The Super Roof Turns 50

On August 26, 1972, athlete Günter Zahn lighted the Olympic Flame, officially opening the Olympic Games in Munich. It would be an event remembered for a devastating terrorist attack, memorable champions, remarkable visual identity, and a ground-breaking architecture symbolized by an undulating roof that covered multiple venues and their surrounding landscape. The roof was designed under the direction of Günther Behnisch, Frei Otto, Fritz Auer, Fritz Leonhardt, Jörg Schlaich, Rudolf Bergermann, and others. The cable net construction in Munich became the progenitor of all lightweight structures and its enduring image continues five decades after its construction.

This past October, Iker Gil, architect and founder of MAS Context, and Bill Baker, structural engineer and consulting partner at SOM, met with Rudolf Bergermann and Knut Göppert to talk about the project. Cofounder in 1980 of schlaich bergermann partner (sbp) and currently a consulting engineer, Bergermann discussed his education, early career, and behind-the-scenes of this remarkable project. Partner at sbp since 1998, Göppert explained sbp´s current involvement in helping to preserve these important lightweight structures.

IG: Mr. Bergermann, you were born in 1941 in Düsseldorf. Tell us a little bit about growing up in the city?

RB: I grew up on a farm outside Düsseldorf. I went to school in Düsseldorf up to the age of twenty, and then came to Stuttgart to study civil engineering.

IG: Do you have any other siblings?

RB: No.

IG: How did you make the decision to study in Stuttgart?

RB: The reason was to be far away from home. [laughs] Stuttgart was known in those days because [Fritz] Leonhardt was a professor at the university, and that was the reason why I came here.

IG: Were there any professors that were inspiring to your work?

RB: Sure, there are many. But, of course, the most impressive was Professor Leonhardt with his lectures at the university. He was my professor for the advanced studies from the fifth semester onwards.

IG: I believe you graduated in 1966.

RB: Yes.

IG: What were your next steps after graduation?

RB: I joined a construction company called Züblin, here in Stuttgart. I was there for a short period. After finishing my studies, I tried to get into Leonhardt’s office, but it wasn’t possible. At that time, Leonhardt’s office had no more than thirty engineers. After two years, I got the chance to change jobs and I started working at Leonhardt’s office early in 1968.

IG: Do you remember how many people worked in the office at that time and the early projects that you participated in the office?

RB: On January 1, 1968, we were eight new engineers entering the office, so it went from thirty-seven to forty-five. I didn’t start immediately with the Olympic structures. This came a little bit later. I first designed pedestrian bridges and civil engineering structures.

IG: Were those projects located mostly in Stuttgart?

RB: They were all around Germany. At that time, nearly all the work in Leonhardt’s office was here in Germany.

IG: Were there consistent teams or collaborators in the office?

RB: No, it changed from project to project.

IG: Let’s go back a couple of years. In 1966, the year you graduated, Munich won the bid to host the Olympics. On February 1, 1967, the City of Munich announced an architecture competition for the planning of Oberwiesenfeld, a former airfield and dumping site among other things, as well as the design and construction of the stadium and other venues. And I believe one of the requirements of the competition was to have a transparent roof as the German Olympic Center for Radio and Television wanted to minimize shadow contrasts for TV cameras, is that correct?

RB: No. If I remember correctly, during the competition there was no requirement for a translucent roof. The winning architect, Behnisch & Partner, used a translucent material for his model [nylon], but just to build the shape. He used this material for the complicated tent roof structure. The requirement for transparency only came during the first design phase.

IG: I see. So, it was not part of the competition.

RB: No.

IG: Do you know any of the other offices that entered the competition and if the competition drew international entries?

RB: There were a few entries from other countries, but I do not remember their names. Here in Germany, there were many famous architects, such as Gerkan Marg [now gmp] for example. Heinle, Wischer and Partner also entered this competition. During the competition, we [Leonhardt and Andrä] partnered with Heinle Wischer, not with Behnisch. The engineer for Behnisch was Heinz Isler from Switzerland.

IG: It seems that there was quite a lot of controversy around the winning proposal, and apparently it was almost eliminated from the beginning. But the Mayor of Munich, Hans-Jochen Vogel, and the chair of the jury, Egon Eiermann, pulled that project back into the competition. Do you remember a little bit of that process?

