Growing and making Flimwell’s chestnut gridshell

All images c/o  Woodland Enterprises Limited

The Weald and Downland Museum’s Gridshell wasn’t the only early gridshell structure in the South East. The Flimwell Woodland Enterprise Centre’s first building saw Bath practice, FeildenClegg, complete an adapted modular gridshell taken from the local weald wood, chestnut. The research and development which went into this early chestnut projects launched this low grade wood as a construction material, so that chestnut has become a core, niche market across many sustainable projects in the South East of England

See also Fourth Doors Unstructured

David Saunders, project manager of the Flimwell Woodland Enterprise Centre, looks out of the glass-less window frame of his near-to-completion building and into the thick foliage of mixed growth immediately beyond. As he looks, Saunders can see chestnut everywhere. “Although I haven’t done the sums, it looks like 5 hectares of chestnut would provide enough wood to grow a new building every year.” We are upstairs in the building which is part of an experiment in turning an historically important, though generally ignored coppiced wood, into a potential core building material for the whole of the South East region.

Seven years since the project was conceived, the Feilden Clegg building went up in 2000, with its adapted modular gridshell roof, cladding, walls and flooring, all using chestnut. Right in the heart of the South East’s once extensive chestnut region, a principal aim has been to demonstrate that chestnut can again be employed as a contemporary material. As with the Weald and Downland gridshell building it is an attempt to embed contemporary use in the midst of a traditional rural context. Unlike the Downland gridshell however, the project has been generated by the forestry, and rural industry, rather than building and architecture.

By the early nineties, with the withering of the chestnuts industries, outpaced by cheap softwoods, it became clear to those working in the region that if anything of the economy was to survive, new uses for the wood would need to be found, epecially as chestnut is such a predominant wood in the clayey Weald of the South East, where there are19,000 hectares of trees. There is an annual excess of timber of half a million tonnes a year, and each year, the core tree capital grows another 100,000 tonnes of potential timber. If this amount is used in a year then the harvested amount doesn’t actually affect the central reserve of core woodlands. Fencing and outdoor materials takes up the best of the coppiced tree, leaving lower-grade material, which hitherto could not be used for primary uses because of the irregularities caused by knots, and uneven surfaces. Up until recently it has gone to be pulped as wastewood. But at the Woodland Centre, a whole new variety of uses are being created.

Before the Woodland Centre was mooted, regional foresters and timber engineers had long felt that if a project could be put together which demonstrated that if chestnut was a strong enough material to pass various structural tests, it could once again be re-introduced to the building and construction market as a home-grown and therefore far more sustainable wood to utilise. Aware of the developments in timber and glue technologies, the engineers were confident that if these were applied to chestnut interesting results would follow. Furthermore, if this could be demonstrated by an innovative building “a living example of how the wood could be used, and packed with examples of designs, including furniture, and other uses the wood could be put to – this could only help the future of the industry.

There had already been a number of experiments using coppiced chestnut. Some of the buildings at the local farm, where one of the woodland partners worked, had been replaced using chestnut. Subsequently, the roof of a local sawmill was rebuilt using the wood, after a successful test hanging weights from beams which demonstrated that the timber would be strong enough to support the roofing. It was also known that there was a strong sustainable argument for this resurrection of the wood as a commercial material.

