network partners:

Workshop/Seminar 2

Digital Crafting March 29-31 2010, Report by Terri Peters

The second Digital Crafting workshop event “How to Join” was held at the Royal Academy of Fine Arts, School of Architecture in Copenhagen March 29-31 2010.  The invited guest Zurich-based Christoph Schindler of Schindler Salmeron led the two-day workshop and one-day seminar investigating ideas about joining materials and geometries using parametric design.  The workshop focused on experimenting with the Zip Shape technology created by Christoph Schindler.  Workshop participants developed variations on the idea of bending two non-identical “teeth” together like a zipper, to hold a bended shape in place.  Rather than using a mould, (the usual way of creating a complex, bended shape) using this method the material is held in place by the design of the “teeth”.  The seminar on the third day related examples of parametric wood construction, with lectures by Michael Hensel from AHO Oslo, Sigurdur Omasson from DTU Copenhagen, Silvan Oestererle from ETH Zurich and Sebastian Gmelin from Arkitektskole Aarhus.

Working back and forth from parametric model to physical prototype, early studies explored: Where is the strength coming from? How do we design the “teeth”?  What bit will bend? How is the material buckling under pressure?  Does the strength come from the glue?

Interesting discussions happened during the workshop and at the seminar including ideas about:

How can we work with material tolerances?

While digital models may assume zero tolerance, it is impossible to make a production or material technology that delivers this performance.  For example, what about the glue?  The glue is a material in itself so it has to go somewhere, and we found that the parametric model has to be designed with the material in mind.   The material tolerances allow a hand made quality, like with steam bending timber for chairs as in the examples in Schinder´s lecture about Zip Shape.   The foam prototypes are precise and machine cut, but that doesn´t mean they are all performing exactly the same, even from the same or similar digital files.  The material breaks, tears and crumbles in unexpected ways under pressure.  Even in the foam prototypes, Zipshape is about using the material to hold the form with glue, rather than using a stamp or mould, which treats the material in a different way.

What is the relationship between variability and complexity?

During the workshop and seminar we discussed definitions of complexity – and the difference between something “complex” and something ”complicated”.  Handcrafting can, of course, be extremely complex, and it seems to not be about scale, but rather connections and joining.

In Schindler´s lecture, he traced the way that tools have developed and how they have impacted technology.  In his examples, the tool changes the material, and creates variation.

What about processes that are reversible, like weaving, where stitches and joining can be undone and redone again?  We talked about hyper-specified, high performance surfaces where the designer is able to add material exactly where the force will be and about how important it is to understand the specific material performance, rather than the general material properties.  Or is it more about understanding overall behaviour, less about how individual pieces work, but rather about how they work together?

Ways of connecting and joining

With Schindler´s lecture as a starting point, we talked about how joining materials and parts usually works in architecture, and thought about examples of conventional methods.  Schindler showed examples of dovetail joints where all the parts are different but not interchangeable with joint marks needed to identify what goes where.  In our group discussion we touched on ideas of joining materials without adding materials or processes, like in weaving, where the joining is in the material itself, rather than in an additional binding agent.  The Zip Shape and Snap Fit examples are more like weaving than like traditional timber framing, because the joining is an extension of the material performance.  (but what about the glue?)

Progress and technology – what is the future for architecture and design?

Just because we can do something, should we? In the discussion based on the Rosalind Williams reading, “Retooling“ we had a group discussion about progress and technology.  What are society´s agents of change? What leads progress? Technology? Societal “needs”?  We discussed how as a profession, architecture is not in agreement about the aims of digital design and fabrication – is it to improve design quality? Cost? Efficiency? Is there a “sustainable” agenda?

To our group, it seems that digital design and fabrication is nearly the norm, we are thinking parametrically and we are generally convinced it is the future.  But is this really widespread considering the vast majority of the building industry values other criteria?   To paraphrase science fiction writer William Gibson, “the future is here, it is just not very evenly distributed”.  We discussed the idea that digital tools allow designs or communication that “could not have been done by hand” –but is this really true? The group discussed ideas about this and asked “does it even matter”?


