Skip to content
Only a person who has polished a 66-foot-long, 33-foot-high, 125-ton, hollow, stainless steel kidney-bean-shaped sculpture until it was so shiny you could use it as a shaving mirror could tell you what a cumbersome chore that sort of thing can be, and how long it will probably take you if you decide to attempt the task yourself.
And yet it seems that practically everyone in Chicago has an opinion on whether or not it has taken just a little too long to finish and finally unveil the gargantuan sculpture in question -- Anish Kapoor's dazzling, crowd-pleasing "Cloud Gate," a.k.a. the Bean.
Come Sunday, if things have gone as planned, it will have been completely divested of its massive Quonset-hut-style tent for several days, and the plaza will be opened to the public. While it's not the first time the tent has come off, and the Bean won't actually be "finished" per se, the plan is to leave it alone for a month, then have the workers back on-site beginning in October to finish polishing the portion known as the "omphalos," the interior vortex underneath Cloud Gate. Aside from closing off that area, the plaza will remain open to the public and the Bean will never be forced to wear that dowdy tent again. City officials say the job should be complete by winter 2005-06. But you never know.
And if you ask Danny Kozyra -- the project's 38-year-old general foreman, who has headed the crew of ironworkers since construction began in January 2004 -- there's not much way anyone could have predicted exactly how long such an uncommon task would take. And he should know. He's one of only 25 people in the world who've ever tried it, and they all belong to Ironworkers Local 63.
"Nobody had ever done anything like it before," Kozyra was saying on a recent morning, a few days before the tent-removal process began. It was as hot as an oven inside the tent, where the Bean rose like a pampered space-age egg out of a jumbled nest of wires, stuffed garbage bags and scaffolding. Kozyra and the rest of his crew were wearing protective respirators and white polyethylene jumpsuits, which in spite of their substantial physiques made them look like Oompa Loompas as they buzzed around the gargantuan Bean, buffing and polishing and scrutinizing it from precarious-looking positions on scaffolding and ladders, or stretched out underneath it on the ground.
"If this had been done previously anyplace in the country or the world," said Kozrya, "you can bet [we] would have had someone on the phone asking them to come out here and show us how to do it."
Did testing on process
According to Senior Project Manager Steve Belcaster, of U.S. Equities Realty: "We did some testing last year, to get a better understanding. We welded steel panels in the shop and ground them down to a smooth finish, so you couldn't see the weld anymore -- like a mirror. . . .
"And then when we got up on top out here, it was a totally different story. We couldn't use the method used in the shop, where there were no compound curves. The panels we tested were flat. So the instruments didn't work the same way."
So with Chicago watching, waiting and worrying about money, Kozyra and his crew set about "defining and refining" a process for executing a job that will likely not be done again anytime soon. "It's been extremely painstaking and difficult. Because there's no tolerances and absolutely no room for error," Kozyra said.
The Bean was structurally engineered in London and California, fabricated in California, and then shipped to Chicago in pieces, which Kozyra and crew put together on-site after drilling holes and sinking anchor bolts into the concrete slab above the Park Grill to support it.
But to hear Kozyra tell it, building the Bean was an absolute breeze compared with polishing it.
"From an engineering standpoint it was phenomenal -- just unbelievable," he said, with his respirator pushed back on his head and sweat streaming down his face, as he conducted a private tour of the goings-on in the tent, yelling over the constant roaring noise of sanding equipment, air-cleaning HEPA filters and cooling fans (which weren't helping much).
After erecting the subframe structure -- the two huge rings that rose up in 2004, connected by a network of what looked like erector-set pieces -- the crew placed 168 panels of 3/8-inch thick steel (weighing 1,000-2,000 pounds) over the frame like a dress pattern and sewed them all together with bolts. And it fit almost perfectly, something that surprised Kozrya after 17 years in the business.
No cutting, filing
"I've never been on a job where you could take pieces, in the form they arrive, put them together, and not have to cut them or file or do anything to them."
Instead, the resulting 2,442 linear feet of welded seams and the general dullness of the steel have been the real challenges, because the sculpture had to be brought to a high-grade mirror finish -- something hard enough to pull off on a flat surface much less a curved one as big as a house.
Kapoor's design was inspired by a ball of mercury, and in a way it has been a metaphor for the job. You put your finger on a solution to a problem, but it gets away from you again.
"When we put it together we thought, `Holy cow, that's going to be easy. . . . It looked great, but when we came back we noticed that while the edges were fine, an inch or two in from the edges there were imperfections. Now if you start grinding it, you have to start expanding your grind zone, because if not you have dimples and ripples . . . so now you're trying to make this good, but you've got to pay attention around you because if this is good and this isn't. . . . '"
The non-existence of specialized Bean grinding, sanding and polishing equipment was one obvious problem.
"You get in a car accident, and you have to fix your car with a hammer and a piece of sandpaper," Kozyra said. "And bring it back to showroom quality. That's basically this job. There are only so many tools available to do it, and we had to figure out how to make them work.
"We had already modified the instruments; we had to refine them. Each step we had to stand back and say, `OK, you gotta build this, move this, twist this, try this -- OK.' So it's taking the machine and working on it for hours to get it right.
Kozyra went over to a big metal tool trunk and rifled through a messy pile of equipment including assorted buffers, grinders and air hoses. "This is a platen sander," he said, handing over a heavy piece that looked like an industrial lawn sprinkler.