Those involved felt sure chestnut would measure up, but the wood needed to be comprehensively and formally tested. With a strand of EU money, called Life funding, and wood donated by a local sawmill, the Building Research Establishment (BRE) were brought in to the research picture. Firstly, English oak, (in 1995/6) and then the chestnut, (in 1997) were examined for their structural properties, and stress strengths. BRE’s research demonstrated that chestnut was indeed both safe and strong. What emerged from testing strength, stress, and durability capacities to sections of both chestnut and oak, was that much less hardwood was needed compared to softwood, to support or carry specific weight-loads. If a building uses smaller sections, it will use less wood, and that wood immediately gains a competitive edge. This was good news for chestnut, as was the fact that the results have made their way into the review of British Standards for Temperate Hardwoods (BRE Digest 445) which will lead to the revision of BS 5268: Part 2. The implication was that chestnut could be considered for use by architects and the building industry. It also meant that research shows that the best of the low-grade timber could be used for building. With deformations removed, remaining timber could be cut into small 350-500mm sections, and using finger jointing technology and, as in Weald and Downland, Collano polyurethane glues, reconstructed into long slats – up to 30 feet – from which the modular roofing sections could be created. Structural engineers, armed with the materials research would now be able to model large-scale structures, such as modular gridshells, using the chestnut.

Out of this positive research exercise a part-private, part-public company, Woodland Enterprises Ltd, was set up to develop ways to first steer, and then exploit the research alongside other concurrent initiatives to breathe life back into woodland industries.

Now, the idea of a complete building – constructed from chestnut – could be envisaged. The Woodland Enterprise Centre began to look at this as a serious option. The concept of an actual physical woodland visitor centre had been in the ether for a while, but with the structural and materials research in, a feasibility study was commissioned. Once completed, this again showed the potential for developing a market for locally produced hardwoods, and chestnut specifically, and a bid from Woodland Enterprise was, in 1996, one of only six to receive a grant of one million pounds. From there a competition produced an imaginative entry from Feilden Clegg, (these days, Feilden Clegg Bradley), chosen because of how Woodland Enterprise envisaged trees from the locale being integrated into the building plan. After some to-ing and fro-ing Feilden Clegg adapted their plans to concentrate a chestnut as the building’s primary material. The project evolved into a working woodland centre, aimed at providing training, work and accessiblity to the woodland industries community. Feilden Clegg was able to develop a new building with a modular roof, rather than a purely gridshell roof as the original brief required.

The benefits of the research began to kick in. Feilden Clegg’s rejigged building plan integrated slats as part of the modular gridshell roofing structure. For these, chestnut was cut into short 28mm x 75mrn sections and sent off to Newcastle to be remoulded into the 10m lengths, which could then be put together as the gridshell slats for the modular bays – modelled by the structural engineers Atelier One, who’s empire at the time included a
Brighton office. The chestnut used was freshly felled greenwood, which meant it was partially dry though retaining high moisture. This meant that less washing and preparation was required before drying, producing a much more stable timber. The suppliers were a local, Kent based Woodland Trust coppice, and a Sussex-based chestnut forester, Richard Cope. Even If obtaining the supply was relatively straight forward, certain difficulties ensued with the joiners and industrial carpenters inexperienced in working with the hardwood. Indeed what became, was the need to introduce industrial carpentry and finger-jointers to the special requirements by hand, as it is a difficult and unknown material for those used to softwood. Specific machinery designed for hardwood, as softwood-centred technology is not entirely suitable.

When the wood arrived, it did so in varying conditions and quality. The key to ensuring that the chestnut is good enough to be used is through the initial selection process, what must be done by a skilled eye, and experienced assessor. The chestnut research consultant, and traditional timber architect, Nigel Bradon, has developed this to a fine art, identifying where the slope of the grain is too steep, or the proportion of knots too heavy, both of which can lead to failures. Once identified, the potential defects are removed, and visual timber grading is applied for future usage, and then the timber is sawn into 75mm by 25mm sections.

The building, which sits off the edge of the A21, south of Tunbridge Wells, deep in dense Weald woodland, is a compact and attractive mid-sized structure, fitting well amongst the surrounding trees. The large-span modular gridshell carries five bays, the sawn and fingerjointed laths, bolted together, comprising the barrel vaulted arches. The curve of these arches is slightly eccentric, rather than symmetrical. The five bays were constructed on-site by Cowley’s, the timber carpenters. These modular sections can be constructed anywhere, in prefabricated units in a factory, and transported to the prospective site.

