Mokuzai Kaikan: Japanese timber revisited in Tokyo

Timber construction is an art, specially when deployed by the crafted Japanese carpenters of past centuries. Apparently, even if there are still artisans who keep some of its secrets today, the subtleties of Japanese timber joinery have disappeared from Nippon architecture. Kengo Kuma is the only name that comes to our mind if we think timber and contemporary Japanese architecture. But projects as interesting as One Omotesando in Tokyo (see the images below), with its main facade protected by thin vertical fins of larch wood, use timber more like a skin care rather than as a construction material. Timber here is the wrapping, not the real thing. The same technique was used by Kengo Kuma at the Nagasaki Art Museum, this time with louvres made of stone instead of timber - although it's difficult to find out from a distance!

Has timber joinery disappeared from contemporary Japanese architecture then? Well, not yet.

'The Art of Japanese Joinery' by Kiyosi Seike is the best introductory book into Nippon wood joinery as an artistic craft. The book starts with the history and philosophy of Japanese architecture as it relates to joinery, then follow many pages of great black and white pictures of wood joints. Only 48 types of joint are presented, selected from among the several hundred known and used today. Joints range from the simple scarf joint to the insanely complex ones. Some of them are truly puzzle-like in construction. The text continues with a chapter on the functions of Japanese joinery, then a chapter on Tsugite or splicing joints and finally Shiguchi or connecting joints, both of which have drawings showing the construction of the joints with hidden lines for further clarification (or obfuscation?)

Wikipedia will help us enter the world of splice joints or Tsugite. A splice joint, in Japan and elsewere, is a method of joining two members end to end in woodworking. The splice joint is used when the timber pieces being joined are shorter than the length required by the construction. Splice joints are stronger than (unreinforced) butt joints and have the potential to be stronger than a scarf joint. The most common form of splice joint is the half lap splice (see below left), used in building construction to join shorter lengths of timber into longer beams. Connection has to be achieved using glue, nails or screws. The beavel lap splice gains profit from geometry, with its dovetailed shape. Things start to get more complex with the tabled splice joint, where glue or nails aren't working in shear any longer.

A variation of the latter is the wedged tabled splice joint (see below right), where two interlocking wedges close the gap and secure the connection of the two timber pieces with each other. Here we don't need to nail or glue the joint and even better, we can disassemble the two pieces if and when needed. We got it: this is Tsugite, we have just entered the Japanese timber joint world.

Some may think this is nice stuff for DIY aficionados with time to spend on the weekends. Is there a way to apply the intricacies of joinery shown in Kiyosi Seike's book to real life construction? Where will we find the carpenters and the time for doing this - not mentioning the money to pay for it? There is a way, and it's called CNC woodworking machinery. State of the art woodshops today are employing computer numerically controlled pin routers to cut wood, and are using vacuum holding fixtures and autoclave-like devices for joining solid wood. They can even glue wood with glues that harden only in the presence of radio waves. A longish Youtube video - but worth seeing for a couple of minutes - shows one of these Japanese timber frame joinery machines at work.

So timber in Japan is an abundant material, there is a milennary joinery knowledge and there exists CNC machinery to cut and join timber for construction purposes. One would expect to see hundreds of buildings in Japan using timber as structural or as finishing material, both for outside and inside applications. That is not the case though, and this was the reason why Mokuzai Kaikan (the Tokyo Lumber Wholesalers Association) decided to promote timber as the material of choice at their new headquarters in Tokyo.

The Mokuzai Kaikan office project was commisioned to a big architectural firm in Japan, Nikken Sekkei. The numbers in this studio are impressive: founded in 1900, the firm was 29 strong by 1904 when they finished their first big project, the prefectural library in Osaka with a neo-classical style. Now they are almost 2,900 between architects and engineers, and their projects extend all along the Pacific rim. The Wholesalers Association project was directed by Tomohiko Yamanashi, principal at the architectural design department, in coordination with Takeyuki Katsuya. Yamanashi and Katsuya embraced the idea of promoting the use of timber as requested by their client, and wood became the leit motiv of the project.