RB: I remember that many thought that this roof could not be built, and that is why it was not the favorite of the jury. But it was pulled back in. Heinle Wischer’s design featured individual shell structures and the proposal finished third place in the competition. I think that, at one point in this discussion, when it tended towards the Behnisch design, [Erwin] Heinle himself, the architect of our team, proposed that we should change to Behnisch’s team. At the time, it was already clear that Isler, who had advised during the competition, had no capacity to deal with these structures and he didn’t have a big office. Professor Leonhardt from our team was asked by the client about the feasibility of the design and he publicly declared that the roof was buildable. He was a very famous engineer at that time, and it gave confidence to the client to proceed. Perhaps this was also another argument to change to the Behnisch team.

IG: Let’s talk about the different team members. The architects Behnisch & Partner presented a model that was based on the roof that Frei Otto proposed for Germany’s pavilion at the Expo 67 in Montreal. But Otto was not part of the team during the competition. However, Isler was part of the team. I would like to know how Otto and Leonhardt got involved in the project and what the role of each person in the team was.

RB: As I said, Isler was in Behnisch’s team during the competition and that is the reason why he had the right to be part of the winning team later on. He did not like to be the designer of the roof, so he got the job of all the concrete structures underneath the roof. That was nearly more than he could handle. During the further development of the design, we changed from the Heinle Wischer team to Behnisch, and we became part of their team, with Leonhardt as our most famous partner. Jörg Schlaich was also part of the team from the beginning and gave a lot of good advice. Frei Otto was not part of the team even during the first month of the preliminary design when trying to find the shape for the roof. But the client, the Olympia Baugesellschaft, insisted that Behnisch takes Frei Otto into his team. Frei Otto was known as an expert in this field, so they asked that he should be part of the architect’s team.

But as you know, Frei Otto was not only an architect but also an experienced engineer, and this was one of the reasons he joined the team. He also had a considerable influence on the shape of the roofs. He was the one who brought some regularity into, for example, the stadium roof. He cut the roof into nine separate sections. He also had influence on the position of the pylons supporting the roof. For example, he insisted that the pylons should not be straight but should be inclined instead. The simple reason was that, based on his experience in Montreal, he knew we would never get such an exact geometry. A small deviation looks bad, a big deviation looks planned. He had considerable influence on the shape of all the roofs, including the tent structure for the swimming arena.

But Frei Otto was a very difficult person to work with. We as engineers had serious problems with him. The architects also had problems with him because he insisted on being the leader and on deciding everything. This led to the point that, after nearly half a year of cooperation with Frei Otto, the client decided to dismiss him.

IG: Was it the client who made that decision?

RB: Yes, but of course on the advice of the architect.

IG: Was it based on a specific issue or just the working process with him that he just made untenable?

RB: The overall working process.

IG: There is a famous photograph that features the different team members during a work meeting. We are wondering about the context of that photograph. When and where did it take place?

RB: This was during the six months of Frei Otto’s cooperation. That is why he is in the photo. We were looking at the models built by the architects and discussing if the shape, from our point of view, would be feasible or not. It was a working session, and as you see in the picture, there was still a discussion going on. If I remember correctly, it was one of the architects working on the models who took this photograph in the architect’s office.

IG: Did the meetings usually take place in the architect’s office?

RB: Yes, they took place in the architect’s office. But all our offices were close to each other as we were all in one building in Munich. It was specially built for this purpose. The Olympia Baugesellschaft, the client, had their office in one wing of this building, and the architect’s office and our office were close to each other. We saw each other several times a day and we had a close cooperation. Most of the discussions took place in the architect’s office because the model was there.

IG: Who had the ultimate responsibility of the design if there was a disagreement?

RB: The architect. But I do not remember any decision which had to be taken where the architect decided against Jörg Schlaich’s advice. It was a very close cooperation and a very close development on the roof structure.

IG: Had the different members of the team worked together in other projects before this one?

RB: No.

IG: This was the first project?

RB: Leonhardt had never worked with the architect before. At that time, we already had completed many projects, but we had no experience working with Günter Behnisch.

IG: I’d like to talk about the tools and the processes used during the design of the project. What type of physical models were built, what were they used for, and how were the different elements tested?

