The Perils of Restoring

Krueck & Sexton architects faced tremendous challenges in the restoration of Mie Van der Rohe's 1956 masterpiece, S.R. Crown Hall, at the Illinois Institute of Technology
This course is no longer active
[ Page 3 of 4 ]  previous page Page 1 Page 2 Page 3 Page 4 next page
From Architectural Record
Sara Hart

The sound of glass shattering

"Mies's design was experimental," explains Sexton. "At the time, glass technology was in its infancy, and there were few regulations. Mies relied on his intuition regarding size." The facade consists of two types of windows. The original upper panels were enormous and not tempered-9 feet 8 inches by 12 feet 9 inches-and only 1¼4 inch thick. With no codes to dictate size and thickness at the time, Mies was free to push the known limits of engineering. His experiment had mixed results. Great expanses of crystal-clear glass poured daylight into the studio and created the "barely there" effect he sought. However, "great expanses of glass have a tendency to break and fall out in strong winds," Sexton adds. (Anecdotal evidence suggests that no one was ever injured, or at least not seriously.)

click images to view them larger
1
2
3
The contractor shrouded the entire building (1) to keep the dust from the lead-based paint from escaping into the environment. Students work at Mies-designed drafting tables in 1956 (2). Mies's grandson, architect Dirk Lohan (3), ceremoniously launches the restoration by smashing one of the glass panels.

In contrast to the perfectly transparent upper panels, the lower units were sandblasted, which provided some privacy for the students, but mostly it served to hide interior activity from view, and thus retained a pristine Modernist face to the public. However, the glass was sandblasted on the interior face. Sandblasted glass is porous and absorbs oil from fingerprints and the adhesives students used to mount drawings on the windows. Due to decades of unintentional damage, the lower units became stained and scratched. In addition, the building was subjected to ad hoc alterations and mandatory repairs between 1970 and 1977. In 1975, Skidmore, Ownings & Merrill (SOM) replaced all the glass, including the thin upper panels, which were replaced with 3¼8-inch lites. Then, employing new technology for the lower panels, SOM installed a laminated glass with a mylar interlayer to imitate the translucent qualities of the original sandblasted glass. Although the solution eliminated the staining problem, the result was more reflective than Mies's matte finish, and repairing the repair, as it were, presented a perplexing challenge for Krueck & Sexton.

click images to view them larger
1
2
3
At least five full-size mock-ups (3) were produced to test several glazing options and the blind replacements. Studies of solar radiation transmission show the difference between laminated glass units (1) and single sandblasted panes (2).

Every solution seemed to create more problems. The first challenge, and the one that generated the most controversy, centered on the upper panels. First of all, the glazed area was too large by code (or common sense) to have the original 1¼4-inch-thick polished plate glass replaced. To comply with the code, the architects could have simply specified 3¼8-inch tempered glass. Keeping in mind that Mies's details are deceptively simple, Sexton recognized that tempered glass could be less than perfect. "When glass passes through an oven during the tempering phase, it can develop the slightest surface wave from contact with the rollers," he explains. The waviness can sometimes be noticeable when viewing reflected images from a distance. This would be unacceptable to even the most forgiving critic.

Now, 1¼2-inch glass doesn't have to be tempered, so Sexton was confident that he could deliver a perfect surface. But, of course, every solution yields a new problem. In this case, the problem was color. Iron in glass gives it a green tint, so the thicker the panes, the greener the tint. Crown Hall was intended to have glass so clear as to seem barely there, which was possible with 1¼4-inch panes. The architects eventually found a manufacturer that could make low-iron glass in such large sheets, in order to achieve maximum transparency, high-fidelity light transmission, and the kind of brilliance usually reserved for jewelry cases and museum displays.

The 68 original lites of the upper panels were replaced with PPG Starphire (low iron) glass, and the cycle of problems and solutions continued. Thicker glass is heavier, of course. The new panels weighed 700 pounds, making them way too heavy for the original stops. In other words, the stops had to be enlarged from 5¼8 inch to 3¼4 inch in depth. As the wall sections show, this redesign is subtle enough to be called invisible. However, it has a slope. The architects felt, and the mock-ups confirmed, that a deeper reveal would look heavy. Sexton argued that by sloping the stop from 3¼4 inch at the glass to 5¼8 inch at face, it would read the same as the original. The purists rebutted that it would be blasphemous to introduce any amount of slope in a rigidly rectilinear structure. They also argued that Mies used off-the-shelf extrusions, and a sloped stop would have to be custom fabricated, a clear violation of his principles. Sexton, with the support of Dean Robertson and Gunny Harboe, prevailed, because they convinced all the interested parties that, first of all, the slope cannot be seen. And secondly, compromising on the custom-design issue was better than specifying a heavy, and thus inappropriate, stock stop.

 

[ Page 3 of 4 ]  previous page Page 1 Page 2 Page 3 Page 4 next page
Originally published in Architectural Record.
Originally published in January 2006

Notice

Academies