This is the outline of the Mokuzai Kaikan project as it appears at Nikken Sekkei webpage:
This project involved the relocation of the offices of the Association of Wood Wholesalers in Tokyo. It serves as a showcase to demonstrate the possibilities of wood as an urban construction material. Engawa, or Japanese terraces, allow a natural breeze to enter while shutting out strong sunlight for a comfortable indoor environment. Lumber were integrated into the building's structure, and architectural exposed concrete was cast in cedar formwork. Since the building uses a large amount of wood, great attention was given to fire safety measures. The design focused on creating spatial continuity with the use of layering and natural light.

The building also revives and adapts another of Japan’s architectural traditions through the use of the Engawa (see night image to the left), a terrace space prevalent in traditional homes. In accordance with earthquake regulations, the 7,582m² seven-storey building employs reinforced concrete for its structural frame. Beyond this, timber was specified wherever possible. The architects paid close attention to detail, fitting the main concrete frame with the secondary timber elements. Concrete was cast in cedar formwork, maintaining the scale and grain of the timber (see below left). 

In terms of the timber elements themselves, everything that can be seen is formed in 105 x 105 mm sections of Japanese cypress; a standard off-the-shelf product. These sections are used in composite panels to create the distinctive cubic Engawas, but they also form the remarkable longitudinal beams that span the full length of the 25m rooftop assembly hall. Here is where the art of Tsugite reappears, and in a way that mixes tradition with modern requirements. 

Each cypress element is just 0.105m high x 4.0m long, and it had to be connected to those above, below and beside to conform a 1.6m high x 25m long beam. The connection system owes something to the traditional joinery - you can see timber wedges in vertical, combined with wooden oak plugs that connect every two pieces in horizontal. Tabled spliced joints can be seen both at the top and the side of every element. But there are also stainless steel rods - or should we say long bolts? - that keep all timber pieces working together as a 1.6m high beam. In order to avoid the concentrated tension around bolts to fracture the wood under extreme stress, cilindrical aluminium rings have been added around each passing bolt. 

The exploded detail can be seen here below, extracted from an article about the Mokuzai Kaikan office appeared on The Architectural Review. The following images, the assembly room plan at the upper storey, the vertical section and the Engawa 3D drawing have also been taken from the same article.

The Mokuzai Kaikan office was built between Nov 2007 and June 2009. The project received a Special Jury Award at the 3rd annual MIPIM Asia Awards 2009 held in Hong Kong. I invite you to see some stunning images of the interior here (go to the bottom of the page)

How will the lumber perform on the outside say 10 years from now? This remains to be seen. I have not found any information about the surface treatment these cypress logs have undergone - if someone out there knows more, please shout. One final critique has to do with the side facade, visible at the images from the link above. That facade doesn't seem to be the best part of the building. But all in all, Mokuzai Kaikan is a great example of architecture well delivered at all levels, from the concept phase to the 1:1 details, from programme to materials selection. The front facade is an example of intermediate, filtering space, one that moves forward from the typical flat, barrier-like glazed facades we are used to these days. 

There is another project I find vaguely similar to this one, located in a very different place place and context: Louis Kahn's Salk Institute at La Jolla in California (1959-1966). Sorry, but I can't find a better way to finish this post than by paying a little homage to the old master.

The Ledge at the Sears Tower in Chicago: glass is the limit

There can't be more stories written about the design, engineering and construction of such a small part of a building as for The Ledge, the latest attraction at the tallest observation point in Chicago. The Ledge is really small in number: just four glass boxes, all the same. And it's small in size too: each box is 1.3 x 3.2m in plan, and close to 3.6m high. But there's one detail that makes all the difference: these four fully glazed boxes are cantilevered from the 103rd floor of the Sears Tower, at 413m above the ground. Glass Google pages and technical papers at recent facade conferences are filled with stories about it.

Why so much fuss about glass boxes? Well, the image says it all.

The Sears tower (since 2009 sadly renamed as Willis tower) remains the tallest building in the Western hemisphere. Completed in 1973, its 442m (without counting the spires) are still today an impressive height for a building. The Sears tower observation deck, called the Skydeck, is located on the 103rd floor of the tower, and its view 412 m above ground is one of the tourist attractions in Chicago. Visitors, up to now, could experience how the building sways on a windy day. Now they can also feel a different sensation: that of sheer vertigo.