RB: At the beginning, the approach to designing the roof was to build models, load them, and measure the deformations and forces in these models. We also started with models at Frei Otto’s Institute [for Lightweight Structures] here in Stuttgart. It was clear for Professor Leonhardt and our team that we needed to build models. At that time, nobody thought of computer programs as they were not available.

We started with models at scale 1/75 in Frei Otto’s Institute. We built complete models. We lost a lot of time loading these roofs and taking measurements as well as taking measurements for the geometry and for the lengths of the cables. We found out that this would not work as the geometry resulting from such a model was too inaccurate. The accuracy we had to assume was bigger than the elastic elongation of the cables, which meant that maybe one cable in the net had no force at all, it didn’t participate in the load carrying of the structure.

After some time, we came to the conclusion that we had to find a way to use computer programs for the calculations and determine the geometry by computer. We started looking for somebody who could help us, and we finally found someone at the University of Stuttgart. He was not within the Civil Engineering department, but in the department of Aeronautics. Professor [John] Argyris was using some special programs for the helicopter rotators. ASKA [automatic system for kinematic analysis] was the name of this program. He confirmed that he could develop this program into a program for designing cable nets. We had a very short time so we had no choice but to develop this program, which took about four to five months until we could finally start doing analysis of those subjects.

IG: Was Professor Linkwitz involved at that time?

RB: No, not at that time.

IG: Okay.

RB: We started with Professor Argyris at his institute [Institute of Statics and Dynamics] and we finally succeeded in having a model for designing the sports arena as one unit, as one model. For the stadium, for example, the nine individual roof segments could be designed only individually with no equilibrium at the contact. But that is all we could do there. For the tent structure for the swimming arena, there was no time left to use a computer program. But the swimming arena has the simplest geometry, it is just like a tent.

The results of those models were used. At the outer edges of the tent, you pull as long as you have the real force in the structure, leading to some packing plates under the cable sockets of more than nearly a centimeter. That demonstrated how inaccurate the geometry from the Frei Otto model was. For the other structures, it would have led to disaster. However, for this tent it was possible to pull and elongate the cables at the outer edges.

IG: The computer software that was used was customized specifically for your project. Did the architects and the engineers both use that computer software? How was the exchange back and forth between engineers and architects?

RB: The architects had no contact with the computer calculations we did. The overall geometry, which the architect liked, was not changed by these computer calculations. The changes were only a few centimeters, which did not affect the overall shape that the architect was interested in. But this was mainly an exchange between the institute of Professor Argyris and the engineers.

BB: Was the Force Density Method by Professor Linkwitz used?

RB: No.

BB: Did that come later?

RB: Professor Linkwitz got involved mainly to supervise the shop drawings that had to be done for this project. The drawings created from the model were not very accurate. The outer strip of the roof, where we have the connection to the edge cables and where the total length of the cable with these elements was determined, was very inaccurate. Their drawings, scale 1/20, were produced just by hand. This was done by a team that was supervised by Professor Linkwitz. At that time, Linkwitz and Hans-Jörg Schek started their own program to design cable nets, but it was not yet used for Munich.

IG: Was the team doing the set of drawings part of the university, or was it part of a professional office?

RB: The drawings were produced in the office of Professor Linkwitz at the university. He had no private office. The work on these 1/20 scale drawings was done by twenty or thirty people. These were no prisoners as it has sometimes been suggested. There were no prisoners involved. These were soldiers from the Bundeswehr, the armed forces of Germany.

IG: Do you remember how long it took to do all the drawings or what the drawing set would look like for a project of this scale and complexity?

RB: It didn’t take long because we were under extreme time pressure. As I said earlier, we lost a lot of time finding a way to do the computer analysis. It was a question of three to four months, not more.

IG: How was the transition then to the construction phase? Was there a contractor hired by the organization? Was there a contractor that you had all experience working with?

RB: There was not a contractor large enough who could handle such a big project and who had experience with cable structures, because there hadn’t been big cable structures before. The client, the Olympia Baugesellschaft, organized a competition and they received only two entries, one from Germany and one from Austria. The cost was extremely high, at least compared to the cost estimates done by the expert, and the offers included many risks. They said that we could simplify it or that they could build it as designed, but that it would cost much more and take much longer. They didn’t guarantee that they could build this roof within the schedule. There were a lot of discussions with them without any result. Finally, the client decided to put the two companies together. That way, at least that would increase their capacity and the risk of not being on schedule would be reduced. And in terms of cost, this team signed a cost-plus contract. They got paid for all their work with a 30% profit. They did a large number of works on site which, in our opinion at that time, was not necessary. But they were paid everything. And, of course, the cost went up. The final cost for the roof was 200 million Deutsche Marks [102 million euro]. The initial estimate was 20 million, but it was known to everybody that this could not be a serious estimate.

