept within the box, but the mega-trusses were twice as high, ten meters, and therefore did not fit within the box. This was acceptable to the architects as they were placed at the border of the box. It would not make much difference if the mega-trusses was ten or fifteen meters high, I thought, so for the alternative option with only two megatrusses the height was raised to fifteen meters. This made the mega-trusses more efficient, and it turned out that the overall steel weight

156: Top left: Moveable roof structure in context with the rest of the stadium. Top right: Moveable system adapted from other projects. Bottom: Structure of one retractable part

285

KALININGRAD STADIUM PRESENT

was lower for the option with two fifteen meter high mega-trusses. The competition option also had the disadvantage that it would be difficult to construct the intersection between two mega-trusses where the shorter one supported the longer one. But none of the mega-truss options where our, or the architects, favourite: we all preferred the space frame. As described previously, the space frame

286

would be supported by an oval ring of columns placed at the back of the temporary upper stand and columns in conjunction with the façade. At the corners of the stadium the oval ring was furthest away from the façade and thus creating a lever arm which could be used to hold up the roof. At the middle of the four sides the oval ring almost “touch” the façade. At the same time the shell effect of the roof implied that each of the four sides of the roof

157: The two mega-truss options; the initial competition concept with four mega-trusses on top and below the alternative option with only the two longest spanning mega-trusses

158: Top: Overall geometry of the stadium roof and supports. Bottom: Faรงade view showing the two five meter high lines Bottom render by Wilmotte

287

159: 3D rendering of the columns placed at the back of the upper stand, here seen in a corner of the stadium Image courtesy of Wilmotte

KALININGRAD STADIUM PRESENT

Option 1

box in some way supported each other. It was hard to predetermine a 3D-grid that would be the optimal, as the structure both had to have a good strength for bending – both to have a good overall stiffness and to use the lever arm between the corner columns – but it also should be able to transfer normal forces that would appear due to the shell effect. Another parameter for optimisation was the height of the roof. The architects visual demand was the five meter high border line towards both the exterior border and the interior border towards the hole. In between these two boundaries the height of the 3D-grid could be increased to have more stiffness. Therefore I could raise the top part of the 3D-grid above the five meter, as no one would be able to see this from outside. The only constraint was the retractable roof that needed to slide over the roof parts place behind the goals. For the lower part of the 3D-grid the architects did not have a problem with giving shape a “belly”. I tested five different options and compared with the initial flat shape. Only the top grid was raised for the first three options and for the last two options the bottom grid was lowered as well. The setup only used round tube profiles, and all elements were dimensioned in each option in order to compare the diameter. The task was to find a diameter that could work for all profiles disregarding the forces, and then have different thicknesses. As the diagonals have an increased length with the higher variants, it become more difficult to decrease the size of the tubes, even though a higher lever arm between the top and bottom grid in theory

290

Option 2

Option 3

Option 4

Option 5

160: The five different options for the height optimisation study

should be more efficient and thereby require less area per cross section, as a longer profile will buckle under less load. The evaluation of the results showed that the impact on the weight was higher in the beginning, for options 1 to 3 with a lever arm from 5 m to 8 m, but then the curve slowly flattens more and more. I evaluated the stiffness in form of the deformation. An optimum for the stiffness was found for option 3 with a lever arm of eleven meters, but this might be due to the shift from only raising the top grid to also lower the bottom grid. Together with a colleague, I developed and tested more than twenty different layouts for the 3D-grid itself. We took five basic geometries and tested them with different grid sizes. In order to calculate the different variants rapidly, the same profiles were used everywhere and only the deflection of each type were calculated and compared. In that way more or less 100% 95%

Percentage of initial flat geometry

90% 85% 80% 75% 70% 65%

Weight

60%

Deformation snow

55%

Deformation self-weight included

50% 5

Lever arm / maximum height [m]

161: Top: Schematic plan indicating the areas where height optimisation was possible. Bottom: Results of the height optimisation study in graph form showing the relationship between height lever arm and weight and deformation

291

KALININGRAD STADIUM PRESENT

B1 Rectangular Grid basic geometry

relation upper lower grid grid sizes

292

eB1 m

shifted non-directional

B2 Diamond Grid eB2 m

shifted non-directional

B3 Triangular Grid eB3 m

non-shifted non-directional

B1.1 e= 8.25 m

B2.1e= 6.5 m

B3.1 e= 8.5 m

B1.2 e= 10.4 m

B2.2 e= 10.4 m

B3.2 e= 10.4 m

B1.3 e= 12.5 m

B2.3 e= 12.5 m

B3.3 e= 12.5 m

B1.4 e= 14 m

B2.4 e= 14 m

B3.4 e= 14 m

163: Overview of the different types of grid that was tested

shifte direct

B3.2

B3 Triangular Grid

B4 Hexagonal Grid

eB3 m

non-shifted non-directional

eB4m

shifted x directional

B3.1 e= 8.5 m

B3.2 e= 10.4 m

B3.2a e= 10.4m

shifted y directional

shifted directional

B3.1a e= 8.5 m

B4.1 e= 4 m

B3.2b e= 10.4 m

B4.2 e=5.2 m

B3.3 e= 12.5 m

B4.3 e= 6.5 m

B3.4 e= 14 m

B4.4 e= 8 m

shifted directional

B4.3a e= 6.5 m

293

KALININGRAD STADIUM PRESENT 294

B1 Rectangle

B2 Diamond

B3 Triangle

basic grid

shift x y

non shift

shift x y

connections

164: Principles of the five different grid compositions used

B4 Hexagon

elements were under-dimensioned, but if every single member had to be dimension in each test, we would never have finished. Therefore one initial option was dimensioned and the weight from this test were used as a guideline on the other types: the mass of the structure was always 200 kg/m² no matter the type, so for a large grid big profiles were used and small profiles for a denser grid. Again the deformation was the parameter to compare, as it gave me an idea of which grid was the stiffest. The size of the grid had to be evaluated together with the substructure, as the extra weight of a longer spanning substructure should be added to the final total weight. The different grid types were based on four different polygon shapes: 350%

rectangule, diamond, triangule and hexagon. This basic grid was then offset by the defined height of 5m either straight down or down and shifted sideways. By connecting the top and the bottom grid a Polyhedra – a structurally efficient space packing – was made. Comparing the overall test results of all the tested grids we saw how both the diamond option and the shifted triangular options with respectively element lengths of twelve meters and ten meters performed efficiently in terms of deflection and they also had the advantage of a low number of connection nodes – having many nodes would increase the cost. When further analysing the two best options, the diamond grid option had nodes where eight members meet, where the

OVERVIEW GRID DEFORMATION RECTANGULAR GRID DIAMOND GRID

300%

TRIANGULAR GRID REGULAR TRIANGULAR GRID SHIFTED

250%

HEXAGONAL GRID

200%

150%

100% 5

Element length [m] 165: Graph showing the deformation for each investigated grid compared as percentage against the best performing

295

KALININGRAD STADIUM PRESENT

shifted triangular grid had nine members meeting. We therefore suggested choosing the diamond grid. Thereafter a layout of the 3D-grid combining the height optimisation with the diamond grid was dimensioned and I checked how the structure would work under temperature variations as well as lateral stability. The steel tonnage was calculated to 160 kg/m² including the substructure. Comparing this with 230 kg/ m² for the best mega-truss option, the space frame option saved 30 % of the steel. In parallel to the optimisation studies I had carried out statical concepts and calculations for the stands; the temporary upper stand should be done in steel elements so all the connections could be bolted and ease the dismantling process and the lower stand should be done in prefabricated concrete as this was thought to be the cheapest structural solution. Wilmotte did not care too much about 3D-grid of the space frame; they left that study completely to us. For the height study, they asked for it to not be too noticeable if a helicopter would film from above but besides from that I was given free hands. For the inside visual perception of the roof they actually preferred a little “belly” underneath. What they really liked from us was the opportunity to show the client that we as a team were able to find optimised solutions that could save money without making architectural compromises. To find even more savings than 30 % on the steel, we also proposed not to lower the roof in

296

the Post World Cup configuration. The teamwork with the architects went really smooth, they listened to all our ideas and used our arguments of efficency as their own. I think everyone in the team had the feeling to have given something to the design. The design belonged to all of us, not only the architects or the engineers individually. After we delivered our package for the schematic design phase, the architects went to Kaliningrad to present our work. The Russian engineering team had only calculated the initial mega-truss option, but as they also found the steel tonnage around 230 kg/m² they believed our further development with the space frame and accepted the change to this concept. The project then went political, as some friends of the local governor new some other architects who tried to push out Wilmotte from the project. To make a long story short, we are not sure if we will continue to work on next phases of the project despite the credit we received for our work so far.

166: Top: Rendering of the chosen structure. Bottom: Render of the cores and cables connecting the cores with the roof for horizontal stability

297

167: 3D render of the final design of the (empty) stadium Image courtesy of Wilmotte

300

301

Case 2: Large-scale

302

KALININGRAD STADIUM PRESENT

303

ANTWERP PROVINCIEHUIS PRESENT

Chapter twelve

Antwerp Provinciehuis

In the spring of 2011 the province government of Antwerp launched an open architectural competition to design a new office- and multipurpose building for the province. Together with Xaveer De Geyter Architects (XDGA) BollingerGrohmann participated in the competition. In December 2011 our design of a remarkable twisted tower was announced winner of the competition. I started in Bollinger-Grohmann in the beginning of 2012 and the Antwerpen Provinciehuis was my first project, though I did not participate in the development of the competition design. At present time the province already disposes over a tower and a series of smaller buildings located at the site of the competition. The province government found the existing buildings, constructed in 1972, to be untimely and not modern enough. In the winning design almost all the existing buildings were proposed demolished, to make room for a new park. Only

304

the lower part of one existing building, called “Front Building” as it faces the street, was to be kept and the new tower was constructed over this building. XDGA wanted to keep this existing building’s two lower stories, as today it is used for formal receptions and as a space to welcome important guests. The new twisted tower will have 15 floors above ground and will be 57 meters high, thirteen meters lower than the existing tower. On ground plan the tower forms a cross with the existing building, but as the tower rises towards the top it is twisted 17° around the northwest corner. The twisting creates an angled façade and the tower leans towards the south. On sunny days this inclination creates shade on some parts of the south façade. The building’s shape is optimised towards more northern light and less direct sunlight. The characteristic triangular windows help to bring natural light deeper into the building than regular rectangular windows

168: (Previous page) Rendered view from the street of the new Antwerp Provinciehuis Image courtesy of XDGA

Existing volumne

of the same size would have done. The Bollinger + Grohmann engineering team consisted of members from both the Paris and the Frankfurt office. We, that is the Paris office, had done the competition with XDGA and were in daily contact, but the Frankfurt office had the knowledge of designing large towers, for instance the new European Central Bank (EZB) in Frankfurt. In the beginning the main engineering task for us was to make the tower stable – ensure it did not turn over because of the twisted geometry – without compromising the architecture. Later on we started focusing more on the existing building, with architecture on a bit smaller scale than the tower.

Obstacle in the park

Element in the park

Moving deputies

TOWER The structural concept of the tower was designed with a load bearing façade and two cores on each side of the existing building. The load bearing façade was chosen to keep the internal space as free as possible, but also because the twisted form created a rather large cantilever, which would not be easy to carry without a load bearing façade. The façade, two concrete cores and a row of columns between the two cores supported each floor slab. A two-story high truss, spanning between the two cores above the existing building, supported the columns. We proposed to make the façade in concrete, as the architects wanted to expose the material in the façade and also preferred to have the ceiling as exposed concrete. This implied that

Sculpture

New Province House

the floor build-up was designed as a reversion of a normal one, so instead of a dropped ceiling to hide the ventilation ducts and electric wiring, we used a raised floor with all installations accessible from above.

