THE PISHWANTON HAND-BUILT GRIDSHELL

Both the Flimwell Modular gridshell and the large scale Weald and Downland building have required large amounts of public money into them. It could be it is fortuitous chance that Sussex is the epicentre of the current British gridshell phenomenon. There is, however, a third building, completed in 2002, which is a gridshell, but it has been built on much smaller budgets, done independently and through the energy and enthusiasm of a small group of people.

Image David Tasker

The building is the craft centre at the Life Science Centre Trust, Pishwanton, East Lothian, a centre devoted to the renewal of the largely unknown Goethian approach to scientific research. Designed by the respected architect, Christopher Day, it is deeply organic, quasi-rustic building. It mention of the core of the building - the craft room, where ears have been perking up, for its roof is a ten metre span self-built gridshell, constructed overwhelmingly by the craftsmanship of working hands. The £1.8 million it cost to build Weald and Downland, is beyond the reach of many communities who might want to explore such forms at an equally valid, if far smaller scale. This is where the Pishwanton building comes in. Through what appears to be a mixture of fortitude, foolhardiness and staying power, the Pishwanton building has been completed over four years in a stop-start, almost hand-to-mouth fashion. It is testimony to those involved that against the kinds of odds where more mainstream professional operations would have despaired, the Life Science Trust has seen the project to completion. And they have done so at a cost which is a fraction of the institutional amounts needed: £50 and £60, 000 .

Christopher Day began sketching out the architectural ideas in 1997. The first work commenced in 1998, the buildings function being to provide both sheltered facilities for Life Science Trust activities in bad weather, and for pottery classes, wood carving and other woodland crafts. The original engineer’s envisaged an apparently sophisticated but prohibitively expensive truss system, which would apply beams and columns to the central area. It was at this point that David Tasker, a structural engineer, who had already been involved in the construction was brought in to reassess the situation; and Tasker, upon seeing the column planned, thought “this wants to be a gridshell.” The resulting plan envisaged three pod-like gridshells resting on an irregular hexagon, nine metres in diameter. Today, on each side of the completed hexagon room are two wings, one for storage the other comprised of two smaller rooms.

Image David Tasker

Tasker describes the next steps as working through limited structural analysis, "backyard engineering", mainly scaled modelling and “basic ‘shell theory’”, but without any software programmes; a full 3D structural analysis wasn’t applied because of uncertainties over the wood and the prohibitive cost. Tasker also acknowledges the jointing as very basic - partially as a result of its computer independence.

Apart from the wood used in the gridshell, beams and truss, the building used primarily locally sourced limestone, stone, sand, and wool. No Portland cement was used at all. In its place an imported environmentally friendly limestone was used for the foundations, and as mortar for the stone walls. A second, St Astier lime, this time from France, was applied to all the above ground walls. The walls were built up by a contracted stonemason from local stone brought in from a farmer's derelict dry-stone wall a handful of miles along the hill ridge. The sand used was from the local quarry. Mineral wool was used in the external walls and cavity walls, while in the other openings sheep wool was used. Marketed as Cosywool, this unusual organic mineral wool came from the Wool Marketing Board; being made into 'tops' on site, small long sausage shapes cut from 100 metres of the material. The work took time because the wool needed repeatedly folding over, but was pleasant on the hands - unlike many comparable glass and other industrial wools.

Image David Tasker

At first there was some scepticism about gridshells. They didn't appear to follow the way the wood wanted to go, rather, it seemed as if the form was fighting against nature. But soon the challenge became interesting for those involved, including a forester and joiner, Malcolm Lemmon. Quite quickly the technical problems of realising the project, apart though causing mighty headaches, became fascinating issues with which to wrestle and seek solutions.

Image David Tasker

Lemmon found 40 tonnes of larch from a project in Perthshire, which had foundered and was closing down. Lemmon bought this after checking its sustainability factors. The timber for the window frames was Highland Douglas fir - an increasingly popular, and durable, tree crop in Scotland - manufactured into windows in Edinburgh. For Lemmon, as for all those centrally involved, it was a steep learning curve. What analysis there was happened on site as the building progressed. Happily, once construction was underway the roof turned out to be not as difficult as Lemmon had presumed. Tasker had prepared a small experiment with a miniature pine dome model, and Lemmon used this to see how the wood would respond. From this he realised that young green larch would work well; the sapling’s elasticity ensuring the bending needed for the curvature. From there, Lemmon went looking for the right potential larch, finding it eventually in one of the Earl of Rosebery's estate's forests, a few miles south of Edinburgh. Trained to be able to uncover woods from eye, he hand-picked the larches, and took them away as part of a management thinning project with the estate. These were prepared as laths around four metres long, and next, initially by trial and error, Lemmon began constructing longer laths, scarf jointing three together to make twelve metre lengths, until there were two amounts of 20 laths, able to cover twelve metres. These were made up in two layers of 35mm by 25mm by 600 cm, with the bits drilled in to form the nodes. The nodes were checked using pilot bolts, on the jig, to hold each square of the grid together.