"So hold that," he said, handing it over. "That's maybe 8-10 pounds." It seemed like 50, at least, but it also seemed like a very tiny sander for such a large surface. And according to Kozyra, it works on only a 2- or 3-inch area at a time.
"So picture this," he said, holding the sander over his head. "You're kneeling, hanging onto the side of the Bean by a rope, moving yourself up and down. . . . "
A rope? Why, with the network of scaffolding surrounding the Bean, would Kozrya or anyone else in his crew be hanging from a rope?
To work on the very top of the Bean, they could stand in stocking feet. And scaffolding was perfect for the sides. But for that hard-to-reach area where the giant Bean begins to curve away from the top, they had to think outside the box.
"We were standing here trying to figure out, OK, how the heck are we going to come down over the roll? Some of the guys had done some mountain climbing. So one guy went out and bought some climbing gear, put it on and let's go! Boom, boom, boom! We were all tied off on ropes and had harnesses on."
Which sounds like fun -- who doesn't love to rappel? -- especially if you're actually climbing a mountain on a cool spring day rather than welding or sanding for an eight-hour stretch, inside a tent that reaches temperatures of 120 degrees or more near the top.
"You're working in a very uncomfortable position, trying to weld and then trying to grind," Kozyra said. "And there's someone working beneath you. If you drop your sander it's going to hit somebody, or if you slip and gouge the Bean you have to start all over again."
"Stainless steel is a very forgiving surface, but if you scratch it, it can be a bitch," he added. "If it's a light scratch you can get it out, but if it's deeper you have to start all over to get it out. One little nick, one little gouge and you've basically created four, five, six days a week of work for yourself."
Down at the south end of the Bean, the one not poking out of the tent for tourists to dote on, a guy was standing on a high layer of scaffolding, one foot on a bucket, the other on a plastic milk crate, stretching to go over a patch that was not quite mirrorlike but still looked awfully shiny. "He's refining," Kozyra said.
"It's a lot of repetitive work: grinding and grinding and grinding and grinding, and then you hit it with some sandpaper. And then you have to say, `Look, we missed a spot,' so we have to go back up."
Down on the floor, a few steps away, a pair of workers were double-teaming the Bean: one was lying underneath it, using a buffer; the other was sitting down next to him, helping him apply the rouge, which is a little like Soft Scrub, according to Kozyra. "You put it on the wheel to clean it up and polish it -- they're taking out whatever scratches are left and bringing it to a high luster."
Kozyra went outside, to the end of the Bean protruding from the tent. Tourists were checking themselves out in the reflection, and smiling and waving at the workers inside, through the plastic window. "It's a little like being in the windows of Marshall Field's at Christmas," Kozyra said. "Everyone is usually pretty nice, but people have been upset that this is all we can give them."
Even though they're pretty big guys, tough enough to take the heat literally and figuratively, the criticism can hurt, especially after eating and sleeping Bean six days a week.
"People say, `Why is he just lying there?' Kozyra said, as we watched a worker touching up the underbelly "But imagine lying on your back doing that for 20 minutes at a time, 10 hours a day," he added, imitating the weight lifter's motion.
From the looks of it, though, the crew has been relatively undistracted in its mission -- in spite of its public nature. While Kozyra was talking, an ironworker wearing a Packers cap came out and balanced himself, like a high diver, on some boards protruding from the tent above the crowd. He hugged the curve of the Bean, placed his cheek next to it with one eye closed and began checking for imperfections, circling them with a marker.
"We understood from Day One: This is not your typical job, and it's got to be above and beyond everything you've ever done. . . .
"It's like Michelangelo painting the Sistine Chapel. Once you put the scaffolding up, it's going to take a while, because it's a very meticulous job. You can't just go up there and paint a bunch of stick people."
- - -
January 2004: anchor bolts set; construction begins.
March 2004: tent erected for the first time.
July 2004: tent removed; plaza opened to the public.
January 2005: tent erected for a second time.
May 30, 2005: north end modified for public viewing.
Aug. 28, 2005 (projected): plaza opened. Bean free for good.
- - -
Bean by the numbers
What was required for each 10-foot section
Rough cut: A 5-pound, 4 1/2-inch electric grinder, using three 40-grit sandpaper disks, was required to remove the welding seams.
Initial contour: a 15-pound, 2-inch, air-driven belt sander (a.k.a. the Road Grader), using 15 each of 80-grit, 100-grit and 120-grit sanding belts.
Sculpting: an air-driven 10-pound, 1-inch belt sander (or platen sander), using 40 each of 80-grit, 120-grit, 240-grit and 400-grit belts.
Refining: a DA (for double action sander), using 200 each of 400-grit, 600-grit and 800-grit disks, to remove the fine scratches that are left from the previous step.
Polishing: a 10-inch electric buffing wheel using 10 pounds of "rouge," or polish.
2,400-plus: linear feet of seams welded on the 66-foot-long, 33-foot-high sculpture.
6: levels of scaffolding during sanding/polishing process.
3,000: estimated number of protective polyethylene jumpsuits used for the crew of 24.
100: estimated number of protective respirators for the crew.
3,000: estimated pairs of respirator filters.
77: number of 50-gallon barrels of debris removed from site (spent sanding belts/disks, jumpsuits, respirators, etc.)
Source: Steve Belcaster, senior project manager, U.S. Equities Realty