169: Architectural concept explained through eight schematic drawings Image courtesy of XDGA

305

ANTWERP PROVINCIEHUIS PRESENT 306

170: Exploded view of the structure, cores, columns and truss in red Structural render by Bollinger + Grohmann

One of the primary arguments for the twisted design was the favourable effects on the indoor climate. The province wanted to act as an example for everyone, and its new headquarter should be as energy-efficient as possible. The aim was to get a BREEM-certificate for a passive house. The design team had proposed natural ventilation in the competition design, which required the air to pass through the concrete floor slabs to be either cooled in summer or heated in winter. As the flow of air in a natural ventilation system is slow, large ducts were needed and therefore we needed 40 cm thick slab. We investigated several structural solutions for the slabs, as a 40 cm thick full “fat” slab in concrete would be rather heavy. As the tower in general is not symmetric, a heavier self-weight would have a big increase on the overall turning moment. The governing load for the overturning moment was in fact not wind load but eccentric self-weight. A solution to reduce the weight was to pour air into the concrete in the form of balls or bubbles – a solution known from the prefabrication industry. We also discussed using a reversed rib solution, with concrete beams placed on top of the slab – this solution could also be done with steel beams, as the steel would not be visible from below. Finally two prefabricated solutions were investigated, the simple solution with prefabricated concrete elements and a newer solution called “slim-line” where steel beams are poured in to a thin concrete slab. The architects did not want to see any lines

from joints between prefabricated elements, so these options were dismissed early on in the process. After more detailed calculations the rib solution was also dismissed, as it would require 70 cm high rib beams. A combination of a full fat slab with pre-stressed reinforcement and the “bubble deck” solution was chosen in the end. As the span length of the slabs were not the same on each floor, as the floors spanned between the twisting façade and the strait cores, some areas with longer spans required extra stiffness in form of pre-stressing and for shorter spanning areas we could pour in bubbles. We had to combine our need for pre-stressing with the environmental engineer’s (heating, ventilation, and air conditioning - HVAC) need for ducts evenly distributed over the floor plan. This caused problems, especially near the cores, which were sometimes hard to solve. Besides the cores and the façade, the slabs were supported by a row of columns placed on top of a two-story high mega-truss. The truss spanned between the two concrete cores. The truss itself was designed rather quickly in the schematic design phase where we basically only dimensioned the steel elements in the truss. In the next stage we had a more thorough look at both the joints in the truss and the connection with the cores. The job was done by our colleagues in Frankfurt who had experience with connections of this size from the new European Central Bank tower - here two separate towers are bound together by a vertical truss through the otherwise none-structural atrium between the two towers to secure the

307

ANTWERP PROVINCIEHUIS PRESENT 308

171: Rendering of an open floor office Image courtesy of XDGA

309

ANTWERP PROVINCIEHUIS PRESENT

lateral stability. The Frankfurt office also helped to evaluate the façade structure. The engineering problem was to determine to which degree the structure functioned as one connected structure or weather the different structural elements – the façade structure and the core walls – acted independently. When modelling this in a regular calculation program, one can easily make great mistakes by assuming that everything is connected rigidly. The stiffness of the concrete floor slabs are determining how good the different vertical elements are connected and this can be difficult to evaluate. A specialised calculation program, called “ETABS” was used to evaluate this, taking in to account time dependent factors as shrinkage or creep in the pre-stressed concrete bubble deck. At the same time the program can simulate the erection sequence to find the biggest stresses the

structure will undergo during its lifetime. The special situation for the twisted tower was that one core was touching the south façade at the bottom levels, but as the core was vertically strait and the façade not, the core was in conjunction with the north façade at the top levels. It was clear to us that the core would function together with the north façade at the top and the south façade at the bottom, so for instance the cores walls were in compression and the façade elements in tension when the building was exposed to a turning moment from the wind. It was difficult to analyse how the structure would work in the twisted part beforehand, where the core stands somewhere in between the two façades, so here the program was of great use. A topic for many discussions between the architects and us was the geometry of the façade structure in combination with the windows. The vertical parts of the façade acted as fixed boundary conditions for us, as the grid of triangular window-panels shifted half a grid for every floor, so that half a window was placed at the corners on every second floor. The façade grid was 3,6 meters, and the top of the window 3,0 meters wide leaving 60 cm for the structure at the top of the floor where the structure was connected with the slab edge-beam. Each floor was always 3,55 meters high, except the lower entrance level which was higher. The bottom floors with a straight façade were 26,6 meters wide and the top strait part 19,4 meters wide, both being 72,0 meters long. The double curved surfaces in the middle twisted part of the façade made the tangential

310

172: Rendered image of the principle floor buildup Image courtesy of XDGA

height of the façade longer than on the straight parts. As the architects wanted the windows to go all the way from the raised floor to the exposed concrete ceiling, each window on the twisted part would have different geometries. Also the length of a whole floor decreased due to the twisting, so the grid of 3,6 meters did no longer fit, and the normal 60 cm structural space between two windows decreased on these floors. Structurally we needed to keep the structure as continuous as possible, and we also tried to make as many elements as possible similar to not create a big extra cost. The exterior impression of the façade was of course very important to the architects,

and also to us. I thought that a lot of differently sized windows would become a big extra expense, so I started to model the façade in order to be ahead on this subject. The doubly curved surfaces in the middle were actually ruled surfaces, meaning that they could be constructed with straight lines – as in a hyperbolic parabola. I started proposing options where the triangular windows had one vertical side so our structure could fit with the straight lines of the hyperbolic parabola, but this was not an acceptable solution for the architects. Then I went on to propose an option where the structural lines where not vertically straight, but still continuous in between the windows. This meant that windows on the twisted part were

173: 3D-drawing of the final façade layout, showing how the grid is develop from the straight part of the façade and highlighting the differing windows. Image courtesy of XDGA

311

ANTWERP PROVINCIEHUIS PRESENT

all different which we tried to avoid, but this was visually the most appealing solution for the architects. I proposed an alternative version in which all the windows had the same size, but this created a strange visual effect on the most inclined parts of the façade when seen from the inside, where a part of the façade needed to be filled. As previously explained, the architects wanted the windows to extend from the exposed concrete ceiling to the raised floor, but in the twisted parts we would then have either a beam dropped from the ceiling or the windows would not stand on top of the raised floor. The difference in length also made the line of the forces none-continuous in a way that could be acceptable, but the connections would need more reinforcement and therefore cost more in both labour and material. I tried to combine the two options in a parametric option; I had developed an algorythm where it was possible to choose how many different types of windows we wanted to use and then the script found the most suitable solution. It created a sort of facetted interior visual impression, but the effect was not too big once the amount of different windows went above ten. Due to the risk of fire spreading from one floor to the one above, a minimum of one meter’s clearance between the top of a window to the bottom of the windows on the floor above was needed. The windows had to be less high to make enough room for this demand, so the architects choose to place the now smaller windows so the top of a window would be in line

312

174: Principles of the different solutions investigated. Bottom picture show the scripted solution with 10 different sized windows

with the exposed concrete ceiling, and then have an up-stand beam at the bottom. This made the option with same-sized windows favourable again, as the up-stand beam would be anywhere now. In the end, the architects chose this option combined with groups of different windows to solve the problems at the façade’s ends. Structurally we investigated three options for the façade; in-situ concrete structure, pre-fabricated concrete structure and a steel option. The prefabricated options was preferred as it was the cheapest, but the irregularly façade

had almost no repetitions which might bring the cost up significantly. The steel option was working well, but it was expensive and the elements had to be protected against fire. The architects also preferred to show the structure, which could not be done if the steel had to be covered. We chose to continue to investigate the insitu solution. We could still image two different scenarios: a full cast surface with triangular holes for the windows or only poured elements so that the façade would become a big truss. The last option would save both material and

175: Combination of the two options. Red lines show the line of the forces or the centre of the concrete elements.

313

ANTWERP PROVINCIEHUIS PRESENT 314

weight. Connections could be done in two ways, rigid or with charniers; rigid connections would mean a greater stiffness of the structure, but the joints would require more reinforcement. The difference between deflections of the charnier connection compared with a rigid connection were calculated to be around three percent.

In areas with concentrated loads, steel beams were integrated into the concrete elements. Different joint types were developed according to the loads acting on it. Four different reinforced concrete nodes were designed, as well as one steel beam node. Another faรงade issue also called for great coordination effort within the team. The air intake for the natural ventilation should be placed in the faรงade. Structurally we had placed an edge-beam here with the same height as the total floor package, including the raised floor, to distribute the loads from the floor slabs to the supporting faรงade structure. At the same time the HVAC engineer needed to place big tubes through the faรงade, and combined with the triangular windows this needed coordination from the whole team. The architects did

176: Handsketches showing one principle joint with the chosen option with in-situ concrete elements. On top the principle for integrated steel beams

Joint designed for pre-fabricated concrete elements

Joint designed for full concrete panels

not wish to have a big visible hole in the faรงade every 3,6 meters for the air-intake, so the air had to seep through smaller openings around the window panels. The hole in the structural part of the faรงade also had to be coordinated with the fire regulations which secured that a fire could not spread from one floor to another. This was especially critical on the south faรงade where the inclination is outwards, making it easier for a fire to spread upwards. Unfortunately the natural ventilation concept was dropped due to cost savings. The concept

was made by a German company, Transsolar, who cooperated with a local Belgian engineer for the detailed planning and to comply with the Belgian norms for the BREED-certificate. The local engineer did not have much experience with natural ventilation system for an office building, and they proposed a hybrid solution where a mechanical system would take over at very hot or very cold days. The extra cost for this hybrid solution was too much for the client, so in the end a complete mechanical ventilation system was chosen.

177: Sketches of the how the joints could be done for the two other options investigated: prefabricated concrete elements and full in-situ surface

315

ANTWERP PROVINCIEHUIS PRESENT 316

178: Rendering of the new Congress building Image courtesy of XDGA

317

ANTWERP PROVINCIEHUIS PRESENT

CONGRESS The existing building was previously used for offices on the two top floors, and with space for receptions and meetings for VIP guests on the double high ground floor. An archive and a kitchen were placed in the basement, which also was a double height level. The idea was to transform the existing building into a congress, with open spaces for receptions, art exhibitions and an auditorium for lectures. Further the building should house the province council room, called “Raadzaal”. The two upper office floors were to be demolished, as well as the existing façade. The ground floor should be kept more or the less intact, the open space was only obstructed by a few monumental columns. Structurally we had much bigger problems in the basement, where the auditorium and the council room were to be placed. A one meter high crawling space made with concrete ribs was located underneath the basement floor, which I thought was created either to withstand the groundwater pressure or to distribute the heavy loads from the archive on the foundation plate – the building was directly founded on the soil. In the competition design the architects had described how the sloping structures of the auditorium and the council room were to go down into this crawling space. The council room was placed to fit within two continuous monumental columns, but for the auditorium we needed to demolish several smaller columns in the basement to create enough open space. The difficult engineering problem for re-arrang-

318

ing the structure of both the crawling space and the auditorium was mainly to evaluate the capacity of the existing concrete structures. No calculations existed in the archives, only formwork drawings were at our disposal. For the slab above the basement, where we needed to cut the supporting columns, I was afraid that changing the static system of the slabs, by adding new supports and suppressing others, would eventually lead to cracks in the concrete. The existing floor was not done in the same way regularly through the whole floor; in some areas the slab was thickened and spanned four to five meters, while in other areas a grid of beams carried a thinner slab. To our knowledge, all areas were poured together as an in-situ construction. I proposed to make a steel truss, that only supported the slab at the points where the existing columns where suppressed. In that way the concrete slab above the basement kept its boundary conditions and should be able to function just as well as it had been doing for more than forty years. The different structural build-ups made it very difficult to evaluate the effect new supporting conditions could have on the structure. In the crawling space I had to evaluate forces from the up drift from the groundwater when the concrete ribs were demolished, to make room for the auditorium and council room. After further studies, it turned out that the existing one meter rib floor was probably done in a combination to distribute the heavy loads on the soil and to resist the up drift pressure from

the groundwater. With the archive gone in the new scheme, I had to secure the resistance of the bottom concrete plate when exposed to the up drift forces, after some of the ribs were demolished. We had difficulty to retrieve the necessary information on the water table, as the geotechnical survey had been started too late, so in the end I used the information on the water table from the measurements taken during the construction of the existing buildings in 1972. In the up drift situation, the static system had to be turned upside-down, the foundation plate no longer supported the columns but rather spanned between them, as their weight

was holding the building back from floating.. I proposed a set of steel beams to be bolted to the foundation plate to act as a composite structure, but the amount of beams and size should only be determined once we had the correct information on the water table. This was done for the schematic and conceptual design phases. The architects wanted to use the new roof level as a terrace. They had promised to the client that it would withstand 6,2 kN/mÂ˛ or more than 600 kg per square meter. The existing floor was not able to take up this load, and it also had to be insulated and have a new surface made of heavy concrete

179: Rendering of the proposed steel truss solution that would only support the floor at the places of the suppressed columns. Rendering by Bollinger + Grohmann

319

ANTWERP PROVINCIEHUIS PRESENT 320

tiles. They were not willing to make a compromise for how much load we should calculate the floor for, so the only solution for us was to demolish the slab above the ground floor as well. And once we choose to do this, we might as well save ourselves the trouble of redoing the basement level, so in the end all that was left from the existing building was the concrete basement “basin”- outer walls beneath the ground and the crawling space. Morally I felt somehow obliged to re-use at least some of the existing building, as I thought the main concept of the competition design was a tower than spanned above an existing building. The architects did not feel this obligation to keep the existing building, as a new building, with the same outer dimensions, would give them a much bigger flexibility for the interior design.

