Donald MacDonald has designed more than a dozen bridges
since establishing his architectural firm in 1966. Today, he has bridges scheduled for construction in Dubai and Portland, OR, and he's in the midst of building the world's largest self-anchored suspension
bridge: the SF-Oakland Bay Bridge's eastern span (due to be completed in 2013). We visited MacDonald's San Francisco offices to find out what it's like to build a $6.3 billion bridge, how things have changed since the "old days", and why young architects should learn to just "Draw the bloody thing!"BBG: You have a history of updating and redoing classic bridges like the Golden Gate. What goes into reworking another architect's creation?
MACDONALD: Anytime you work on a historical structure, there's already a design vocabulary. I have a whole ritual of doing an analysis of what the original architect intended. With the Golden Gate, I went into [Irving] Morrow's documentation to see how he handled light and traffic and functional things. I try to find out all the ingredients. With the Golden Gate, I picked up certain forms that are redundant. They're not engineering, not structural. They're there for effect, to play with light and shadow. But those forms can be applied to safety. For instance, when I did the bike rail [on the Golden Gate], it has a vertical i-beam which picks up the vertical shape that's in the pickets and the light poles. So if you add something new, you use that shape and it just works.
When you want to modify a heritage structure, there are Federal 106 guidelines
that state that if you do any changes, they have to be state of the art, of the moment. On the north end of the Golden Gate, underneath you can see new trusses and old trusses. The new trusses are square tubes where we plasma cut the shape to mimic the [look of the] old trusses that were riveted together in a crisscross pattern. Our tresses are much stronger structurally, but if you're a layperson and you're not cognizant, it looks the same. There's an analysis called "solids and voids." It's very simple: there's a hole (void) or something that's filled (solid). So when you do something new, the idea is to pick out those patterns and carry them through to the new form.
BBG: Your vision for the Bay Bridge is a total reimagining. The most obvious change is replacing the double-deckered east/west bound lanes (above right) with the Skyway. What was the thinking there?
MACDONALD: The design of the initial bridge was the cheapest way to build. But you're riding in this tremendous long tunnel that's like a bridge of darkness. The idea was that if you put those two flows of traffic side by side [above left], you open up views of the whole East Bay. If those two sections were attached, you'd have 14 lanes wide plus the bike lane
. Opening them up helps to get light into the water below to help the fish and plant life.
BBG: All architecture is a reconciliation between design aesthetics and what's possible from an engineering perspective. With this project, how difficult was that process?
MACDONALD: The project is so big and expensive. It's $6.3 billion, which is the largest project Caltrans has ever had. So there are many players involved. You have to present to their peer-review panel, which has I think 21 people and this went on for two and a half years. We've had a tremendous amount of control on the design, but early on the panel set up some guidelines: the new bridge should pick up the west-bound half, meaning that it should look like one continuous bridge on both sides
[of Yerba Buena Island]; they didn't want the tower any higher than 525 feet (that's what the towers are on the west bound); they also decided it should be either a cable-stayed bridge
or a self-anchored suspension bridge
. [By going with] the self-anchored system, there's a real harmony to it. If you get up in an airplane or look at the bridge from the hills, you can see the island has the reverse form of the cabling.
BBG: So the shape of the cables as they curve carries the same wavelength as the island?
MACDONALD: Yeah, basically. It picks it up so it ties the bridge into the context of the island.
BBG: And that was a design choice?
MACDONALD: It's a coincidence. The silhouette of the island is somewhat similar to the silhouette of the catenary
form on the
main structure. But that wouldn't have happened with a cable-stayed bridge, because those bridges are like a sail -- all straight cables with no curve. A self-anchored bridge is also more
flexible in an earthquake.
BBG: Are there any unique ways the bridge is designed to withstand quakes?
MACDONALD: The main tower is like a cantilever -- the deck [roadway] is not attached -- allowing the tower to have about three feet of movement. Inside the main tower leg are four steel link beams that are like sacrificial beams. They are expected to deform and rupture under the stress of an earthquake, but the bridge will be able to stay in operation until we can go back in and fix the beams.
Engineers design the beams for the steel's elastic limit, so they can tune the link beams to pop at different periods, so as you go up the tower, the harder the earthquake, the more link beams will rupture. And [the stress] can happen in any direction, because there are four legs to it. It's a pretty unusual tower.
BBG: What's your take on the free-for-all we're seeing in Dubai?