In January 2009 the tower owners began a major renovation of the Skydeck including the installation of four glass balconies, extending approximately 1,2m over the west facade from the 103rd floor. The all-glass boxes allow visitors to look through the floor to the street 412m below. The Ledge opened to the public on July 2009.

The Skydeck renovation project was awarded to Skidmore Owens and Merrill (SOM), the architects who designed the tower. The picture above shows a realistic image of their intention: glass all around, no steel structure at all if possible. “The Sears Tower set architectural and engineering standards when it was first built and now we are able to carefully craft new elements that expand the capabilities of the original design while retaining its integrity,” said Ross Wimer, design partner with SOM and one of the fathers of The Ledge idea. A 360º view of one of the glass boxes (this is not a render) is the nearest you can be to the real sensation of stepping in (or out?) there.

Architects must be praised for the idea, but engineers - glass specialists - were needed for its detailed design and realization. The building owners contracted Halcrow Yolles in Toronto as engineers for the observation boxes, and gave them the responsibility of fully design and detail the glass and steel components. Halcrow's senior principal and structural glass engineer was at the time John Kooymans, one of the few who can say 'The Ledge is my baby'. Around the end of 2009 John moved from Halcrow to another Canadian engineering team, MTE Consultants in Ontario (the name means More Than Engineering). If you visit Halcrow and MTE webpages you will find how both companies claim having authored the engineering design of The Ledge - and both are right!

There are two articles describing the design of the glass boxes both written by John Kooymans, one as part of Halcrow Yolles and the other one written this year as member of MTE. According to the first paper, publised in Glass Performance Days 2009, the challenges to solve were many:

• All glass elements had to be brought up to site using the internal elevators, which limited the size of the box elements.
• The observation boxes had to be moveable. This was required to allow the facade maintenance equipment to operate along the facade without interruptions. But, even more difficult, it was decided that the glass boxes were to be retracted an additional 1200mm into the floor space, so that glass maintenance and cleaning could be done from inside the building.
• There was tenant at the floor below the Skydeck, so all the glass box loads had to be hanged and framed from the ceiling above, in order to avoid interferences.
• Design loads and movements to be imposed onto the glass structure would be very high from the calculation stage. The frame and glass box had to be stiff enough to allow the movement of the box without creating large deformations or stresses at the glass connections.
• The details around the glass box had to include weather seals in both the extended and retracted position, allowing for the movement of the tower, and for the seal to be temporarily broken while the assembly moved from one position to the next.
• The architectural intention was to obtain maximum transparency, including the floor and the roof, minimizing the visible structural steel elements.
• Structural redundancy (safety) and protection of the glass were requested. But, to speed the use of the balconies, visitors would not be forced to protect their shoes when entering. Safety issues excluded the option of double glass, and laminated glass was the solution.
• Outer temperatures should not create a risk of condensation on the glass, or worse allow the formation of ice shards outside the boxes, falling onto the walkway behind. So, the design had to introduce some kind of heating system as well.

The second paper by John Kooymans is the one that originated this post: I read it at the proceedings of Engineering Transparency, a conference held in Glasstec Düsseldorf last 29 and 30 September. Carlos Prada and María Meizoso, two colleagues from Arup Facades Madrid, attended the conferences and brought the news back to us. You can have access to the paper here, through the MTE webpage.

Each box face can withstand design wind pressures of 4.6 kPa, and the roof and floor wind pressures of 6.0 kPa. At the floor there is an additional live load of 4.8 kPa due to its intended occupancy. As Kooymans puts it, it can essentially hold more people than it can fit.

The corners and intermediate joints where the different wall panels come into contact with each other are simply stitched together with stainless steel angles and through bolts. The floor is stitched to the glass walls creating small local opaque connections that allow for the transfer of external loads into the hanging glass panels and subsequently, into the steel cantilevered frame.

From John's Glasstec paper:
The glass had to be designed with enough redundancy to ensure that any accidental breakage would not result in a total collapse of the system. For this reason, three layers of glass were selected for all the elements. The structure was designed so that only two layers of glass were required to resist the design loads, and only one layer of glass would be able to support the self weight of the structure. In addition to this design decision, the glass floor was constructed using an ionoplastic interlayer (SentryGlas Plus) captured by the through bolts in the floor which would ensure the stiffness of the tempered floor panel would remain intact in the remote possibility that all three structural glass lites failed.