180: Renderings of the two structural options explored Image courtesy of XDGA

So while the design of the tower started to come to an end, we restarted designing the conference building. After some months we decided to completely demolish the whole existing building, as the architects’ new design brought bigger punctual loads on the foundation which I could not ensure it would be able to resist. In the new design, the conference building was – function wise – similar to the old scheme; the upper floor was arranged as a big open space and two meeting rooms in one end, and the lower floor housed the auditorium and the council room. The existing ground level was lowered so the lower level would have smaller windows along the edge of the ceiling. The façade was completely glazed and not structural. In the beginning we explored two possible structural schemes. The architects had first proposed a monumental structure with few, but massive, concrete columns, a scheme they had already shown the client. The other option we came up with was a grid of thin columns, spaced five meters apart. The five times five meter grid was defined by what I thought would be the cheapest solution. The auditorium structure was now being turned around, so instead of going into the ground it was going up to the lowered ground level – this provided an easy escape route in case of fire. The council room still moved down into the foundation level. In these areas I still could not place columns, so I proposed two solutions for each particular situation. In the auditorium I wanted to hang the floor by a one story high truss placed on the upper level, which could

be fitted within the separation walls between the open space and the meeting rooms. For the council room, as it was going down, I could place a bigger structure underneath the floor package. I would let the bits of the above columns continue down into the space and the hang them up with cables. In that way, one would still have an idea of the structure above and in the same time we would create some sort of an artistic structural ceiling in the council room. Both the architects and us thought that this was a quite good scheme, incorporat-

ing the structural needs in a useful way into the architecture. Unfortunately the client had already decided on the monumental scheme, so the option was skipped. The architects’ idea of the structural concept for the first “monumental” scheme was rather similar to Le Corbusier’s Dom-ino House principle, with a free standing façade and columns offset from the façade. I proposed to have a composite structure with steel beams connected with studs to a concrete slab on trap-

321

ANTWERP PROVINCIEHUIS PRESENT 322

181: Renderings of how the different column sizes from the two options would be interprerated Image courtesy of XDGA

323

ANTWERP PROVINCIEHUIS PRESENT

ezoidal steel plates. The composite structure had a strong resistance for positive bending moments, meaning that the concrete was being compressed and the steel beam was mainly in tension. The positive bending moment would take its biggest value mid between the columns, but as the columns were offset from the faรงade a cantilevered part created a negative bending moment above the columns. The composite structure did not have the same efficiency for negative bending moments, as the concrete would crack and could not be taken into account for the calculations, so basically only the steel beam was left. I created

a detailed calculation program in Excel, taking into account all load cases, the weight of the faรงade and live load located in different positions. The Excel calculation was made with the offset distance of the columns as a variable parameter, so when I ran the program I could see the optimal offset distance for the columns, taking into account the different resistance in the composite cross section towards positive and negative bending moments. Some of the architects at XDGA liked this, because it resulted in a thinner structural package and was therefore cost saving also. Other architects from the team were more worried

Pd Mnegative

Mnegative

pd,cantilever

pd,mid-part

Mpos itive L

2700 2200

Moment [kNm]

1700 Negative moment 1200

Positive moment Symmetric cross-section

700

None-symmetric cross-section

200 -300 -800

324

Cantilever length [m]

182: Diagram showing the result of the optimisation program created in Excel - negative moment is above the column and positive moment is in the middle of the span. From the Excel program by author

about the impact on the auditorium, where some seats had to be cut. In the end it was decided not to use the optimal position, but it did move the columns more inwards than initially proposed as the architects recognised the positive effect of this. The concept of the composite beams was repeated on both floors. The structural system was designed so that columns were placed on every second structural grid line, and a transfer beam was needed between the columns. To avoid a massive steel connection at the columns, the transfer beam was designed as simply supported by the columns. This rose the bending moment in the middle of the beam, and as I did not want to have a composite structure in two directions â&#x20AC;&#x201C; even though it would have been theoretically possible â&#x20AC;&#x201C; the beam had to be so high that its top flange was placed in the concrete. In the centre of the building a monumental stair was placed, connecting the terrace with the open space floor beneath. This stair was cutting a composite beam, normally supported by the transfer beam, so the load had to be transferred to the two nearest beams. As the cross section was already calculated quite close to its limit, there was not enough spare capacity left to take up this load. Initially I thought that the hole the stair created would reduce the load on the beams equally, so that the beams were not loaded more than the rest of the beams. As I had promised a certain height of the structural package in the previous phases, a solution to stay within this height had to be found.

183: Pictures of models of the final layout of the congress building Image courtesies of XDGA

325

ANTWERP PROVINCIEHUIS PRESENT

The beams with the extra load were placed in the middle of the building, where the building was connected to the tower. I could create a hyperstatic structure by holding the otherwise cantilevered parts down at the connection to the tower. This would not affect the architecture of a freestanding façade, as the support could be hidden within the connection to the tower. The hyperstatic structure still had too big internal forces when calculated with our FEM calculation program. Supporting the beams at the cantilevered ends attracted more forces towards the column supports. So instead of having a positive bending moment in the middle of the span that was too big, I now had a too big negative bending moment above the column supports. In a hyperstatic structure, the distribution of internal forces is normally – and for instance in most FEM computer programs – solved by the superposition principles developed by Navier almost two hundred years ago. The theory is a lower bound solution, and to solve the system, e.g. to find the internal forces in the beams, one must know the statical boundary conditions, assuming that the internal forces are always in equilibrium and know the specifics of the material used. The theory is correct as long as all boundary conditions are true, but for the example of our beam, the outermost fibres in a section above a column would have started to yield long before the final load was reached and the beam would no longer be straight.

326

I further developed the idea of a hyperstatic structure by using a bit of an unusual way for Belgium or France, a way normally not accepted there; to distribute the internal forces by the theory of plasticity – and upper bound solution. To solve the system of an upper bound solution, the failure mechanism must be known. The geometry of the failed structure is difficult to derive, so normally the breaking mechanism is guessed. If the guess is not the correct one, the internal forces will be underestimated and therefore it is considered an unsafe theorem. Plasticity theory is in Belgium and France accepted as an approach to calculate the resistance of cross sections, a local phenomenon. It is normally not used as a way to distribute internal forces in a global system. By using the plastic theory I was able to prove the sufficiency of the beams. To model it in the FEM program, I put springs as supports at the intersection with the tower. By manipulating the spring constant of these supports, I came to a result where the negative bending moment above the columns was equal to the one in the middle. I could prove this would work, but I was not sure if this method would be good enough for the Belgian control office. I knew it would work, but I would like a better way, an accepted way, if one could be found. After a few weeks thinking I found the solution: the “Gerberette”. Developed by Heinrich Gottfried Gerber in 1866 and used by Eduardo Torroja, Peter Rice and others throughout the history, the Gerberette converts the hyperstatic structure into two static determinable struc-

184: Sketch of the development of the calculation method used for the beams exposed to extra loading from the monumental stair. Sketch by author

327

ANTWERP PROVINCIEHUIS PRESENT

tures. If I introduced hinges, or charniers, in the right place, I could shape the curve of the bending moment as I wanted. By analysing the curvature of the bending moment found when calculating the beams according to the plastic theory, I could see the exact points where the bending moment went from being positive to negative and place the hinges at these points.

These hinges would never be visible for visitors, but to us it brought great satisfaction. I was also responsible for the design of the monumental stair, that caused all these structural “acrobatics”. The stair was to be placed in a glass display, which also functioned as the thermal barrier – the stair was placed outside even though it went down from the terrace. Being called the “monumental stair” and placed as on display, we needed to design something visually apperaring very nice. The architects had designed the overall geometry; they wanted a spiral stair turning one-and-ahalf time and the diameter of the spiral was also fixed. Initially I tried with two trusses on the edges. The inside edge was 22 meters long and the outside 36 meters long. These spans are hard to image when being seen in elevation. I discovered how the outer truss did not help on the overall stiffness, so instead I tried to make each step cantilevered from the inner truss and suppressed the outer truss completely. This, though, created great torsion forces in the truss. The inner truss could be quite high, as it could be incorporated inside the handrail, which needed some sort of structure anyway. But the torsion made the structural elements too big to be acceptable. Then I thought of connecting the cantilevered steps with the bottom of the truss, creating a triangular box to resist the torsion. The box was working quite well, but I realised that the upper part of the truss no longer helped the stiffness.

328

185: The edge details for the stairs was investigated by the architects and the preferred option sent back to us

I might as well just suppress this upper part and instead increase the stiffness of the box with more material. This was the solution in the end. The architects preferred to have everything done with steel plates instead of regular profiles. This, I thought, was difficult especially for the lower part which was a doubly curved surface. It was an expensive solution, but the

one they wanted. The bottom plate had to be done with techniques normally used for the shipbuilding industry â&#x20AC;&#x201C; but it would look good.

186: The monumental stair from the FEM calculation program RFEM with colors showing the deformation under live load

329

330

331

Case 3: Small-scale

332

ANTWERP PROVINCIEHUIS PRESENT

333

HERMÈS MIAMI PRESENT

Chapter thirteen

Hermès Miami

Rena Dumas Architecture Interieure (RDAI) is responsible for designing all Hermès stores around the world. Bollinger + Grohmann started to work with RDAI and Hermès for the new Paris flagship store, Hermès Rive Gauche, which opened in 2010. The store was designed in an old 1930s swimming pool that was declared a national monument in 2005 because of its strong architectural character representing the art deco style. The interior design of the Hermès Rive Gauche is characterised by three wooden structures standing on the floor of the former pool area. The wobbling shapes were designed by RDAI, as a modern interpretation of the art deco style. Starting with that vision, RDAI worked in a close collaboration with Bollinger + Grohmann to turn the concept into a feasible, selfsupporting “objects” considering the feasibility of double curved wooden elements. Besides the Hermès Rive Gauche project, Bol-

334

linger + Grohmann have most recently been involved in refurbishment projects of Hermés stores in Los Angeles and Shanghai. Hermès are planning to open a brand new flagship store in the new Miami Design District in Miami, Florida. A private contractor who owns the entire area is developing the Miami Design District, and rents out individual slots to brands that want to open a store. The construction process for the project is done in a bit of an unusual way: the landlord is in control of both tendering and construction of everything in the area, including paving, pluming and electricity, but Hermès is still in charge of choosing the design and layout of their building. Bollinger + Grohmann came into the Miami project from an early stage. RDAI had had a discussion with Hermès and the landlord where they had made a proposal for a twostory building with a roof garden. Once the sketched proposal had been agreed on, we

187: (Previous page) Rendered view from the street of the new Antwerp Provinciehuis Image courtesy of XDGA

For the HermĂ¨s Rive Gauche the geometrical analysis and the formfinding for the structures were made on rough 3D-models with a pure volumetric approach delivered by the architects. The shapes were optimised with parametric design software linked directly to a structural solver. All element parameters could be manipulated, as for example the geometry of the volumes, while the constraints and dependencies between the structural elements were maintained. The wood slats were produced in a factory on falsework. Three wooden elements with 14 mm thickness and 60 mm width are cut from flat plates and placed on the falsework. The slats had to be slightly bent and fixed in the correct geometry manually. After placing and aligning the laminates, they were glued and held together with bar clamps, to create a composite effect. At the crossing points of the rods it was necessary to develop a connection detail between the laths that could absorb torsional moments, as each wobbling shape was only braced by a horizontal compression ring in the upper part of the structures. The centre shoulder bolt was

188: The three wobbling volumes when they were just finished inside the former swimming pool Image courtesy Bollinger + Grohmann

335

HERMÈS MIAMI PRESENT

sunk in the laminated wood laths and covered by wooden covering plates, so that the connection detail was nearly invisible. one guy was responsible to collect all the parts that were drilled out in order to make room for the connection to be collected and glued back in place to hide the bolts. It has been considered to use an invisible screwing system, called “Invis Lamello”, where special male and female parts of a screw were placed on the inside of each wood slat, and screwed into each other with a magnetic screwdriver mechanism. The solution was dismissed as the insert parts had to be placed too close to the border and would not be sufficient for the torsional forces. One of the main indicators for the success of the project lay in the collaboration between the architects, with their ideas of a sustainable emerging of the former swimming pool and the wobbling shapes, the engineers with the knowledge about the geometrical constrains and the load bearing abilities of wood, and the manufacturing expertise of the fabricator Holzbau Amann.