The discussion began to be about whether or not all design needs to have a “sustainable” agenda.  We did not define the term, which led to everyone having their own idea about what it could mean and therefore not really talking about the same thing, but most agreed it is an overused term but important nonetheless.  Someone asked, “Is there a “need” for digital crafting” and discussion followed about the role of research, design, and innovation.  Surely “sustainability” is not only about “what we need” at the expense of “what we want”?   Some of the group believe it difficult to be critical of sustainability and sustainable agendas but of course this cannot be an excuse for not engaging with such an important issue.  Design quality is also a “difficult” subject but it must be debated and understood.   The group wondered if Digital Crafting should shy away from issues of sustainability or develop a considered understanding of key concepts.  We talked about cost and efficiency as possibly relating to sustainability.  Developing new ideas about material performance is a possible step towards minimizing waste and creating innovative ways of working that could be considered sustainable.  With no mould, the same cost applies if one make one or one hundred Zip Shape prototypes because its about how much the material costs and the time that goes into it.  Zip Shape is not well suitable for mass production in that way but do these ideas belong in a discussion of sustainability?  Is all good design “sustainable”? Is all “sustainable” design good?

Workshop 2 Investigations - "How To Join"

The workshop was held on March 29 and 30, with participants working in groups to explore the parametric possibilities of Zip Shape.  The formal and geometric experiments were materialized using digital fabrication, allowing the participants to create foam prototypes.

Anders Holden Deleuran and Tore Banke were trying to challenge the idea that the zip-shape fabrication method could only be used for rectangular sheets and single profiles extruded along a vector.   They began by projecting (somewhat) freeform shapes onto the rectangular zip-sheets and trimming them, resulting in generating more loosely defined and open-ended surfaces.  Then they attempted to use developable surfaces in the zip-shape method to generate surfaces with a more complex curvature than simply an extruded profile.  Due to the limited time, they decided to make a digital model and series of physical models, which combined both techniques. They discovered that when using the “Projection” technique, it is better to use a surface created by extruding a single profile, otherwise the projections will become offset.  They also discovered that when using developable surfaces it is important that the surfaces are actually developable and do not have too much curvature. If not, the two zip-sheets and their teeth might be hard to generate and actually fit together.  They found that the foam did not work well with steep curvatures where the “bit in-between teeth” would break easily.  They said they could build on these studies by designing smaller teeth using less brittle materials, although generally they found the foam worked quite well for quick prototyping of concepts.

Sebastian Gmelin, Mette Ramsgaard Thompsen and Claus Peder Pedersen were investigating complexity through standardization in a series of prototypes that were about the idea of experimenting with alternating the thickness of the zip and also about branching, and possibilities of skipping “teeth”.  They made a “zip worm” prototype and discovered new ideas about covering the sides of the zip and also about how the prototypes perform in tension (they break).   The group also tried to create a Zip shape that not only bends, but that varies in thickness.  They didn´t quite get it to work but they think it could be done. They had to change their digital model during the design process as the algorithm was based on the two flanks of the shape running parallel.

Niels Martin Larsen, Johannes Rauff Greisen, Karen Marie Hasling, and Asbjørn Sondergaard were interested in the space defining possibilities of the material, and became interested in interlocking spaces.  Could a design be developed that could grow – with more pieces being joined on the end to interlock and weave?   They looked at three different curves and tried to understand which part is doing what.  The group says their biggest challenge was to figure out how to get the splits to curve in different directions individually. Using a parametric model, they developed a design in which three separate tracks of zip-shapes were positioned parallel to one other. Each zip-shape track was aligned in one end to the track next to it, and then due to their different curvatures, the tracks would gradually get out of sync towards the other end. This way, the 3 tracks were linked to each other in a kind of Z-shaped formation.  They had some trouble getting the material to curve enough in the physical model.  They learned that the slicing of the shape in the digital model would have to happen in 3D to avoid the Zip tracks not lining up.  They found this was not so visible using the blue foam but they figured it would be visible if they used another material such as wood.

Interview with Christoph Schindler following Workshop/Seminar 2

Interview with Christoph Schindler following Seminar/Workshop 2 by Terri Peters

TP: In your lecture, you said that architecture is becoming more about showcasing the process, the “how” of a building and maybe less about the “why” – How does this relate to parametrics?

CS: This seems to me one of the big questions raised in our discipline today. Most of us grew up with an education that understood architecture as part of the arts and thus history of architecture as part of art history or at least cultural history, applying terms such as «styles» to describe and structure its development.
Today, nobody doubts the central role information technology has taken or is about to take in the building sector.  3D-modelling tools, parametric models, research on self organization and computer-aided machine tools are not only omnipresent means, but appear to have become the self-sufficient focus of the discourse. It is difficult to relate those technologies and their formal output to art history as we know it since the 19th century. Therefore I wonder whether it could be history of technology itself which might help us to describe and contextualize contemporary development in architecture…

TP:  Does this relate to the discussions we were having relating to mass customization and how it could relate to parametric design?