In the end, the glass box elements are all created with three layers of 12 mm tempered low iron, heat-soaked glass. The requirement of heat-soaking helped eliminate the potential for spontaneous breakage due to nickel-sulphide inclusions.

If the engineers of The Ledge were Canadians, the facade contractors are pure Chicagoans, and making part of the city building history. MTH Industries, located in Hillside, Illinois, started building glass facades in Chicago back in 1886. Upon first hearing about the project, Ludek Cerny, vice-president of glazing at MTH Industries, thought it was pretty unusual. Because of that, MTH wound up taking on a design-assist role.



Load tests done in-house by the contractor (see images above) involved loading a glass lite that was half the size of the actual floor of the bays to 2½ times the required code load for a 24-hour duration. “The test was later repeated with fracturing one of the lites with the actual design load,” Cerny says. That wasn’t enough for this team: “Out of curiosity,” Cerny says, “we actually broke more lites and realized that you could still stand on the glass floor with all of the lites broken.” Miracles of ionoplastic interlayers.

In addition to avoid damage from breakage, the design includes ways of protecting the bays from daily wear as well. There is an anti-graffiti film on the inside of the vertical glass units. The laminated floor has a 6mm sacrificial layer of fully tempered, heat-soaked glass on top that can be removed or replaced if it gets scratched, cracked or damaged. The stainless steel fasteners that support the glass panes to each other are bespoke and have been custom-machined by MTH.

The motorized system that projects and retracts the boxes from the building utilizes steel LinearBeam mechanical linear actuator systems. The systems operate with a rigid chain technology. A rigid chain is a mechanical actuator that is flexible in one direction and forms a steel beam in the other direction. The contractor worked with the supplier to design the locking pins and the control systems that secure the bays.

Because of the movement, the perimeters of the bays are lined with inflatable seals. When the bay is in the viewing (outside) or in the maintenance (flush) position, the seals inflate to create a secure air and water lock for the building.

Vertical movement - that is, transporting the box material up to the 103rd floor - proved to be one of the bigger challenges for the contractor. The installers moved the glass units and the 5.5m suspension beams up on the top of elevator cars. To ease the material handling, MTH ended up creating custom tools to help hoist and carry. “It was all a conglomeration of things that already existed modified to work under these conditions in the space allowed,” Cerny adds.

The laminated glass units forming the walls and the roof have three tempered 12mm lites of PPG's Starphire low-iron glass. The walls and roof are laminated with clear PVB, while the floors are laminated with 1,52mm DuPont’s SentryGlas Plus interlayers. The glass fabricator was Prelco of Montreal. Prelco delivered its last panel in April 2009, six months after the company began fabrication and two months before the end of the installation on site.

This article from The New York Times has a very interesting short video abour the building of the Ledge. Scroll down and you'll find it on the right. Not to be missed!

The final image is my personal homage to the vision of all the people involved in getting these glass boxes real: from the owner to the architect, engineer, contractor and every supplier. Compare this image with the first render, drawn by the architects in 2008. One year later, the built thing is astonishingly similar to the design intention. In fact, it is even better: by selecting low iron glass, the green aspect of the standard glass used in the first image has disappeared.

Could such a small job have been done better? 

Arup and facade engineering

This is my post nº 22. This blog has had more than 1,100 visits up to now, in less than two months after I started writing it. Not bad!

It's time to get a bit more personal, and tell you, dear reader, what I do for a living. I am a facade engineer, or a facade specialist, or a facade consultant - it's all the same more or less. I work in Arup, a big engineering firm based in London but with offices in all continents. My desk is in Madrid, but my projects are - and have been - in many places around the world. That's of course a fantastic experience.

There are almost 300 facade engineers in Arup if we count all offices. The discipline started in London around 1985, and I think we are now the biggest facade consultant in the world. Our offices with facade dedicated teams are located in the UK, Ireland, Germany, Italy, Netherlands, Spain, Denmark, Dubai, South Africa, Australia, Singapore, China, Japan and the US. The facades team in Madrid started in 2004, and we are 10 people between architects and engineers. We have taken part in projects as interesting as the image below: the Bridge Pavillion in Zaragoza with Zaha Hadid. We have been lucky enough to work with well-known architects as Rogers, Foster, Zaha, Piano, Chipperfield, Arup Associates, or Spanish firms as Rafael de La-Hoz, Lamela, Nieto Sobejano, DL+A, MBM, Ferrater, Vidal or Cruz y Ortiz. We also work for developers, usually helping architects to develop the trickiest parts of facade designs, acting as site specialists during construction, conducting failure investigations or leading the facade refurbishment of existing buildings. Sometimes we also do systems research and development for facade contractors.