IG: Was there any special tool or process that had to be developed to build the project, the same way that to make the calculations you had to use a software that was coming from another industry?

RB: No. The production of cables for the net itself was a process which was known to some specific companies. Putting clamps on those cables in a controlled process was also known to the company who finally produced net cables for the edge cables. These were locked coil ropes, which were known at the time, at least here in Germany. We had difficulty to find a company which had sufficient capacity to build all those cables and, therefore, we had to go to France for the edge cables. For the main cables from the tower tops to the foundations, we used parallel strands that were bundled, which was also a rather new development but known already at that time. It was a Swiss company who built these bundles. For the erection itself, the companies obviously had to develop new concepts. In our opinion, they developed quite conservative concepts. But I can understand that they wanted to go the safe route. For example, for building the stadium roof, we proposed to build the main cables on the ground and then to pull it up like we would do for many other structures later on. This was deemed too unsafe for the company. Perhaps, the cost-plus contract led to the decision to build a huge scaffolding up to 40-meter high to place the cable on a scaffolding.

KG: You also had the cast steel junction points.

RB: The geometry of the nodes in this cable net was very complicated. Very soon, we came to the conclusion that it could not be made as a welded structure from individual plates. It was too complicated. We had to go for cast steel, which in civil engineering structures was very unknown at that time. Of course, the material itself was known for machines used in the steel industry. We had to deal with this material, do a lot of tests to get to know what we could do, and how we could cast this material into the special shapes.

The pylons were standard tube elements. The size was bigger than the companies were used to, but there was no specialty required. They were fabricated in segments in the shop, brought to the site, and assembled there. The joints between these elements had the gaps open. We had a uniform pressure compression in these joints. There was no special screwed connection necessary.

To build the 80-meter masts, the company preferred to build them vertically, but the final position was inclined. We had ball bearings at the base so that we could quickly rotate the masts to the inclined position. After that, the steel ball bearings were cast in concrete.

IG: Regarding the techniques that you are mentioning, were they already in use in the construction of bridges and other infrastructural projects?

RB: No. The construction industry already used cables but in a different way. They used cables for cable-stayed bridges, but those cables were straight, with a socket at each end. They were much easier to install. Here we had a very difficult cable geometry. We had a lot of clamps where the cables had to be connected. The clamps on those cables, for example, were a development in themselves. Because clamping them with the clamps being used until then would have led to a much bigger and very bulky solution. We did a lot of tests and we found that using a much thinner clamp, acting not in bending but which acts as a tension bend, created a lot more friction to the cable. That led to this very small clamp solution. We added a lot of different sizes or forces in the edge cables and decided to simplify this by using a spiral cable with an 81-millimeter diameter. This was a type of standardization for the details that we designed.

IG: Let’s talk about the roof and the acrylic glass. How was the material selected?

RB: The requirement for a translucent roof came after half a year of our design. We mainly dealt with the structure, not with the cladding. The television team came and asked for a transparent roof to avoid shadows on the pitch. The decision for the cladding was very easy because there was only one company which offered this type of cladding. The company [PLEXIGLAS] guaranteed a transparent sheet of significant strength so they could build a larger area of these individual segments. We had a cable net with a 75-centimeter module. Of course, this was too small for the sheets, and it would have meant an enormous number of joints, something the architect was trying to minimize. We finally came to the solution of using three-by-three-meter pieces for the cladding. I think the estimated lifetime was twenty-five years, and the cladding had to be replaced after twenty-five years exactly. The translucency was excellent, it was really like glass.

The geometry of the individual panels on the roof could not be predetermined. The geometry for these pieces was determined directly on the roof. The pieces were not regular, they had a certain shape. Each three-by-three-meter element was taken up to the roof, placed in its final location, marked, and then brought down to be cut to the exact geometry. Then, it was brought back up again and supported on the rubber cushions every 75 centimeters.

IG: Did that grid change? It is our understanding that the 75 centimeters of the cable net geometry had been 50 centimeters before.