336

189: (Opposite page) One volumne during erection at Holzbau Amannâ&#x20AC;&#x2122;s shop

190: (Above) The formfinding process, all elements could be planar. Image courtesies Bollinger + Grohmann

337

338

HERMÈS MIAMI PRESENT

191: The HermĂ¨s Rive Gauche shop seen from the stair which leads customers down on the former pool floor Image courtesy of Michel Denance

339

HERMÈS MIAMI PRESENT

started out designing the new flagship store. One of the first changes to the overall scheme was the addition of an extra story making the building fifteen meters high and with a footprint of 425 square meters. The flagship store will in principle have a quadric footprint, which is divided into four parts. The north-west part of the quadratic constitutes a public palm court, and on the other three parts the building are erected. The building is facing the street on the southand west side, on which the façade is made of a curtain wall with glass panels without spandrels. Approximately 80 cm in front of the glass façade, a semi-transparent wall of tubes is placed. The tube façade leads from the roof terrace’s railing to the ground floor, only divided by a horizontal beam in the middle. The structure of the tube façade is independent from the rest of the building, and is supported only at ground- and roof level. On the building’s others sides, walls of concrete masonry are erected. The construction site is located in the hurricane zone, with a rated wind pressure of 3,75 kN/m² or more than 350 kg per square meter. The project was mainly a façade project, but the rest of the building was also in the field of my responsibility.

INTERIOR STRUCTURE The building was to be founded directly on the ground, with a slab-on-grade foundation principle. This was out of our scope of work, as the

340

client had obtained a geotechnical survey that made proposals for the foundation. In the architects’ design stock rooms and staff facilities were placed in the back of the house – away from the street – and with big open spaces towards the glazed façade. In the beginning they had imagined that the outside columns could carry the interior floor slabs by having small “arms” sticking out through the façade. In my bachelor’s project I had made some research on fibreglass profiles, and knew of their thermal capacities. We initially discussed to construct these arms in fibreglass elements to avoid cold bridges, but eventually fire issues and the lack of shear strength in fibreglass prevented us from going further into that. We instead proposed internal columns, which actually were no problem for the architects. The conditions given to me for designing the floor slabs, were just to avoid placing additional columns. The height of the slab was not of great importance to the architects. I proposed to do the slabs as a composite structure, as the boundary conditions with supports placed on the periphery of the slab only created bending in one direction. The composite structure had several advantages: we needed something to put the floor onto and here a thin and easy to construct concrete plate was the cheapest solution; we could use this concrete that spanned between the beams structurally for compression; the local – American – construction methods make great use of steel structures and they rarely have experience with

192: Renderings of the interior store: (top) seen from the main entrance; (bottom) seen from top floor down Image courtesies of RDAI

341

342

HERMÈS MIAMI PRESENT

193: Structural drawings of the first (mezzanine) level (top left), second level (top right), and elevations of the street-facing faรงades. Drawings by author

343

HERMÈS MIAMI PRESENT

concrete. To suppress the need for formwork, the concrete was poured on top of trapezoidal steel plates. Two fire escape stairways constructed in reinforced concrete secured the buildings lateral stability. I spent a lot of time on developing the details that impacted the architecture, with the connection between the glazed façade and the concrete slab receiving the most attention. The dropped ceiling, made of plaster boards, was shaped to go up towards the façade to let more light in. On top of the slab the architects wanted to have a continuation of the finished floor out to the top of the façade transom elements. I had placed an L-profile as an edge-beam for the concrete to be poured against, and it would have been reasonable to use this for

344

the connection as well. Initially I designed the L-profile to be slightly higher than the finished level of concrete, so we could bolt through the L-profile and directly into the façade transom. But this did not allow for much room for the slotted hole, and the architects also wanted to have the rounded ceiling going all the way up to the bottom of the transom. So instead I flipped the L-profile and welded in a triangular plate where the slotted hole could be placed. Turning the L-profile around also made it possible to move the rest of the interior structure back which created room for the ceiling to have a more rounded shape, as wished by the architects. An “arm” in shape of a flat steel profile was then welded on the back of the façade mullions and could be bolted together with the triangular plate. The façade was seen as two parts; the glazed thermal barrier and the external sun shading.

194: Structural detail of the façade connection, showing the dropped ceiling, interior structure and façade. Drawing by author

GLAZED FAÇADE Hermès being one of the world’s most exclusive brands called for very exclusive architectural solutions, which in this project – as in most store designs – meant the façade. Before this project I had limited knowledge of how mullions and glass fixation where designed, so I was up for a challenge. The normal design of the connection between a glass panes and aluminium mullions are done with an external cap, hiding a screw that fixes the glass panes against the mullion. Alternatively the aluminium can be replaced by

wood or steel. Normally when using steel mullions one choses a T-profile. We proposed a newer design, patented by the Italian façade specialist Frener+Reifer, called “minimo”. In the minimo design, the external glass panes are glued to the spacer between the two glass layers, and the screw is replaced with a system where the glass is “clicked” on the mullions. The only problem, besides the cost, was that the project was in Miami and the extreme wind loads forced us to design the mullion ourselves. No standard high-end product was

195: Rendering of the façade, showing the entrance in the palm court to the Hermès sister brand Saint Louis Render by RDAI

345

HERMÈS MIAMI PRESENT

available. To build something in the hurricane zone in Miami requires either a pre-approved designed in which certain products have been certified by the Miami-Dade County, or the design has to go through a so-called missile test. The Miami-Dade County wishes to assure that debris – like wood pieces or other objects – cannot break the façade. The design of the mullions had to be as slender as possible: preferable just a flat steel plate. The first floor of the building was made as a mezzanine floor that did not all the way through to the façade, so the slab did not support the façade horizontally. We could make the mullions with full steel plates of 50 x 220 mm. One problem was still left to be solved, as the second floor had an open area where a stairway was coming up and therefore did not support the façade. In this area the façade mul-

lions spanned from the ground to the roof, and the dimension of the mullions would increase. The architects preferred to have all mullions the same dimensions, so we needed to find a solution. The problem was not only to fulfil the requirement of the mullions not tp break under extreme loads, but also to limit deformation of the whole system to meet the tolerance requirements of the minimo system – essentially to avoid the glass panes to break. I tried different options; changing the support conditions to be fixed; using the nearby vitrine (showcase) to create an intermediate support; adding a cable structure that could act as a translucent addition to the mullions; or making the transoms (horizontal mullions) structural and cantilevering them from the wall. None of the solutions seemed were very elegant to me; the most invisible solution would have been to make the transoms structural, as they functioned as an intermediate structure with overlapping glass panes in all the other areas. But that would have required a connection detail with the wall that did not seem plausible. In the end I came up with the idea of having the transom at the level of the second floor structural. This still required a rigid connection at both sides to fulfil the deformation limitations, but as the transom was in line with the second floor slab the connections were simpler. On top of that this solution allowed the vertical mullions to have the same boundary conditions as in all other areas of the façade, which I be-

346

196: Sketch made for the architects to approve my design of connection detail between structural mullions Drawing by author

197: Sketch of an alternative option were both the mullions and the transom were activated by connecting them rigidly with fully welded connection. For this option I had to consider the construction sequense, as we wanted to avoid welding on site

347

HERMÈS MIAMI PRESENT 348

lieved would make the production easier and cheaper because of more repetitions and less special cases. As neither the transom nor the mullions were integrated in to the concrete slab, a rigid connection of the transom would create torsion forces in the mullion which was unfavourable. I chose to make the transom continue to the next mullion and thereby create a hyperstatic system where the transom could rotate in connections with the slab but still function as a rigid connection would have done.

We started discussing with local façade companies to price the façade and to check the feasibility. We realised that American façade companies were not able to deliver a high-end product, and whenever we asked for something outside their catalogue they were unable to deliver. Together with Hermès we are now considering to maybe chose a standard façade approved for Miami hurricane zone or to use a European façade contractor and ship it to Miami.

198: Preliminary sketch of the steel mullion connection with the glazing

199: (Top left) “Regular” glass fixation connection with external cap. (Top right) Invisible connection without a cap. (Bottom) Initial sketch of how we wanted to create a sleek and slender design for the glazed façade

349

HERMÈS MIAMI PRESENT 350

SUN SHADING FAÇADE

mm and 250 mm.

Where the glazed façade mainly will be experienced from the inside, the sun shading façade is the first impression the customer gets. For the architects this is of highest priority. The sun shading façade was designed with the intention of a “monumental” look; columns cladded with marble and thin ceramic tubes in between. Two different sized tubes where used, with a diameter of 50 mm and 70 mm and the distance between them varied between 60

The visual intention was to have the tubes continuous from top to bottom of the façade, with only one horizontal beam supporting in the middle. Each part, below and above the horizontal beam, was 6.9 m. According to manufacturers, the maximum length of a ceramic tube was 1.5 m. Each vertical profile therefore had to be made of at least five tubes. The joints between these profiles should be as small as possible, in order to

200: Exterior view showing the façade with the sun shading. On the right the main entrance and on the left the palm court Rendering by RDAI

201: Sketch of how I thought the system of ceramic tubes would deform when a steel tube was placed inside them Sketch by author

351

HERMร S MIAMI PRESENT

make them visually appear continuous. We could not assume the ceramic to be able to withstand any high stresses. Therefore, the tubes should not be load bearing elements themselves - only transfer the load acting on them to the joints. The primary load on the tubes was the wind load, and it was assumed that the tubes were able to carry their own weight.

steel rod was strong enough, but the deformation was quite large. As the sun shading faรงade besides the visual appearance had no other function than to create shades, I had no visual deformation criteria to comply. The ceramic tubes gave me a deformation restraint though; as they were divided into at least five pieces, they would follow the deformation of

I considered many variants of the structure of the ceramic faรงade. One option was to pre-stress the tubes against each other with a cable, but this option was skipped as we were afraid of putting too much stress on the ceramic tubes. The first idea was to place a simple steel rod inside the ceramic tubes. The

352

203: Sketches of how I imagined a joint made with rubber connections. (Left) The joint shown in an exaggerated deformation situation. Sketches by author

the rod at their edges but in principle stay straight. The large deformation resulted in the steel rod touching the inside of the tubes in the middle, and thereby putting stresses on them. This could be solved by dividing the tubes into more pieces, but this was not as favorable for the architects, as it also resulted in more joints which they wanted to avoid. I calculated the deformation of a steel rod with the diameter 5 mm smaller than that of the inside of the ceramic tubes, to allow for enough room for this internal deformation. The 5 mm tolerance was an optimum between stiffness of the rod, which would increase if the tolerance was lowered, and how much deformation I could allow. An alternative solution to the simple steel rod was to put pre-stress on the rods. This would require a stiffer frame structure, but it lowered

204: (Left) Sketched principle of the deforming steel rod following a hyperbolic curvature and straight tubes. (Right) Sketch of how I imagined an inbetween support in shape of a thin steel plate could be done. Sketches by author