MACDONALD: If you're an architect or a bridge designer, it's the place to be, because there are no rules. They're letting you fly. There's a lot of criticism, sure. All of a sudden you've got an instant city in the desert, and how are you gonna support it? But for me, it's a relief to really let it fly. But just about every project in Dubai is on hold, so we'll see.
BBG: What's on the horizon for bridge building, aside from huge spans?
MACDONALD: There's no smashing new technology in bridges. It's a very slow progression of form. The Brooklyn Bridge was when Roebling
introduced wire cable and he wasn't even sure about how to use a suspension bridge, so he's got two systems: a cable-stayed overlapping a suspension system, which is redundant, but for him at the time I think he felt he didn't have enough mathematical analysis available so he wanted to use both systems. And nowadays you can use one or the other, but if you mix them, it's more expensive. There's a whole scale for bridges. The most expensive is the
suspension bridge. The next one's the cable-stayed bridge, then the arch and boxed girder. The engineers help decide which type based on a variety of dilemmas. The budgets have become very much a part of the process.
BBG: How has your work changed as technology, rendering software, etc. has gotten easier to use and cheaper?
MACDONALD: Computers haven't done a damned thing as far as the economics of producing a building. They're just as expensive as just five guys drawing. There's tremendous inefficiencies in computers because just about everybody doesn't just work on a computer. They check the news or check the stock market, personal emails. There's a lot of inefficiencies and a loss of manpower and hours that are billable. In the old days when you sat and drew, you didn't get on the phone all day.
But another downside [to technology] is you can't evaluate a young architect's credentials very well. Since everything is done inside a computer, you can't understand how he thinks. Some architects become really good at [rendering with] computers, but putting things together, they screw up. There's been a lot of lawsuits because of leaks in buildings where it's taking some design, one detail, and pulling it into your database and nobody checked it out, so it goes onto 20 buildings. But in the old days, when you re-drew it each time, you'd check it out.
The other thing is you don't have the same flexibility to move personnel. It's like now I gotta get a guy that's on ArchiCAD
. I'm old enough that I'm not that patient with it. Draw the bloody thing! Just sketch it out! And [a lot of young architects] can't draw anymore.
Universities are going back to teaching drawing. I lectured at Notre Dame about six months ago, and it's my understanding they're going back to drawing. Students focus on [drawing by hand] for the first few years and then they go into computers after that. So the school produces guys who can draw beautifully and render well. And they're in high demand now.
Of course, it's not all negative: there are great efficiencies in emailing photographs and drawings. We used to do a lot of work with Germany, and shipping used to take 3 days over, 3 days back. But as far as people and production, it's the shits, as far as I'm concerned.
BBG: What about your own use of computers for rendering?
There's tremendous redundancy in computers. Instead of erasing a line, you go into the computer and change the line, and then print. In the old days, you would just erase the line and change it [on the same draft]. We have CDs here that must have hundreds of drawings, just for the Bay Bridge. They want to archive it at the Bancroft Library
, but who's going to sit there and go through hundreds and hundreds of illustrations?
BBG: That's a problem with a lot of industries. We're creating tons of data we don't know how to sort through.
MACDONALD: They don't even know how to archive all that stuff. I say, "Well, should I print it out on archival paper?" Libraries have a tremendous problem... Anyhow, there's good stuff and there's bad stuff [to technological progress and computers], but in general it costs the same amount of money, takes the same amount of time to get the buildings up.
BBG: Why else is a hand drawing preferable to a computer rendering?
MACDONALD: It's better to show a drawing to the public, because it's general enough. With a computer, it's got to be pretty well worked out, because it's real; it represents reality. So it's not a good way to begin a design process, because you're stuck with the damn thing. Whereas, with a hand drawing you've still got moves -- eyewash, so to speak. You can start with generalizations and slowly melt them down with details.
So now there's a real demand for drawings. I'll draw my little hand off doing 10 drawings in two days, because no one wants to make a presentation with a computer rendering. They also take way too long. I can do a rendering in about 3 hours [by hand]. For a guy on a computer, it takes a goddamn week just about. It's just a riot.
BBG: You've been working on the new Bay Bridge since 1997. What's it going to be like for you when the old Bay Bridge is demo'd into the water and the last piece of your new bridge drops into place?
MACDONALD: If I'm still around in 2013 [laughs]. I hope I'll be sitting right there at the front of the line, probably on the main deck near the tower. It'll be spectacular. Once in a lifetime, old boy.
[photo by Dan Cressman]