RB: Yes. This was a very hard discussion and was one of the reasons why Frei Otto finally left the team. Frei Otto fought for 50 centimeters, because he said that, during the placement of the cladding, people had to walk on the cable net and the danger that they could fall was very high. With 50 centimeters, nobody can fall. The person can spread the arms and there is no danger of falling. But the number of joints in the cable net, of course, would have been so big and the cost impact would have been so high that we fought for a bigger size of the net geometry. The final compromise was then 75 centimeters.

IG: Were there any injuries during construction?

RB: No, there were no injuries during the original construction. There was one during the cladding replacement done after twenty-five years. Just before replacement, the individual plastic sheets were so brittle that you could not touch them anymore. You had to move stepping on the joints. There was one failure and one worker fell through and died. This was later and there was no accident during the initial construction.

IG: Can you share if there was any collaboration with the University of Stuttgart? Was there any other type of knowledge that was transferred between the experience of designing and building the project to what was happening in the university?

RB: That was no contact with the university, neither the University of Stuttgart nor the University of Munich. The University of Munich only interfered at the very beginning of the project with the proposal to build this roof not as a cable net but as a timber structure. This was supported by the Bavarian timber industry who offered to build the roof at their own cost just to get involved in this project. But the Olympia Baugesellschaft declined. At that time, it was already known that it had to be a transparent roof, and with timber it was not possible. But this was the only interference from the universities that I remember.

KG: But there were many professors involved as experts for corrosion, building physics, and so on. They had smaller contributions as part of the bigger team. Most of them were from the University of Stuttgart.

RB: Yes. Also, in Germany, every project has to be checked by an independent proof engineer. Even for a small house, it is still mandatory today. We also had a proof engineer for this structure, and our proof engineer was a professor from the University of Munich. This was a very good cooperation, although he also had no experience. He was a professor at the university for concrete structures, but it was a very successful cooperation.

IG: After the completion of the project, were the ideas incorporated into the universities’ curricula? Was this something that was added quickly or was it dismissed?

RB: I think there was not a lot of interest in this type of structure, even after Munich. I do not know any university that included it into their curriculum.

IG: Really?

KG: Very little, still.

RG: For the engineering industry, there were a lot of new things developed during the project in Munich. There was a special research program where we could explore in more detail what we did, for example, how local ropes behave, their elongation, and their non-elastic behavior. Or how cable sockets should be designed and how they should look. There were many of these points which had to be investigated in more detail. The results that we got during the Olympics were studied later by individual groups at universities, especially at Leonhardt’s Institute for many years in order to make those results available to other engineers.

KG: You could say that it was a different process than it typically is. The research came after they had built it. The research was based on the design assumptions, which were made in order to build it. And then it went back to the university to get more depth, more proof, and probably, in some cases, more general results.

IG: What was the lasting effect that this project had on you professionally and, perhaps, personally?

RB: Nearly the whole team that worked together in Munich came back to Stuttgart. Of course, we were interested in designing more cable structures and this really influenced not only this design team, this group in Leonhardt's office, but also us personally. A big number of structures we designed had to do with cables. We designed only a few cable nets. The ice-skating rink in Munich was one of the next cable net structures. We also designed a second stadium roof with a cable net. We had finished the design and then, unfortunately, it was not built because the contractor decided to do a cheaper roof with columns within in the spectator’s area. We designed many bridges with cables, not only the cable-stayed large bridges. We also did a lot of pedestrian bridges with very complicated cable work later on.

KG: I always say that the Munich period was when you and Jörg, more or less, decided that you would start schlaich bergermann at one point. I am not sure whether you discussed it or whether you felt it. Is that true?

RB: Our collaboration started in Munich. It is clear that it started there. In the initial phase of our own company, that we started in 1980, we tried to get as many cable structures of any type as possible. We also designed our first solar project—the tower in Manzanares with stayed cables. Munich has given us a lot of knowledge about cables and how to design with cables. We clearly have tried to make use of this knowledge.

IG: Now that the project has turned 50 years old this year, what do you think is the biggest success of the project?

RB: The biggest success is that it was built. It was really a surprising success for us. Professor Leonhardt, when he said to the Olympia Baugesellschaft that this roof could be built, had no idea at all about what problems would come up. He just thought we could build it like in Expo 67 in Montreal, which we very soon found out was not possible. We had to find a lot of new things during this process, which was under a very strict schedule. In my opinion, this was the biggest success: that it was built, that it is still there today, and that it still is considered a success. Although, unfortunately, it is not used as much today as we would like it to be used.