353

HERMĂ&#x2C6;S MIAMI PRESENT 354

205: (Top) The tubes are placed over small stumps welded on to the bottom of the frame. (Bottom) Frame sides are then welded on and in the end the top frame part is welded on, fixing the tubes losely in the frame

the deformation. I also tried a solution in which a flat steel plate was added in the middle to halve the span, but visually this was not such an appealing solution. We also proposed to investigate glass fiber reinforced concrete (GRC) or pure steel tubes, but the texture of both materials could not be as smooth and glossy as the ceramic would be. We had three European façade-specialists to price both the glazed façade and the sun shad-

ing façade early in the process to get a realistic view on feasibility, both in terms of price but also the solutions we proposed. We had made our own cost estimation, and the prices we got from the companies were way above our own. They were afraid to take risks, and to use prestressing in a façade was not normal for them. As we explained the principle and our design they eased up and reworked their estimate. It was still higher than what we expected, but one reason we discussed in the office was the fact that they knew it was for Hermès – who is known for having big budgets for projects like

206: Two frames are connected to each other at the middle. I designed the connection to be as invisible as possible with a U-profile welded on top of the bottom frame that could slide inside the bottom of the top frame

355

HERMÈS MIAMI PRESENT

this – and all companies thought they were the only collaborating contractor on the project. Hermès were getting afraid of the feasibility of the project due to the high prices, and asked for a downscaling of both the glazed façade and the sun shading façade. The architects had actually never received a final budget from Hermès, so they were not happy to downscale. We decided to go for a complete steel solution for the sun shading façade, and I proposed to construct each frame in a shop and then stack them onto each other on site. The reason was that I had heard very bad things about the American steel work on site, and in general. Allegedly Americans are not very good in fine steel work. If we could create most in a shop, maybe even a European shop, we could secure the quality of connections and finishing much better. The frames were designed to be removable, as we needed a way to access the glazed façade in case a panel would break. My idea was to make a frame welded together, with the steel tubes placed loose inside. I wanted to weld small stumps onto the top and bottom of the frame, and then glide the tubes onto the stumps. They would be held in place by the top of the frame, that would only be welded on in the end. All the steel had to be galvanized in order to withstand the salty environment close to the sea, and in this way each frame part and each tube could be treated before assembled together.

356

207: Rendering of the new HermĂ¨s Flagship Store in Miami Design District Image courtesy of RDAI

357

358

359

HERMÈS MIAMI PRESENT

Case 4: Micro-scale

360

361

PLANETARIUM SORROW PRESENT

Chapter fourteen

Planetarium Sorrow

I had already made one art piece with the architect and artist brothers Laurent and Cyril Berger, when they approached us to help them out with their latest idea Planetarium Sorrow. The previous project was a seven meter long cantilevered steel tube, placed in an art gallery at eye level. The idea was to make the visitors think about their perception of space. What did one long steel bar in the middle of the room do to their feeling of the room? The exhibition area was approximately 50 square meters, and besides the tube there were also a piece of compressed earth and a painting on one wall. Berger&Berger had already ordered the tube, so my job was limited in terms of designing the tube. They wanted the tube to be as horizontal as possible, so initially I calculated the deformed shape of the tube under its self weight and flipped it. It was not an easy shape to replicate, and the builder only had limited tools available at his shop, so

362

I suggested making the shape by bending the tube using only two different bending radii. It was not possible for the builder to do this, so instead we agreed with Berger&Berger to construct the tube with an angle at the support. The tube was welded on to a plate, which then was to be anchored into the galleryâ&#x20AC;&#x2122;s wall with Hilti anchors. The gallery was located in Le Marais in Paris, and I had explicitly told them to check the conditions of the wall before they started, but apparently they had forgotten. I received a call just after they started to construct it, saying that they had made a hole through the wall into the galleryâ&#x20AC;&#x2122;s neighbour... It turned out that the wall where they wanted to attach the tube was only made of porous stone. Under small time-pressure I had to design a new solution. They could not relocate the tube, so the only place I knew there must be something structural was above and below in the form of the floor slabs. I placed a C-shaped column

208: (Previous page) Rendered of the Planetarium Sorrow in the Tuileries Garden and with the Louvre Museum behind Image courtesy of Berger&Berger

209: (Top) Screenshot from calculation model of the cantilevered tube. (Bottom) The tube installed in Galleri Torri Images by author

363

364

PLANETARIUM SORROW PRESENT

210: The seven meter long cantilevered tube seen through the window of Galleri Torri Images courtesy Gallerei Torri

365

PLANETARIUM SORROW PRESENT

with welded end-plates and bolted it into both slabs. The tube’s end plate was then bolted to the C-profile a little under its middle point. It worked, and the tube did not fall down. I had also pledged them to put a note up, stating that the tube should not be touched as I was afraid of dynamic movements could cause a failure, but as I went down there to see the tube myself there was no note of course. But it did not fall down and it was a success.

TUILERIES 2013 AND OTHER BOUNDARIES FIAC (Foire Internationale d’Art Contemporain) is a French organisation that holds an annual art fair in Paris since 1974. Galleries from around the world have their little stand in the Grand Palais, the Petit Palais or in the temporary tents placed outside the palaces. The last years some artists have been asked to do installations in the Tuileris garden between the Louvre Museum and Place de la Concorde, and for the 2013 exhibition Berger&Berger

were asked to participate. Their idea was presenting a broken geodesic dome: the Planetarium Sorrow. The geodesic dome was invented, first patented and built by Walther Bauersfeld in 1929 in Jena, Germany. After The Second World War the Americans acquired all of Germany’s patents, and it was allegedly through the old German patent that Buckminster Fuller found the Geodesic principle and started to develop his domes – for later to acquire an American patent on the principle. The Planetarium Sorrow will be a geodesic dome with random elements culled. Berger&Berger asked us for support on ensuring that the dome would stand up securely. Geometrically a geodesic dome can be constructed in many ways; it is basically a way to make straight lines on a curved surface. The denser the grid, the closer the original sphere. Berger&Berger had decided on a geodesic dome made of 270 elements of three different lengths. The geometry was slightly rotated and cut with a plane just below the middle, which left back 184 elements. This fixed grid was one of the boundary conditions. Berger&Berger had had meetings with a sponsor who also wanted to take over the installation after it had been exposed in Tuileries. Together they had made a budget and Berger&Berger had shown renderings for a design with 108 bars. The second boundary condition was to develop a design that could stand up with only around 108 bars.

366

211: Two Alumina tubes before enamelling Image courtesy of Berger&Berger

Culling elements made the design of the node-connections more complex; in the form Berger&Berger had chosen there was normally three different joints; one regular with five connecting elements meeting, one regular with six and one none-regular with six (meaning that elements were not distributed with the same angle). Starting to cull elements made almost all the nodes different from each other, which meant a lot more work and a higher price. We needed to find a way to deal with this problem.

Berger&Berger had bought all the tubes they wanted to use before the project was even started. It was tubes of 1,5 meters with a diameter of 2 cm. They were made of Alumina Alsint 99,7%, a ceramic material, and they wanted to enamel them to retrieve a white glossy finish. We therefore had to research the material properties of Alumina but more difficult; we had to find a way to make the nodeconnections with a similar finish.

212: First rendering of the Planetarium Sorrow showed to the sponsor and FIAC organisation Courtesy of Berger&Berger

367

PLANETARIUM SORROW PRESENT

FORM FINDING THROUGH GENERATIVE OPTIMISATION We thought that this was an excellent opportunity to use scripting methods to achieve a good result and still retain randomness. I started to script an algorithm in Grasshopper that would cull (suppress) random elements, and then perform a live test of the stability using Karamba. The algorithm automatically determined where the structure was in contact with the cutting plane and placed supports at these points. The number of supports was varying as well as the number of elements. An “Evolutionary Solver” (called Galapagos) is

368

built into the Grasshopper software. It produces a chosen number of random “species” and tries to mix their good “genes” through “mutations”. The solver keeps a certain amount of the best mutation – best in the sense of what the user chooses to define as best – and then repeats the process. The solver can be used to minimise deformation or internal forces of a structure, but also just geometrical problems in general like minimum surface area of a volume. I had made my algorithm so the evolutionary solver could optimise the structure for me. As not much load will be acting on the structure, I initially simply used the deformation under the structure’s self-weight as the parameter for optimisation.

213: Sketch to explain Berger&Berger the principal relationship between number of elements and internal forces and deformation. Sketch by author

It should be noted that the Galapagos solver does not find the most optimised solution unless the options it can chose from is very limited. Instead it can be described as finding the “gene” that is most likely to succeed. The geometries that came out of the initial algorithm had the tendency to be clumped together leaving big holes in the dome. I had made sketches for Berger&Berger showing how I imagined the most structurally efficient shape a “broken” dome could take. The test results had many similarities with my sketched ideas, where the parametric development resulted in continuous arches, cantilevered denser parts and having most elements closer to the ground. These results were not reflecting the visual expression Berger&Berger sought. I started to manipulate more with the script. We tried to optimise for a minimum of bending moments, which would imply that the structure worked more as a dome with primarily normal forces. The results were still not as “random” as we wished, but more promising. We then tried with an algorithm where we had split the dome into three pieces, top, middle and bottom, and had the script take out some elements of each piece. This also gave us promising results, but still not completely satisfying. In our final attempt to design an algorithm that could produce what we saw as random, we made the script aim for having three or more elements meet at each node. This algorithm was giving us the best visual results. I had not made the number of elements a defi-

nite criterion in the algorithms, as the budget was not completely fixed and we could argue for more elements as the options with more elements seemed to work best. Another assumption that I had purposely done for the generative optimisation was to use elements rigidly connected to each other at the nodes. I did not take into account if the joints were making the structure less stiff, but as the design of the joints was being developed simultaneously with the generative form finding it was impossible to incorporate correctly at the time.

3D-PRINTING AND DESIGN OF THE NODES When we first saw the design we thought a good way to make the node-connections was to use 3D-printing. With 3D-printing it did not matter if all nodes were different, as it was only extra work for us to make the digital models.

214: Screenshot of the evolutionary solver called Galapagos while running an optimisation 212: (Next page) Generative modelling led to thirty optimised options for the overall geometry

369

370

PLANETARIUM SORROW PRESENT

371

PLANETARIUM SORROW PRESENT 372

216: Rendered of the Planetarium Sorrow in the Tuileries Garden and with the Louvre Museum behind Image courtesy of Berger&Berger

373

PLANETARIUM SORROW PRESENT 374

But how good and how strong were the materials we could print in? We established contact with one of the leading 3D-printing manufacturer in Europe: Materialise. They were able to print in many different materials, for instance titanium but this was out of our price range. They proposed another material, called Alumide. Alumide models are constructed from a blend of gray aluminum powder and polyamide, a very fine granular powder. Strength wise it is really good, around 50 MPa in tension capacity. We should never reach such values if the overall shape was designed well. In terms of elasticity it had its flaws though; the material is very easy to deform. The modulus of elasticity in the Alumina tubes was more than 100 times higher, meaning that the node design probably would have an im-

pact on the form finding process. During a meeting with Berger&Berger we came up with the connection method between the 3D-printed joints and the tubes. They wished to have as invisible a connection as possible, and they had themselves designed something where two pieces were bolted together. But in this way the viewer would notice two bolts for every arm in a node. This would only be done for connections with bars under tension, and for connections with elements always under compression their idea was to simply slide a smaller arm, attached at the joint, inside the tubes. Instead I proposed to use a magnetic screw that could be completely hidden inside the tubes. The connection had been considered for the HermĂ¨s Rive Gauche project, and we therefore had a sample connection laying in

217: Screenshot of the 3D-model made for one joint to be printed in 3D for testing Image by Bollinger + Grohmann

218: Image of a 3D-printed node with two arms for the simple connection and two arms for the glued in tension connection Image by Bollinger + Grohmann

375

376

PLANETARIUM SORROW PRESENT

219: The female part of the Invis Lamello connector glued in to the 3D-printed test piece Images by Bollinger + Grohmann

377

PLANETARIUM SORROW PRESENT 378

our office. The Invis Lamello connection, as it is called, consists of a male and a female part. The male part has magnets placed around the thread, and with a special magnet “bit” for a screwing machine, the thread can be turned when it is placed inside a ceramic tube – or in wood, as it was the initial intension of the Hermès Rive Gauche project. We would then glue in the female part in the 3D-printed nodes and the male part inside the tubes, using epoxy glue.