Image David Tasker

Once ready, the grid was put together and bolted up in one day, with nearly two dozen volunteers helping at the site. Two hours later the volunteers had manoeuvred the grid into position as part of the roof. In the event only two joints needed replacing. Completely floppy, the gridshell was at this point both very supple and fragile until the boards covering the laths were also put in place. After this with further boarding applied, and the structural shape forming around a scaffolding tower and temporary timber struts, the shell began to co-elesce.

While the corners were held down by rope, the first two layers of 100mm boards were screwed into the lattice, the shell, according to Lemmon, becoming more uniform, as it began to dry out. The board wood was found from a local forest four miles away, an economic choice since it was small enough to be cut into shape. In fact, the whole gridshell cost was minimal, in the region of £2500, although as Lemmon points out when you add labour, his work and two intermittent paid helpers, it inevitably increases the cost. As it was put into place, each layer of boarding soon reinforced the emerging shape. Alternate layers were placed at 90 degrees to each other, to reinforce the strength. With three layers across the whole dome, and four at the edges this produced an effectively extra-tough roof. An extra complication was that the shell was being placed onto the hexagon. Domes have been placed on square and circular rooms before, fitting these proportions easily, but they have not been set upon hexagons. Not surprising, perhaps, they do not fit so readily. Lemmon pulled the gridshell dome down upon the two opposing sides of the hexagon walls initially, until they began to bend into shape. Once the gridshell dome and boarding was in place it was covered with a turf roof, and a Rubberfuse membrane was sourced as the best eco-friendly option available. Insulation continued to be a problem and cork was eventually applied to the membrane. The cork, another import, this time from Portugal, compared well with polyutherene and other polyester insulations. It was finally completed at the beginning of the year, then covered with horse and cow dung, and presently sports a thick grass matting.

Before it went up, no-one knew whether the gridshell would actually work and although Tasker felt confident there wouldn’t be any real structural problems, there were quite a few nerve-wracking days wondering what they were going to say to the funders if it didn’t work. The load testing was unconventional. Rather than computer analysis, the building group hand calculated 700 bags of sand which when massed equalled 13.4 tonnes of sand, so that the deflection on the roof could be carefully measured – it came in at 25 millimetres, equalling the worst snow cover scenario.

Happily, the building did work out, and by the summer of 2002 it was beginning to be used. The building represents a success story for the small-scale and self-built, and it also demonstrates the tenacity of a small team. One can easily speculate that were local building contractors let loose on such an unknown quality as a gridshell, since so few have any experience of the form, let alone worked on one, they would price the risk into budget oblivion. Since only one or two companies have the relevant experience, this must be one of the main constraints on its growth in use, with specification and design generally so expensive. The Pishwanton gridshell demonstrates that with resolve and patience people can construct for themselves. All those involved state they learned a variety of things from the construction, and could improve on it if it were being constructed again. Certainly such smaller-scale applications seem obvious; offices, chalets, community halls, restaurants; the list goes on. Gridshell pod roofs could be used for 10 m classrooms both in, and outside, the Steiner-Waldorf context. Further uses in a variety of contexts are also easy to imagine.

One surprise is that gridshells were new to Day. Even if there are very few, they suit his ‘Spirit and Place’ aesthetic. He believes the double-curved parabolic form contains considerable strength, and describes the internal space as a very gracious and harmonious gesture. All of which helps make the form so satisfying, in Day’s words, “at a preconscious level". Indeed the Pishwanton Lothian project deepens the range of forms of this band of buildings represent, pulling another example of this genuinely exciting, elegantly ecological and twenty-first century form into material existence.

Image David Tasker

End.

 

Contents . Project Research . Fourthdoor Review . Unstructured1 . Unstructured2 . Unstructured3