Feilden Clegg have declared themselves pleased with how the chestnut has performed from the sustainable angle; it is durable, and does not need chemical preservatives. Beside the chestnut, the building’s main environmental features include its mineral wool insulation, making it a low-energy building. Roof lights provide the upper floor with considerable natural lighting. The aim is to develop a wood-fired heating system, from excess waste woodthrown out by the site’s separately owned and operated sawmill, which should provide enough heat for all of the prospective buildings. The foundations are concrete – unavoidable because of the Weald’s clayey ground soil.

This said, the principal sustainable innovation is in bringing chestnut into the environmental building equation. The centre demonstrates how chestnut and other hardwoods can be used in mid-sized structures, and as building components, all locally sourced from across the South East. Where before the only obvious options have been softwoods, and/or tropical timber, Flimwell shows that there is a real alternative. Cost-wise, the structure is coming in at around £1000 per square metre, compared with rock-bottom steel and concrete costs of £500. While this is a significant difference, in environmental terms there is investment going into both the regional environment and the local economy. This may not be persuasive for smaller buildings, but as far as mid-sized buildings are concerned it does look potentially competitive. Set against softwoods, and due to the implementation of tougher building regulations, chestnut could well become an attractive option.

The viability of chestnut as a building materials is demonstrated in Woodland Enterprises’ plans. They are intending to set up production within the Centre’s grounds, and the next proposed building is a factory, which could prefabricate the modular sections and other hardwood building components. As such, it provides a persuasive alternative option to the main new choice being developed for low-grade woods; its use as wood fuel energy. If fuel is a high volume industry, by contrast the Woodland Centre chestnut initiative employs a value-added approach to a low-value material.

Whereas regional building requirements are considered, both for mid-size and domestic buildings, there seems to be considerable scope for growing a rejuvenated chestnut coppice timber industry. With the arrival of new building regulations, whether you’re mainstream or at the margins of building, the use of wood, already enjoying a national resurgence, will likely increase. Major housing developers are already reintroducing timber, including Wimpey, Westbury, Laing, and Wilcon. But a main difference with previous timber industry growth-spurts is its use in low-rise medium-size buildings, of which there are about 15% average across England, including hotels, nursing homes, community-centres and commercial and leisure centres. Here, large-span structures are clearly an obvious option, and are one of the markets being eyed-up by the Flimwell Woodland Enterprise. The other primary point is the sustainable sourcing of local wood.

David Saunders, the Enterprise Centre’s manager, believes that around 5 hectares of chestnut would provide enough wood “to grow a new building every year,” If Saunders is right about his back of the envelope calculations – and there is no reason he isn’t – there are 5 tonnes per hectare of chestnut growing each year. Thus, 15 tonnes would be enough to grow another Flimwell. An annual 100,000 tonnes of chestnut could provide the core of many, many, mid-sized and other buildings for the region, as well as a versatile variety of other applications. These are provocative figures to be considered by other regions – within their contexts – when contemplating how to regenerate rural economies. It is early days, but through wood and glue technology, Flimwell has become a temple to the chestnut for aficionados of the tree, and a laboratory for possible woodland futures.

Footnote –  The building uses 12.5 tonnes of prepared chestnut in the structural gridshell, weatherboarding and joinery. Taking into account the high losses in conversion from small logs, this would be derived from a total of 40 tonnes of coppice roundwood. As chestnut coppice grows at around 8 tonnes per hectare per annum, this amount is equivalent to 1/5 of the production from 1 hectare of 25 year old coppice. Alternatively, the area of woodland required to produce one such building each year is 5 hectares.

The earliest version of this piece appeared in the autumn 2001 edition of Building for A Future (vol 11, no 2) It was reversioned as part of Fourth Door’s webspread Unstructured 3 look at gridshells, Gridshelter Futures in  2003 and appears here in hybrid form somewhere between the two earlier versions.

Leave a Reply