Next thing we did was to design a principle node with different tolerances and have Materialise print a test piece. With the test piece we also asked for different coatings, so we could see which came closest to the enamelled finish of the tubes. We received three pieces and glued in everything. It worked perfectly. One structural problem I had a hard time evaluating was the impact strength for the 3D-printed nodes. The installation would be

220: Image of the connection with Lamello Invis connector glued inside - beside the color different it worked perfectly Image by Bollinger + Grohmann

on the grass of the Tuileris, where people are not allowed to walk. But what if later the piece was installed in a museum; what would happen if somebody accidently fell into it? It could withstand a force of 100 kg according to my computer calculations, but the impact from a person falling is different â&#x20AC;&#x201C; like when a karate practitioner cuts a wooden board or a brick in half with his hand. With Berger&Berger we agreed that if this would happen, a part of the structure would have to be rebuilt. Anyway we wanted to put the strength of the 3D-printed material to a test now we had the test pieces. Cyrille Berger, the strongest looking of us, grabbed a connection with a smaller arm slid into a tube and tried to bend it. He had to use quite a lot of force, he was shaking a bit, but in the end he succeeded in breaking the connection. To our surprise he had broken the wall of the ceramic tube and not the 3D-print. We therefore accepted the material to be strong enough. In the end it turned out that all connections except one had tension forces in at least one load case, so all connections are being made with the Lamello Invis.

CHOOSING THE FINAL SHAPE We ended up with some thirty different models which all seemed structurally good, some better than others, but as the optimisation models did not include the less stiff nodes I wanted to examine the options more thoroughly to find the actual best one. The first couple of tests were omitted because of our visual criteria, but

the remaining twenty-five tests were then imported to our advanced FEM-calculation program (called RSTAB) where we could model the structure with the less stiff nodes. In the generative modelling in Karamba I had used steel as the material and a different sized tube than the real tubes. This â&#x20AC;&#x153;mistakeâ&#x20AC;? did not matter, as all the generated models had the same mistake and therefore was comparable and the optimisation method was still correct. As previously explained, I had chosen not to model the joints in Karamba, so to justify that the optimised models were in fact still optimised solutions when the joints were taken into account in the RSTAB calculations, I needed to compare the results of Karamba and RSTAB. I compared the deformation of the Karamba and RSTAB, and as the Karamba model had steel and bigger tubes it was of course stiffer and the deformation smaller. But the factor between the two calculations should have somewhat the same factor of difference. For instance if one option deformed 10 mm in Karamba and 100 mm in RSTAB, the factor between them is 10. If another option that also deformed 10 mm in Karamba deforms 200 mm in RFEM the factor is 20. Theoretically I thought that the stiffness of the joints should have somewhat the same impact on all options if these were working as a dome with mainly normal forces. It should be an overall effect that just made the whole dome sag. If the more stiff structure in Karamba was holding up a cantilevered part, then this option

379

PLANETARIUM SORROW PRESENT

was no longer optimised for the type of structure we aimed for â&#x20AC;&#x201C; a dome like structure. If this was the case, the influence of the joints had too big impact on the stiffness and I could not validate that the option was still an optimised one.

the deformation could not be the only parameter. I wanted to find the solution which carried the forces around in the best way. The options that responded with the least amount of energy when exited must be the best, but as the options did not have the same amount of elements this was difficult to determine. Instead I calculated all stresses in both the tubes and the joints. The best working option â&#x20AC;&#x201C; structurally â&#x20AC;&#x201C; should be the one with the lowest stresses as it because of that had to distribute the total load over a greater amount of elements, activating a bigger part of the structure.

By analysing the data, I could see how the factor between Karamba and RSTAB calculations usually were between five and ten. Some of the results clearly differed from the rest. I started to make a statistical analysis of the dataset; I plotted the normal distribution and chose not to proceed with options that deviated of more than 25 % of the mean; keeping the 25th percentile. Of the twenty-five test results I had to ignore five due to this criterion.

I had modelled several load cases; everything from asymmetric wind load on different areas to estimating the load from a group of pigeons even... From the calculation program RSTAB, I had all the internal forces individually as normal force, shear force in two directions, bending in two directions and torsion. I exported all the

I then started to compare the results. One comparison parameter was of course the deformation, but as all options were optimised 0,16

0,14

0,12

0,1

0,08

0,06

0,04

25 %

0,02

0 0

-0,02

380

221: Visual representation of the data statistic, showing the Gaussian curve with the 25th percentile indicated and the standard deviation of the different options plotted on the curve. Image by author

data from all the twenty-five options to Excel and calculated the von Mises stress in all tubes and all joints. Some of the options had quite high stresses in the tubes, up to 50 MPa. In most options the joints only had a maximum stress of 15 MPa. I compared each result relative to each other, which gave me the difference in percentage. I compared then stresses in the tubes, stresses in the joints and deformation, and this gave me, what I believed was, the best final option. If I only compared stresses the same solution was still the best one, and this made me confident that this option was the best one.

222: Table of the data and the results. Test 22 had the least bars, Test 07, Test 09, Test 10, Test 12 and Test 24 were all ignored in the final comparison as they deviated more than 25 % of the mean and Test 31 was the overall best performing

381

382

383

PART THREE

FUTURE

FUTURE OF ARCHITECTURAL ENGINEERING FUTURE

Chapter fifteen

Future of architectural engineering

Architecture is moving towards digital design. And the engineer is moving with architecture in this direction. Geometry has always had the highest importance in architecture and engineering, and since Peter Rice as one of the first programmed a computer to assist him on developing geometrical shapes, the computer has become more and more essential as a design tool. The engineer works in the field between science and humanism, and it is important to keep on developing both. Where architecture deals with the humane, science deals with materials. New materials are seeing the light of day in these years, and with new materials come new possibilities for design. In the future we will need to know more about ever more topics. This will increase the amount of specialists. We will see more complex solutions, and deeper collaboration will be inevitable if we are to solve the complexity of an

386

integrated design. Designs will be integrating new structural materials and higher demands for indoor climate and buildingâ&#x20AC;&#x2122;s use of energy into the architecture, with everything being optimised. â&#x20AC;&#x153;Total Designâ&#x20AC;? will be demanded.

DIGITAL DESIGN Parametric design brings new possibilities within optimisation. Geometrical parametric design is made for form finding process. Architects can optimise their layout, HVAC engineers can optimise the buildings energy performance, and structural engineers can find everything from the most efficient column layout to the design of diagonals members in a truss. Nature have always been a source of inspiration for designers and it is now moving into the digital era. Morphogenesis, the biological process that causes an organism to develop its

shape, is strived for in the digital architecture, and with it comes the need to combine digital processes with material research. Science, structure and architecture come together and the engineer is the one to combine it. The opportunities in digital design are endless. As one example we now see the revival of unfashionable materials as the brick being used for vaults because of computer technology. Another example is the adaptive structures currently being researched at the University of Stuttgart, structures that can respond to excitation with hydraulics. It is important for both architects and engineers not to lose oneself in the endless possibilities and let the computer take over. It is too easy to push a button and see what happens. It is important to still think â&#x20AC;&#x201C; yes, it requires thinking to script an algorithm, but we need to think of why we do it and not how we do it. As Ove Arup said:

â&#x20AC;&#x153;Choosing not to think, is that not indolence? Is that not escaping from the demands of your own personality, neglecting to use your skills?â&#x20AC;? 1

To become successful in making good design, it is important to keep thinking and remember the essential time-independent competences found through this book: be creative, be confident in your skills, and seek beauty while always doing your best.

223: Three images on design of diagonals in a truss; (top) the formal theoretical approach, (middle) a painting of a bridge from Romanticism and (bottom) showing the developing design of a pedestrian bridge using parametric design

387

388

FUTURE OF ARCHITECTURAL ENGINEERING FUTURE

224: (Opposite page) The image used as front cover for the very interesting book Guastavino Vaulting 223: (Above) Example on the advances within computational design of tile vault structures from Block Research Group 389

FUTURE OF ARCHITECTURAL ENGINEERING FUTURE

CONDITIONS FOR SUCCEEDING The competences are not a guarantee for success. It also depends on the environment the engineer works in. One can be lucky to work with an architect who wants to participate in an integrated design process, but it cannot be taken for given. The architectural-engineer can only be successful if the rest of the design team is set on working through an integrated design process. It is something that the engineers must push for, as it clearly creates better design and it is the sole purpose for the existence of the architectural engineer. This can be done actively through conversation, but it can also be done indirectly by showing the good results that architectural engineering brings. It can also be through new methods, as structural optimisation with digital tools as Karamba that invokes an interest in the architect. Architects and engineers are not that different, and it is my experience that if you as an engineer show enthusiasm about one solution, whether that is because it structurally is logic or it just looks good, the architects are likely to appriciate it as well. If you as an engineer can tell the story of why your favourite solution is your favourite, you are more likely to get the architects on board as well. Sometimes one crosses lines with an architect who feels that the breakdown of the master builder simply was a split into master and builder, him being the master. This type of architect often believes that the engineer

390

functions as some sort of translator between himself and authorities and contractors. The engineer’s job, in his eyes, is to make his design happen without any changes being made. The architectural engineer should strive for a change of position from this type of architect, whether doing this actively or indirect.

COMPETENCES The architectural engineer needs to be creative, but as Gropius said “design cannot be taught”. We can help it along though: I started this book by stating that when we design something we build on the work of others. It will be helpful to know the history of engineering – in an architectural context. I would also like to repeat the statement of Manfred Grohmann; “one cannot image an architect not knowing of the architectural history”. As said, engineers function in the space between science and humanism, and need to have knowledge on both. Humanism being architecture, engineers should know about architecture, but more importantly: engineers must know about their own history. It is impossible to determine what makes a person creative. It is also hard to figure out what made the successful engineers good in making structural concepts, but a few topics seem to recur: • Keep things simple: the more you know the better you can design • Spatial understanding: geometry has always had the highest importance, and be-

ing on top of things makes one able to see new opportunities and push the boundaries. Pushing boundaries is another thing that recurs throughout the history. The greatest successes are linked to new methods or materials. To push the boundaries one needs to have a perfect understanding of the theory behind the topic, thereby having confidence in oneself to suggest new ideas. A full and complete understanding of theory is important today with the ever increasing amount of norms and codes the engineer needs to comply with. One thing that is needed for the successful collaboration is – and I admit it is a bit contradictory – to not being too specialised. Specialising leads to narrow-mindedness and the specialist can more easily lose perspective. The creative engineer needs to be able to see things in perspective, to have the overview, in order to engage total design. Confidence is another important factor. Believing in yourself and in what you design is of paramount importance. To have this confidence, the engineer must possess some qualifications: • being theoretically well founded • Know how to built it • Dare to bring something of oneself into the design To create something is to open up a window to

your personality. The engineer needs to bring something other than his calculator to the table, he needs to bring himself. It is difficult to describe, so I will use the example of Centre Pompidou and Peter Rice: Architecture can be provocative – architecture is art with a function, and as art sometimes uses provocative methods so does architecture. The design of Centre Pompidou is very provocative, who thinks of erecting a giant machine in the heart of Paris’ oldest quarter? When my grandparents first saw it they asked me: “When do they take down the scaffolding?” The Centre Pompidou is now more than thirty-five years old, it is still provocative, but the design is no longer criticised. It somehow fits now. Why? I am confident the reason is the humane detailing of the Gerberettes. It is Peter Rice’s need to bring his person into the design that has made the provocative architecture survive. He wanted to use cast steel to shape the Gerberettes more humane, more organic – he was inspired by the old iron structures of the structural artists, who showed their personality through cast iron decorations. The engineer needs to take responsibility for the design, just as the architect always does it. When the architectural engineer calls for the integrated design process it is not only the architect who has to give something, the engineer needs to bring something as well. The outcome of the integrated design process should not only be the architects “baby”, it should be a common feeling that the design belonged to everyone in the team. Then it will afterwards

391

FUTURE OF ARCHITECTURAL ENGINEERING FUTURE

not only be the architect nursing on details, it will also be the engineer. And the engineer will automatically perform his best, because only the best will be good enough. When all nursing is done, the engineer then needs to be able to convince others in the quality of the design. The builder must feel safe with the design, otherwise he will make mistakes and the price will be unnecessarily high. And cost is another factor. The engineer needs to be able to optimise the structures, and know of standardisation and repetition. A building with a thousand different windows is rarely affordable (not even for the rich part of Belgium...), and to think construction sequence and method into the design can solve many problems and save cost and hassle. To not only think through the design, but to actually see it through to the end, is very important. It gives the contractor or builder security and confidence, and the engineer can help solve unforeseen problems, which in the end ensure that the quality is intact. It will be a natural part of taking responsibility for the design, to make it personal. It all takes a sense of creativity, confidence, strive for beauty – but most important the engineer needs to dare. One Ove Arup’s favourite philosopher Søren Kierkegaard once said:

392

“To dare is to lose your foothold for a moment, not to dare is to lose yourself.”