CS: The term «mass customization» was coined in the context of customer integration as a concept offering individual products to individual people, such as tailor-made suits. Thanks to computer-aided manufacturing systems these products could be offered for the price of mass production. For some reason, mass customization as a concept of customer relationship marketing has not taken off so far. However, the term was discovered by architects and started a quite successful career in architectural discourse. What lost significance in the architectural context is the individual customer as a good reason for individual production—and thus somehow the socio-cultural backbone of mass customization.

TP: Zipshape does not necessarily relate to the material qualities of wood, did you try anything else and why have you chosen wood to work with?  What challenges and opportunities have you found working with wood?

CS: The toothed section of a Zipshape panel consists of two different areas – the teeth and the thin sheet that connects them. While the teeth are defining the geometry, the thin sheets are connecting them and take most of the tensile forces, just like in a trussed beam. The teeth indeed do not relate to the specific material qualities of wood, whereas the sheets have to resist to tension and be bendable at the same time—a perfect match for wood’s fibre structure and its anisotropic behaviour. For the «Ziprocker» rocking chair we even used different wood materials for these two areas— low-priced pine boards for the core with the teeth and highly qualified plywood layers for the sheets.

TP: In the workshop we used foam for prototyping, rather than wood, so instead of learning about wood as a material, we looked at parametric design and Zipshape relating to form and geometry.  From your perspective what worked and what could have been improved in the workshop?

CS: This was in many regards a surprising development for the workshop organizers, as we had planned to manufacture wooden samples. As the geometry definition relies on the teeth’s shape and the teeth are not making use of wood properties, we started our experiments with Styrofoam, which we could machine quite quickly on the campus. Wood was meant for a second phase to investigate structure, bending behaviour and the time-consuming gluing process with a vacuum bag—but we got captured by the experiments with Zipshape’s geometry and never made it to the second phase within the three days of the workshop.   However, I had never done Zipshape experiments with foam and was thrilled with the manufacturing speed and especially with the ease of bending without vacuum bag. Taking on the experience from Copenhagen, we are planning right now an exhibition design in Switzerland using foam.

TP: You´ve said Zipshape is at the interface between product design and architecture.  Could you describe the upcoming Shanghai Expo project which uses Zipshape?

CS: The DuC-Pavilion at the Expo 2010 is a part of a three-year program for German-Chinese cooperation. The pavilion is actually a bamboo construction. Zipshape was used for some curved ribbon-like elements in its interior design that could not have been realized with conventional methods. Those eight ribbons with a total length of almost 40 meters were manufactured from MDF with beech flanks and linoleum surfaces. The ribbons serve as a platform for an interactive simulation of sustainable urban planning. While taking care of their virtual city, most visitors will hardly notice the unconventional construction. However, it is the first public application of Zipshape and therefore an important test case. The project was realized in collaboration with the chair of Peter Russell at RWTH Aachen University.

All photos Christoph Schindler

Seminar 2: the ideal of the tectonic: The tectonics of the joint

Date: Spring semester 2010: 31.March. 2010

Michael Hensel – AHO Oslo
Sigurdur Omasson – DTU Copenhagen
Silvan Oesterle –  ETH Zuerich
Sebastian Gmelin – Arkitektskole Aarhus

The seminar introduces key thinking into material conceptualisation in architectural thinking. It discusses the history of the term craft and what this has come to mean through the industrialisation of building practice. It critically reflects on the understanding and creation of material and structure and situates the discussion of technological development within a historical frame work.
The seminar exemplifies this development in the frame of the wood building practice. Herein it reflects the link between the traditional understanding and practice of wood building and it’s current state, where a broad field of novel wooden materials, production and assembly techniques emerge.  A special focus is set on the critical understanding of the interdependence between material, tool and imposed information, such as design intent, assembly strategies and tectonic. A novel frame for technology and technology transfer is emerging. The seminar talks give concentrated insights in the presenters practice and inform the discursive speculation about the frame of an emerging building practice that bears the potential of closing the gap between material and making.


9.15    Presentation and Discussion of Workshop results

10.15    Introduction to the Seminar by Christoph Schindler and Martin Tamke

10.30    Lecture 1 – Michael Hensel – AHO Oslo
11.00    Lecture 2 – Sigurdur Omasson Associate Professor DTU Copenhagen
11.30    Discussion

12.00    Lunch

13.00    Lecture 3 – Silvan Oesterle – Professorship Gramazio & Kohler, Architecture and         Digital Fabrication, ETH Zuerich
13.30     Lecture 4 – Sebastian Gmelin – Arkitektskole Aarhus  m.any, MAS project at eth Zuerich
14.00    Discussion

14.30    Coffee

15.00    Sum up and Discussion of further network activities (i.e. date of next workshop in august)

16.00    End of Seminar