I love being a facade engineer because of the combination of skills it requires. As old Vitruvius used to say, it has a bit of firmitas (resistance, durability), a bit of utilitas (performance, confort, modularity) and a bit of venustas (proportion, colour, texture, beauty). Wasn't it the definition of architecture? Precisely. The question is that these days, because of the complexity of the building profession, one cannot be an architect and understand everything of a building in a holistic way. There are two options: either you remain a generalist and rely on teamwork for the project to achieve a global view, or  you become a specialist in one specific area of knowledge, as facades. In this case you can still have a complete understanding of your branch, combined with a minimum amount of details of the surrounding areas. Engineers have always tended to subdivide their bodies of knowledge; architects have up to now resisted such a temptation. As an architect, I think we were wrong. Someone can argue that my work is not that of an architect, but of a building engineer. I take the point: being a building engineer is a way of being an architect, just as being a civil or an electrical engineer are ways of being an engineer.

Facades are a great topic because they involve almost everything an architect did in the good old days (except plan distributions obviously), so you still feel you are in command, and your area of expertise is still very wide. In fact, I now consider myself a facade generalist rather than a facade specialist - it's becoming impossible to be a real specialist in such a wide discipline as ours!

British people love belonging to clubs. Today's equivalent to the classic clubs are professional fellowships, where Brits feel like at home with their peers. Times have changed for good, and these professional societies do welcome women and foreigners as members. Our club is of course the Society of Façade Engineering. And what is the definition of façade engineering to this honourable Society? There it goes:


“Façade engineering is the art of resolving aesthetic, environmental and structural issues to achieve the enclosure of habitable space.” 

You see? There's Vitruvius again, and I swear I wasn't aware of this definition until now. Sounds good to me (and rather Brit as well). The Chairman of the Society is my friend and Arup colleague Mikkel Kragh. Mikkel is Danish as Ove Arup, our founding father, which makes him a sort of square Arupian. He is now living in Milan and leading a growing Arup Facade team there, apart from chairing the Society and doing several research and academic activities. Mikkel has written an article on the role and challenges of façade engineering, "Façade engineering and the design teams of the future". He points out that our trade is not only a business of architects and engineers, but also one for facade contractors:

The façade engineering discipline is embedded in various aspects of the work of Architects, Engineers, and Specialist Trade Contractors and we will see an increasing need for seamless collaboration and delivery of integrated systems as opposed to elements and components. We have witnessed a recent trend of design teams going from multidisciplinary to interdisciplinary, with disciplines interacting and working closer together.

This is a really serious point: integrated systems as opposed to elements and components. The integration of different functions, not just the co-existence of independent systems as part of one skin, seems to be the strategy for the future of façade technology and design. I believe this is the way too. There is more on this matter in a paper from Tillman Klein, who leads the Façade research group at the Faculty of Architecture in Delft, "Evolution or revolution of systems in façade technology". This article is part of the book "The future envelope 1 - a multidisciplinary approach", edited by Ulrich Knaach and Tillman Klein.

But design and construction are just parts of the whole story of facades. New materials, the quest for optimum energy performance or the support for energy generation systems are requirements that meet with predominantly conventional crafts. Our role as façade engineers in every project is to lead a conversation between these diversely interested disciplines into a converging interdisciplinary team, a team that will not put one interest too much above the others. It sounds like a complex task, but the final result should be simple: as the good movies or buildings we remember long after having seen them. 

Should we façade engineers expect to receive prices or accolades? Nope. By the time for the party and the distribution of medals after the opening of a building we are already hands on with the next project, where the action - and the learning - is. Our medal is to have taken an active role in designing, fabricating and building facades that stand the passing of time, perform well and mean something to people. Our medal is to have avoided failure to happen more than once. Our medal is to contribute to the delivery of better buildings that become sounding pieces of better cities. 

Isn't it a great career?