IG: From the opening of the project until now, perhaps there have been design and management changes.

KG: Now you could say that the Olympic Park, and the roofs especially, are well-protected by their caretakers. The caretakers change sometimes because it is very often political positions, but they are well-protected. The capacity to change things is small, but a few things have changed. One change we made together was the suspended ceiling in the swimming hall. I would say that it was designed in a rush at the end of the original project, and it was just put up there with hundreds of pickup points and those pickup points have created problems. When we had the job to redo it, we came up with a solution where the suspended ceiling carries its own weight, for which we had to adjust the shape. Some of those pickup points that were part of the original appearance disappeared, but we were able to convince the authorities as well as the architects at that time that this was the best solution and that it served the purpose of the original design too. But there is very little that has changed.

IG: Knut, you joined the office in 1989. I am curious to know if the project in Munich influenced you during your studies or during the early phase of you career.

KG: First, I was influenced by Rudy and by his passion for cable structures. The project in Munich was not a focus of mine at the beginning. As a young engineer, you are focused on the next project and on the next task. My interest in Munich came a little bit later. But I must say that the Munich project is probably the most emotional project I know. When you go there, you feel the strength of the shape and you feel the forces. For an engineer, it is a very emotional project. This is something I love, and I was really happy to be part of the team who continued to take care of the project. To do changes like the suspended ceiling in the swimming hall, and, at the later stage, a new suspended ceiling for the sports hall. Now, we are going into the next round of renovation and replacement efforts, which will take place this decade.

IG: Can you talk about the ongoing involvement of sbp in the project? Did the owner of the project assumed that it had to work with the same engineers? Is there a public process that you have to be involved with?

KG: I think the overall arrangement of project was very different fifty years ago than it is now. Now it is very formalized. We go through public procurements and all the public administration processes. So far, we have been successful, and we are still involved. At this point, we probably have the most knowledge about this project, perhaps more than the client themself. I think we will find our way to be involved.

IG: What is the current state of the project? What is the project used for?

KG: Eight weeks ago, the Olympiapark hosted the European Championships for ten days and it was fantastic. The crowd was amazing. The roof was great. There were days with sunshine, days with rain, and it was fantastic. And I think you are going to see a new era of more things happening, especially in the stadium. The sports hall is the cash cow of the park. The stadium is more difficult, but we are seeing more things happening and events coming back.

In terms of our work, about four years ago, we recalculated the full project. We did 3D scans, we did force measurements, and we did new form finding. Rudy says, with his understatement, how they worked at that time. We looked into every little detail that was done in the 60s and 70s, and it is amazing. With all the possibilities that we have now, it cannot be designed better than what was done at that time. We found very little deviation from what was built. We found no single element that was overstressed or overloaded. What, of course, we found is corrosion. Cables had to be clad with fire protection material, which doesn't perform very well and sometimes gets wet, especially where cables hit the ground or go into a deep foundation. There is where we really need to look at. And, of course, the cladding will be replaced again.

RB: Regarding the point that no cable was overstressed, the safety levels that we use today for cable structures is even higher than what we used at that time. Today, you have a 2.4 factor of safety. At that time, there was no code giving these factors, so we assumed a factor of 2.0 and, under certain load cases, 2.2. It is not very evident that every cable part was not underestimated in our design. As mentioned earlier, the way we designed some parts of the roof was not very sophisticated.

KG: To be precise, there is only one cable which is overloaded to current codes. And it was overloaded at that time too because during construction things had to be modified. But there was a good reason why that could be accepted. In the new models, we can show an alternative load path and a minimal impact if such cable would eventually fail. Now, it is also monitored differently than it was before.

IG: What is your current scope of work? Can you explain what you are contracted to do?

KG: Four years ago, the contract was to recalculate the structure to make sure everything was safe and to check everything according to the level of safety in the detailing, in coatings, and corrosion protection. There are new floodlights and the number of them has increased, so the floodlight trusses are not safe anymore. We are now in the process of creating a new design for those elements. We are also creating a manual for how those elements need to be protected in the future, what needs to be changed, and what needs to be done. This will be followed by a call for tenders. After the call for tenders, there will be a big process onsite with closures of some of the facilities for certain periods.

