The intent of this project was to produce with the greatest economy of means continuous curved surfaces without the costly computer assisted technologies, and see whether a tensile membrane can serve as an alternative in creating sculptural forms on a smaller scale. To accomplish this, curved surfaces developed in models made of a stretchable fabric were simulated by stretching chicken wire between fixed wooden frames, the metal membrane shaping the curvilinear structure and its interior space as well.

Ground was broken in September of 2002, and the work was completed in July 2003. Since the experimental aspects of the project -- the shaping of the curved metal membranes and the assembly of parts -- progressed as envisioned before construction began, no conceptual changes were made.

The finished shell is between seven and eight inches thick. It consists of two layers of plaster applied to the interior of the wire membrane; a layer of pleated metal mesh attached to the membrane's exterior surface to control the thickness of a waterproof spray-foam application; and two layers of stucco over the foam. Once the stucco was insulated with a waterproof 100% acrylic coating in two colors, the overlapping bands were painted with regular exterior paints, their color intended to reflect the colors so dramatic in this part of the country.

Throughout, the main concern was to follow the membrane configurations as closely as possible. Distortions, however, began to appear already in the shaping of the membrane, as shortcuts, such as laying strips of chicken wire straight up and down instead of diagonally (see Site & Models), produced bulges that appeared only later, and they had to be worked out, not always successfully. In laying the pleated wire units, intended to keep a uniform thickness of the spray-foam and reinforce it as well, required extra attention -- whether the pleats compressed or expanded, it changed the depth of the pleat, and units laid in a different direction also created ridges that were difficult to resolve. The initial attempt to stabilize their expansion by joining the pleats at cross-points on one side, only made the edges more jagged. In all other respects, the pleated wire mesh served its purpose.

I took notes of how many hours were spent on each experimental stage. The three chicken wire membranes -- over the kitchen, the studio, and the upstairs bedroom -- took three people ten days to complete. The installation of the felt-lined metal mesh panels on the underside of the membrane, took two people seven days. Hog-rings, used in upholstery, proved to be the ideal tool for joining the membrane layers together and attaching both mesh panels and pleated units to wire. The wire pleating machine worked efficiently, and less than 150 pleated units, each 36x36 inches square and 5 inches deep, were used. The units were easy to lay, and it took three people eight days to apply them to all three membranes. The application of a 4-1/2 inch layer of foam was done in three days by a professional outfit, and covering the foam with a wire mesh for stucco took three people eight days. Two standard coats of stucco were applied by others, and painting the color bands took two people seven days.

Since we had no foreknowledge and learned on the job as to what kind of detailing the next application required, in some instances preparations for the next step took longer than the step itself. Thus the number of days mentioned above pertains only to each specific application, and is not an accurate account of the overall time spent on this project.

To share the experience, every step of the procedure was documented, and to date the website has been visited by people from 69 countries. There is much room for improvement in method and especially in the use of materials. Those familiar with the industry should have no difficulty to further explore the seemingly endless possibilities.

I had the good fortune to work with a group of exceptional individuals, all local builders. First on the scene was Kirk Cobb, whose practical considerations helped me translate the shapes developed in the models into readily available building materials. Contractor David Benakis bravely undertook to pour the foundations, erect the framework, and install the utilities without structural drawings, while all experimental procedures were paid for weekly, as the work progressed. Richard Lesperance, an energetic hands-on builder, was in charge from start to finish, and undertook the responsibility to complete the project. And Kenny Autry, a perfectionist, paid close attention to details and finishes at every stage of the construction. Their experience and skills, and their unwavering commitment were indispensable to this venture.

Aleksandra
New Mexico, August 2003

Comments? Questions? Email to: kasubaworks@aol.com