393

394

395

Statement of the problem Written February 2013 - struck-through sentences are left out in final version.

Since the industrial revolution the role between architects and engineers has been ever changing. Before the industrial revolution no distinction between the architect and engineer was made - typically all development of a construction project was in the hands of the Master builder. With the industrial revolution introducing new materials as iron, steel and reinforced concrete, new forms and shapes became possible. These new possibilities within construction called for a greater specialisation within the fields of both architecture and technique. Designing is to shape an interface - from industrial interfaces as cars or clothing to architecture. In the time of the master builder, the term “design” denoted a drawing with a guiding principle of how to construct. Nowadays the term design is increasingly understood as an interdisciplinary field that combines information from multiple areas. Engineers and architects perceive designing differently; engineers design methodological to determine the most efficient structure, where architects design imaginative to create a vision of appearance, functions and humanistic subjects. Designing a new construction can be simplified to technical and none-technical issues.

“I would distinguish the difference between the engineer and the architect by saying the architect’s response is primarily creative, whereas the engineer’s is essentially inventive.” 1 -Peter Rice

“These are the engineer’s responsibilities: the respect of the physical laws, the strength of materials, supply, economy considerations, safety, etc. And these are the architect’s: humanism, creative imagination, love of beauty, and freedom of choice. In my drawing, the engineer’s sphere casts a reflection on that of the architect - the reflection of the knowledge of physical laws. Similarly, the architect’s understanding of human problems is reflected in the sphere of the engineer.” 2 -Le Corbusier

In his book, written in 1992, well-known engineer Peter Rice describes the distinction between engineers and architect essentially in the same way as famous architect Le Corbusier did more than 60 years earlier. Now a day’s engineers have access to more information on materials than the first

396

engineers did back in the beginning of the 20th century, but the role of the engineers and architects apparently stay the same? In this thesis it is my intent to analyse which competences an engineer should possess to be able to work within this field. I want to investigate what is essential time independent engineering competences the engineer must possess to contribute to architecture. Herein lays also the questions of what is not time independent universal engineering knowledge, what has become universal knowledge through time and perhaps guess what will become universal knowledge in the future? Contribution to architecture can be done in all stages of the building process, the further in the project the higher level of detail. I want to limit my work in this thesis to concern only the earliest stages of the process, conceptual design. Throughout my work on competition entries and other projects in conceptual design stages at the Paris office of engineering company Bollinger-Grohmann, I will research which competences is needed to consult the architects I collaborate with the best way. To analyse which competences is time independent, I want to see the knowledge I gain in, among other, a historical perspective (historical, local, international, ...), put the work in context with some of the greatest engineering works since the industrial revolution.

BREAKDOWN OF TASKS: • 50 %:

Analysing how historically famous engineers have contributed to the architecture in their projects

• 40 %:

Analysing work projects, split into four categories: mega-scale, large-scale, small- scale and micro-scale

• 10 %:

Future of architectural engineering

225: L’architecte/l’’ingénieur diagram which appears in Le Corbusier’s Précisions sur un état présent de l’architecture et de l’urbanisme from 1930

397

Notes Introduction: 1 http://www.bollinger-grohmann.com/en.about-us.html 2 Manfred Grohmann, interview cited in Christian Schittich, Peter Cachola Schmal (ed.), Bollinger + Grohmann (MĂźnchen: Detail, 2013), 8

Preface: 1 Ove Arup, The world of the structural engineer, 1968 2 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 59

Chapter one: 1 Jacques Heyman, The Science of Structural Engineering (London: Imperial College Press, 1999), 9-25 2 Vitruvius, De Architectura, Book I, Chapter 1, section 1. Translation by Bill Addis. 3 Vitruvius, De Architectura 4 Steen Eiler Rasmussen, Experiencing Architecture (Cambridge, Massechusetts: Technology Press, 1959), 106-110 5 http://www.denstoredanske.dk/It,_teknik_og_naturvidenskab/Matematik_og_statistik/Regning,_algebra_og_ talteori/Fibonaccital 6 Martin Papirowski, Giants of Gothic â&#x20AC;&#x201C; Reaching for Heaven (Film Produktion Stein e.k., Niemcy, 2011, http://www. dr.dk/tv/se/katedraler/) 7 Jacques Heyman, The Science of Structural Engineering (London: Imperial College Press, 1999), 14-18 8 Jacques Heyman, The Science of Structural Engineering (London: Imperial College Press, 1999), 37-45 9 The History of Concrete (Dept. of Materials Science and Engineering, University of Illinois, Urbana-Champaign) 10 R. Mark and P. Hutchinson, On the structure of the Pantheon (Art Bulletin, 1986), 29-31 11 Leon Battista Albert (translated by J. Rykwer, R. Tavernor and N. Leach), On the Art of Building (Cambridge, Massachusetts: The MIT Press, 1988), Book II, 56 12 Vitruvius, De Architectura, 3.1.2-3

398

Chapter two: 1 Galileo Gelilei, Discorsi e dimostrazioni matematiche, intorno à due nuove scienze (1638) 2 Jacques Heyman, The Science of Structural Engineering (London: Imperial College Press, 1999), 28-32 3 Jacques Heyman, The Science of Structural Engineering (London: Imperial College Press, 1999), 76-101

Chapter three: 1 Kenneth Powell, The Great Builders (London: Thames & Hudson Ltd, 2011), 66-67 2 Den Store Danske – Gyldendals åbne encyklopædi (Gyldendal) 3 Thomas Telford (edited by John Rickman and published after Telford’s death), Life of Thomas Telford, Civil Engineer (London: James and Luke G. Hansard and Sons, 1838), 34 4 Mike Chrimes (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 68-72 5 Samuel Smiles, The Life of Thomas Telford (Teddington: The Echo Library, 2006), 156-157 6 Alexander Immo, The Theory of Bridges, New Edinburgh Encyclopedia, Second American edition, Vol. IV (New York: Whitning and Whatson, 1814), 480, 483 7 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 30-44 8 George Kubler, The Shape of Time (Yale University Press, 1962), 16 9 Mike Chrimes (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 70-73 10 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 39-40 11 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 41 12 Steven Brindle (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 100-106 13 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 49-52 14 L.T.C. Rolt, Brunel (London: Longmans, 1957), 411 15 Edward Diestelkamp (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 87-94 16 Théophile Seyrig, Le pont sur le Douro - Memoires des travaux de la Societe des Ingenieurs Civils (Sept.-Oct. 1878), 743-816 17 Gustave Eiffel, Les grandes constructions métalliques (Paris: Association française pour l’avancement des sciences, 1888), 10-11

399

18 Bertrand Lemoine (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 141-147 19 Martin Bressani (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 113-118

Chapter four: 1 Karla Cavarra Britton (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 178-180 2 Gwenaël Delhumeau (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 148153 3 Ulrich Pfammatter, Building the future (Munich: Prestel Verlag, 2008), 106-115

Chapter five: 1 Heinrich Heine, 1843, quoted from W. and K.-H. Manegold, Treue (1966), 85 2 Dansk Arkitektur Center, DAC-Learning, arksite plus: http://www.dac.dk/da/dac-learning/netundervisning/arksiteplus-1/arkitekturhistorie-1/modernismen-1/ 3 Robert McCarter (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 170-176 4 David Dunster (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 184-187 5 Stanford Anderson, Peter Behrens and a New Architecture for the Twentieth Century (Cambridge, Massachusetts: The MIT Press, 2000) 6 Quote from Le Corbusier found in the house’s visitor brochure published by the Centre des Monuments Nationaux 7 Tim Benton (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 188-195 8 Dansk Arkitektur Center, DAC-Learning, arksite plus: http://www.dac.dk/da/dac-learning/netundervisning/arksiteplus-1/arkitekturhistorie-1/modernismen-1/ 9 Dansk Arkitektur Center, DAC-Learning, arksite: http://www.dac.dk/da/dac-learning/netundervisning/arksite-1/ temaer-1/arkitekter-1/baggrund-1/le-corbusier-1/

Chapter six: 1 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 148-153 2 Robert Maillart, Construction and Aesthetic of Bridges, (The Concrete Way, May–June 1935) 3 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 155-178

400

4 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 162 5 Eugène Freyssinet, ”Souvenirs” in Cent Ans de Béton Armé (Paris: Editions Science et Industrie, 1949), 52 6 Eugène Freyssinet, Freyssinet by himself, A Half-Century of French Prestressing Technology (special English translated edition of Travaux 50, April-May 1966, originally published in 1933), 18 7 Eugène Freyssinet, Freyssinet by himself, A Half-Century of French Prestressing Technology (special English translated edition of Travaux 50, April-May 1966, originally published in 1933), 9 8 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 206 9 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 194-297 10 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 220-232 11 Mario Salvadori, foreword to The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958), vi 12 Eduardo Torroja, The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958), 49-51 13 Rudolf Albrect, Heinrich Gerber (Munich: Deutsches Museum, 1963) 14 Eduardo Torroja, The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958), 59 15 Eduardo Torroja, The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958), 53 16 Eduardo Torroja, The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958), 12 17 Eduardo Torroja, The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958), 10 18 Eduardo Torroja, The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958), 7 19 Eduardo Torroja, The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958), 12 20 Mario Salvadori, foreword to The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958), vii 21 Eduardo Torroja, The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958), 8 22 From a speech given by Nervi in London, October 1955, published in full as Developments in Structural Technique in The Architect’s Journal (London, 1955) 23 Nervi, unpublished manuscript cited in: Ada Louise Huxtable, Pier Luigi Nervi (New York: George Braziller, Inc., 1960), 22-23 24 From a speech given by Nervi in London, October 1955, published in full as Developments in Structural Technique in The Architect’s Journal (London, 1955) 25 From a speech given by Nervi in London, October 1955, published in full as Developments in Structural Technique in The Architect’s Journal (London, 1955)

401

26 From a speech given by Nervi in London, October 1955, published in full as Developments in Structural Technique in The Architect’s Journal (London, 1955) 27 From a speech given by Nervi in London, October 1955, published in full as Developments in Structural Technique in The Architect’s Journal (London, 1955) 28 From a speech given by Nervi in London, October 1955, published in full as Developments in Structural Technique in The Architect’s Journal (London, 1955) 29 Riccardo Dirindin (edited by Kenneth Powell), The Great builders (London: Thames & Hudson Ltd, 2011), 201-205 30 Ada Louise Huxtable, Pier Luigi Nervi (New York: George Braziller, Inc., 1960), 11-31 31 Nervi, unpublished manuscript cited in: Ada Louise Huxtable, Pier Luigi Nervi (New York: George Braziller, Inc., 1960), 30-31

Chapter seven: 1 Sigfried Giedion, Space, Time and Architecture (Cambridge, Massachusetts: Harvard University Press, 1967), 24 2 Le Corbusier, Towards a New Architecture (London: Architectural Press, 1970), 7 3 George R. Collins, Antonio Gaudí and the Uses of Technology in Modern Architecture, Civil Engineering: History, Heritage and the Humanities (Princeton, New Jersey: Department of Civil Engineering, Princeton University, 1971), 1: 70 4 George R. Collins, Antonio Gaudí and the Uses of Technology in Modern Architecture, Civil Engineering: History, Heritage and the Humanities (Princeton, New Jersey: Department of Civil Engineering, Princeton University, 1971), 1: 69 5 George R. Collins, Antonio Gaudí (New York: George Braziller, 1960), 23 6 Jordi Oliveras (edited by Kenneth Powell), The Great builders (London: Thames & Hudson Ltd, 2011), 153-161 7 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 183-186 8 Felix Candela, New Architecture, Maillart Papers (Princeton, New Jersey: Department of Civil Engineering, Princeton University, 1973), 119 9 Colin Faber, Candela, The Shell Builder (New York: Reinhold Publishing Corporation, 1963), 14 10 David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983), 189-193 11 Colin Faber, Candela, The Shell Builder (New York: Reinhold Publishing Corporation, 1963), 80 12 R. Buckminster Fuller, quoted in The New York Times, June 15, 2008, Design – A 3-Wheel Dream That Died at Takeoff by Phil Patton 13 R. Buckminster Fuller, Public lecture at Columbia University, 1965

