A story of a Frank Lloyd Wright Residence in the 21st Century
LEAKS
Unfortunately, one of the things Frank Lloyd Wright is known for is leaking buildings. Fair or not, the stories are legendary. He built a remarkable factory and office for the Johnson Wax company and later a residence for its founder, Mr. Johnson. The Johnson Wax Works in Racine Wisconsin and the 1938 Wingspread residence on the outskirts of Racine are iconic achievements in American architecture. He got a call from Mr. Johnson. “I’m having a diner party and it is leaking above my chair at the head of the table.” Mr. Wright’s reply was brief and to the point, “move your chair”. Those who know the Marin Civic Center know that it leaks like a sieve and nobody has been able to do anything about it. This spring, my wife and I visited the 1923 Millard House in Pasadena, one of the textile block houses also considered by Wright to be one of the first Usonians. It had just rained and water was dripping through the living room ceiling onto the floor some 14 feet below. The house is for sale for $4,995,000 leaks and all.
I don’t want to take up the argument one way or the other. It’s reasonable to hold, however, that the first duty of architecture is to build shelters that protect one from the elements. Particularly for an “organic” architecture like Wright’s that wants to get right down to basics, to the union of materials, form and function. Is it fair to claim you’ve achieved that and then tell your client to move his chair? Like I said, I’m not going to get into it. I’m crazy about Wright’s architecture, leaks and all. But as a builder, I can’t build things that leak, even things like Wright’s structures that truly push the envelope of design.
I expressed this to Maynard on a number of occasions when we were struggling over water control details. He tired to ease my mind. “Keith, we’ve lived here 50 years and it’s leaked 49 of them.” I reminded him that I was a boat builder. I don’t build things that leak. “Maynard”, I said, “it may have leaked before, but when I’m finished your house wont’ leak.” He greeted my youthful optimism and unshakable confidence with a slight nod of his head and twinkly in his eyes and a respectful. “We’ll see.”
The Usonian concrete slab was the source of potential trouble. The interior and exterior slab elevations were the same. There were no door thresholds and the redwood walls, as well as the concrete block walls, came directly down on the slab. The threat of water coming under the door or seeping under the walls was an obvious problem. Meticulous attention to concrete finishing at the doorways, the application of brass moisture stops at the wood walls and the application of a concrete key with bentonite moisture stop at the concrete masonry units, were some of the means we used to meet the challenges of Wright’s design. Glass corners where glass meets glass provided obvious challenges. The perforated board windows which were really no more than a section of the board and bat that opened up to let in air, required special attention. Everywhere we turned the challenges of creating a water tight envelope were beyond any previous experience of myself or my workers.
With the exception of the tilting living room roof, the roof was dead flat. A flat roof can be treated to have enough slope to move water, but this was almost completely negated by the thin 8 inch thickness of the entire roof assembly on the Buehler House. Even with this restriction we sculpted a cardboard substrate under the water proof roofing to within 64th of an inch to move water to the internal roof scuppers. Standing water is deadly for any roof membrane.
The soaring living room roof was covered with copper shingles. The field of a roof is seldom the problem. It’s the edges that require attention. Here we had copper shingles ending against wood, sometimes concrete block, sometimes other metals such as the copper fascia or other roof membranes such as the tar and gravel flat portion of the roof. Each intersection of disparate materials required its own unique solution and choice of materials. Wright buildings don’t leak. Poor joints between materials leak. We say, “the devil is in the details”. In a Wright house, the details are a devil.
The house did not leak. Not for the first year. Not for the second. But on the third winter, I got a call. “Hi Keith, it’s Maynard here.” “Oh, hi”, I said. He got right to it. “It leaks”. “What leaks?” I asked worriedly. “It leaks in the living room.” My heart sank. On the other hand, I recalled Maynard’s rather casual attitude about the first 49 years of leaks. I hoped he might be up to a few more. Just to make sure, I asked, “What would you like me to do?” trying to sound nonchalant. “Fix it!” came a familiar voice that never left a doubt as to what was expected.
Cregg and I went out to the house. We applied water to the area in question and were able to reproduce the leak, It was a seam between the copper roofing and the concrete masonry block chimney of the double fireplace. A little investigation turned up an interesting fact. The caulk we used to seal the concrete to the copper did not adhere well to copper. We learned this from a Richmond outfit specializing in caulking materials. They had the right material and we soon had it in place. The right material was all that was needed to turn the tide on this unfortunate Wright leak.
That’s a happy ending. But unfortunately, not the full story…
Keith R. Alward
November, 2011