IG: Can you talk about the process of checking the structure and the technology that is available right now?

KG: 3D scanning was very helpful from the ground but also from the air. Rudy explained how they did the cutting patterns for the cables, and that is still difficult. The cable net consists of two layers typically of two cables. You get an overlap from your scanning, so we decided to merge the four into two, which are in one plane. You still have to do it partially by hand. Those cables come to the edge cable where you have clamps, and the clamps are basically not in the line of the force, they are perpendicular to the edge cable. There, you had to model a lot of things by hand, which was a big task too.

For form finding, we used our own software. In this case, you have a form, you measure the forces, and you try to find the best possible form to fit the given forces. And the rest is, let’s say, not normal software but commercially available software. To look at the details, you need to climb up there, you need to open clamps, you need to look underneath, etc. You need the same thing we had done 50 years ago.

IG: I am curious where the historic documentation for the project is archived. Is there a place where this information is available or gathered for future research?

KG: I think it is in various locations. The client has, from a technical point of view, from what is built, quite a good archive. I would say that four years ago, the information was somehow organized but not accessible to the client. We organized the information regarding the roof for the client and now it is in much better shape. There is a lot of information out there, but it is in bits and pieces. Some is at the South German Archive, and some is at the Architecture Museum in Munich, and I am sure Rudy and Jörg have something in Berlin. But basically, from an engineering point of view, what we need is there.

IG: Is there any myth about the project that gets continuously repeated that you want to take a moment to clarify?

RB: One story is the contribution of Frei Otto to the structure. Internationally, the Munich roof is Frei Otto’s roof and not Behnisch’s, and this is really not true. Behnisch is almost unknown. Otto had a certain contribution to this roof, but it was very limited, and all the rest is Behnisch.

BB: Who were some of the main architects at Behnisch’s office that you worked with directly?

RB: I remember the two main architects we worked with on the roof was [Fritz] Auer and [Karl-Heinz] Weber. For the roof itself, Günter Behnisch joined us in the session maybe once a week, but not more. I think the input from Behnisch was rather smaller during the detailed design. It was mainly Auer. Weber was more involved in other parts, such as the landscaping, but not as much on the roof. The other young engineers, a lot of them were still students at that time.

KG: It is not a myth, but there are two things that come to my mind very often. First of all, Jörg was thirty-four when he got appointed as the project lead.

RB: Yes.

KG: And you were...

RB: Twenty-six.

KG: Twenty-six. If we think of our offices, when do we give the biggest, most challenging project to a thirty-four and a twenty-six-year-old engineer? That is one aspect that is amazing. If you look at the documents that were created, they are extraordinary. And the other topic which I know a little bit from other sources is that this project was developed during the period of the student protests. Leonhardt had a major role in Stuttgart as he was the director of the university. There was no work-life balance then. It was work, work, work. With all these things happening at that time, it is really extraordinary what was created.

RB: On the team we had in Munich, we were about twelve or thirteen engineers working on the roof project. The team was put together in Munich. I think two or three of us came from the Stuttgart office but all the other ten came from Munich and we didn't know each other. It was really an international team coming from Italy, Switzerland, France, etc. It was a mixed group and, fortunately, all the members got along. But of course, there was a strong input from the top, from Jörg Schlaich.

KG: The client side [the Olympic Construction Company] was brave too.

IG: How was the structure on the client side?

RB: There was one figure on top [Carl Mertz] who really made the decisions. Before being in that role, he was director of the German Federal Office for Building and Regional Planning. It is the agency responsible for designing German embassies all over the world as well as public buildings for the government here in Germany. He had a strong and positive impact on what was going on. We could come to him with every request and if he thought it was reasonable, independently of what it would cost, he then would say yes. The large number of tests that we did had to be paid for by the client. That was in addition to what was specifically design work. The client really was important for us.

IG: Thank you, Rudolf and Knut, for sharing your work in this remarkable project.

Acknowledgements

The authors would like to thank Johanna Niescken and Victoria Weht from sbp for their assistance coordinating the interview and providing the images; Sigrid Adriaenssens, Kai-Uwe Bletzinger, Carlos Lázaro, and Guy Nordenson for suggesting questions for the interview; and Julie Michiels for her help editing the interview.