402

14 Loretta Lorance (edited by Kenneth Powell), The Great builders (London: Thames & Hudson Ltd, 2011), 208-216 15 Frei Otto, quoted in The Great builders (London: Thames & Hudson Ltd, 2011), edited by Kenneth Powell, 253

Chapter eight: 1 Sir Alan Harris, letter to Ove Arup cited in: Peter Jones, Ove Arup (London: Yale University Press, 2006), 194 2 Wolf Prix, transcript by author from interview with “What is architecture?” (http://www.whatisarchitecture.cc/2013/05/ wolf-prix.html) 3 Ove Arup, school essay ”What I want to do when I grow up and why”, cited in Peter Jones, Ove Arup (London: Yale University Press, 2006), 10 4 Ove Arup, “On hospitality and entertaining” cited in: Peter Jones, Ove Arup (London: Yale University Press, 2006), 6 5 Ove Arup, school essay ”The benefits of and dangers of wealth”, cited in Peter Jones, Ove Arup (London: Yale University Press, 2006), 10 6 Ove Arup, BBC interview from the program ”People today”, 1964 (not broadcasted) 7 Ove Arup, BBC interview from the program ”People today”, 1964 (not broadcasted) 8 Ove Arup, BBC interview from the program ”People today”, 1964 (not broadcasted) 9 Peter Rice, A celebration of the life and work of Ove Arup, paper delivered to a meeting of the Royal Society of Arts on 1 March 1989 10 Ove Arup, diary from January 5th 1924, cited in: Peter Jones, Ove Arup (London: Yale University Press, 2006), 33-34 11 Ove Arup, BBC interview from the program ”People today”, 1964 (not broadcasted) 12 Letter from Grete Backe to Ove Arup, October 12th 1924, cited in: Peter Jones, Ove Arup (London: Yale University Press, 2006), 38 13 Peter Jones, Ove Arup (London: Yale University Press, 2006), 38 14 Ove Arup, Reinforced concrete in relation to present-day design, Concrete and Constructional Engineering, 21 (3) (1926) 15 Peter Jones, Ove Arup (London: Yale University Press, 2006), 27-28 16 Peter Jones, Ove Arup (London: Yale University Press, 2006), 54 17 Ove Arup, The world of the structural engineer, 1968 18 Ove Arup, Art and architecture, Royal Gold Medal address, June 21st 1966 (RIBA Journal, August 1966)

403

19 Ove Arup, Philosophy and the art of building, 1974 20 Peter Rice, A celebration of the life and work of Ove Arup, paper delivered to a meeting of the Royal Society of Arts on 1 March 1989 21 Ove Arup, cited in Peter Jones, Ove Arup (London: Yale University Press, 2006), 59 22 Ove Arup, The engineer looks back, 1979 23 Ove Arup, The engineer looks back, 1979 24 Ove Arup, Notes for DOMUS, 1983 (Arup Papers, 2/99) 25 Letter from Ove Arup to Malcolm Reading, April 24th 1985, cited in: Peter Jones, Ove Arup (London: Yale University Press, 2006), 60 26 Letter from Ove Arup to Malcolm Reading, April 24th 1985, cited in: Peter Jones, Ove Arup (London: Yale University Press, 2006), 60 27 Ove Arup, BBC interview from the program ”People today”, 1964 (not broadcasted) 28 Nina Rappaport, Support and resist: structural engineers and design innovation (New York: The Monacelli Press, Inc., 2007), 14 29 Ove Arup, ”Introduction” to Air Raid Precautions: Report to the Finsbury Borough Council by Messrs. Tecton, Architects on the Structural Protection for the People of the Borough against Aerial Bombardment (February 6th, 1939), 2 30 Ove Arup, Science and world planning, 1942 31 Peter Rice, A celebration of the life and work of Ove Arup, paper delivered to a meeting of the Royal Society of Arts on 1 March 1989 32 Peter Rice, A celebration of the life and work of Ove Arup, paper delivered to a meeting of the Royal Society of Arts on 1 March 1989 33 Ove Arup, Post graduate education of civil engineers, Belfast conference, September 1958 34 Peter Jones, Ove Arup (London: Yale University Press, 2006), 175 35 Peter Jones, Ove Arup (London: Yale University Press, 2006), 240 36 Peter Jones (edited by Kenneth Powell), The Great Builders (London: Thames & Hudson Ltd, 2011), 222 37 Peter Rice, A celebration of the life and work of Ove Arup, paper delivered to a meeting of the Royal Society of Arts on 1 March 1989 38 Peter Rice, A celebration of the life and work of Ove Arup, paper delivered to a meeting of the Royal Society of Arts on 1 March 1989 39 Peter Rice, A celebration of the life and work of Ove Arup, paper delivered to a meeting of the Royal Society of Arts

404

on 1 March 1989 40 Felix Candela, in Festskrift for Ove Arup, 1965, cited in: Peter Jones, Ove Arup (London: Yale University Press, 2006), 252 41 Ove Arup, The key speech, 1970 42 Ove Arup, The key speech, 1970 43 Ove Arup, The key speech, 1970 44 Peter Rice, A celebration of the life and work of Ove Arup, paper delivered to a meeting of the Royal Society of Arts on 1 March 1989

Chapter nine: 1 Peter Rice, from the documentary movie Traces of Peter Rice (Arup, 2012) by B. Richardson, K. Eichhorn and J. Greitschus: http://video.arup.com/?v=1_q7rgc4aj 2 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 51-52 3 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 57 4 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 57 5 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 59 6 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 59 7 Jack Zunz (edited by Kevin Barry), Traces of Peter Rice (Dublin: The Lilliput Press, 2012), 16 8 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 60 9 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 63 10 Peter Rice to Jack Zunz, reproduced by Jack Zunz (edited by Kevin Barry) in Traces of Peter Rice (Dublin: The Lilliput Press, 2012), 16 11 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 25 12 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 26 13 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 27 14 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 29 15 Richard Roger in conversation with Jonathan Glancey (edited by Kevin Barry) in Traces of Peter Rice (Dublin: The Lilliput Press, 2012), 46

405

16 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 34 17 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 115 18 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 33 19 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 36 20 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 46 21 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 127 22 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 81-85 23 Bill Addis, Building: 3000 Years of Design and Construction (London: Phaidon, 2007), 528-529 24 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 110 25 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 110-111 26 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 111 27 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 88 28 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 90 29 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 119 30 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 125 31 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 125 32 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 100 33 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 97 34 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 100 35 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 101 36 Richard Roger in conversation with Jonathan Glancey (edited by Kevin Barry) in Traces of Peter Rice (Dublin: The Lilliput Press, 2012), 46 37 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 112-113 38 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 53 39 Jack Zunz (edited by Kevin Barry), Traces of Peter Rice (Dublin: The Lilliput Press, 2012), 14 40 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 31

406

41 Jack Zunz (edited by Kevin Barry), Traces of Peter Rice (Dublin: The Lilliput Press, 2012), 18 42 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 72 43 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 72 44 Frank Stella, “Introduction” to Peter Rice, An Engineer Imagines (London: Artemis, 1994), 23 45 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 74 46 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 53 47 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 73 48 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 75 49 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 76 50 Peter Rice, from the documentary movie Traces of Peter Rice (Arup, 2012) by B. Richardson, K. Eichhorn and J. Greitschus: http://video.arup.com/?v=1_q7rgc4aj 51 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 77 52 Jack Zunz (edited by Kevin Barry), Traces of Peter Rice (Dublin: The Lilliput Press, 2012), 19 53 Barbara Campbell-Lange, from the documentary movie Traces of Peter Rice (Arup, 2012) by B. Richardson, K. Eichhorn and J. Greitschus: http://video.arup.com/?v=1_q7rgc4aj 54 Amanda Levete, from the documentary movie Traces of Peter Rice (Arup, 2012) by B. Richardson, K. Eichhorn and J. Greitschus: http://video.arup.com/?v=1_q7rgc4aj 55 Richard Roger in conversation with Jonathan Glancey (edited by Kevin Barry) in Traces of Peter Rice (Dublin: The Lilliput Press, 2012), 45 56 Wolf Prix, transcript by author from interview with “What is architecture?” (http://www.whatisarchitecture.cc/2013/05/ wolf-prix.html)

Chapter ten: 1 Peter Cook (edited by Peter Cachola Schmal), preface to Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 7 2 Klaus Bollinger (edited by Peter Cachola Schmal), Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 14 3 Michael Schumacher (edited by Peter Cachola Schmal), Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 26 4 Enrico Santifaller (edited by Peter Cachola Schmal), Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004),

407

22-24 5 Petra Hagen Hodgson (edited by Peter Cachola Schmal), Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 41 6 Peter Cachola Schmal, Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 57 7 Christof Bodenbach (edited by Peter Cachola Schmal), Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 55 8 Peter Cachola Schmal, Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 14 9 http://www.karamba3d.com/about/ 10 http://www.karamba3d.com/infobox-competition/ 11 André Chaszar (edited by Peter Cachola Schmal), Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 75 12 Peter Cachola Schmal, Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 88 13 Klaus Bollinger and Manfred Grohmann (edited by Peter Cachola Schmal), Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 12 14 Klaus Bollinger and Manfred Grohmann (edited by Peter Cachola Schmal), Workflow: Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004), 18

Chapter eleven: 1 Rod Sheard, The Stadium: Architecture for the New Global Culture (Singapore: Periplus Editions, 2005), introduction 2 Rod Sheard, The Stadium: Architecture for the New Global Culture (Singapore: Periplus Editions, 2005), introduction

Chapter fifteen: 1 Peter Jones, Ove Arup (London: Yale University Press, 2006), 22

Statement of the problem: 1 Peter Rice, An Engineer Imagines (London: Artemis, 1994), 72 2 Le Corbusier, Précisions sur un état présent de l’architecture et de l’urbanisme, (Paris: E. Grès., 1930), 12

408

409

Bibliography Bill Addis, Building: 3000 Years of Design and Construction (London: Phaidon, 2007) Peter Rice, An Engineer Imagines (London: Artemis, 1994) Christian Schittich, Peter Cachola Schmal (ed.), Bollinger + Grohmann (München: Detail, 2013) Ulrich Pfammatter, Building the future (Munich: Prestel Verlag, 2008) Steen Eiler Rasmussen, Experiencing Architecture (Cambridge, Massechusetts: Technology Press, 1959) Kenneth Powell (ed.), The Great Builders (London: Thames & Hudson Ltd, 2011) Peter Jones, Ove Arup (London: Yale University Press, 2006) Ada Louise Huxtable, Pier Luigi Nervi (New York: George Braziller, Inc., 1960) Jacques Heyman, The Science of Structural Engineering (London: Imperial College Press, 1999) Eduardo Torroja, The Structures of Eduardo Torroja (New York: F. W. Dodge Corporation, 1958) David P. Billington, The Tower and the Bridge (Princeton, New Jersey: Princeton University Press, 1983) Kevin Barry (ed.), Traces of Peter Rice (Dublin: The Lilliput Press, 2012) Peter Cachola Schmal (ed.) Workflow: Architecture – Engineering (Basel: Birkhäuser, 2004)

410

411

HOW ENGINEERS CONTRIBUTE TO ARCHITECTURE

This book takes the reader on a journey through the history of engineering; from the

CREATIVE ENGINEERING

Creative Engineering - How Engineers Contribute to Architecture is a search through

CREATIVE ENGINEERING HOW ENGINEERS CONTRIBUTE TO ARCHITECTURE

Kristjan Plagborg Nielsen