xarch start previous up next end

layout and panelization
how to split up an irregular surface into similar bits

large building parts rarely come in monolithic structures; usually they are made of lots of smaller elements.
to find an appropriate layout for these elements is a key issue in building design. for the kunsthaus project we had to find subdivisions for the shell structure that respected the non-orthogonal nature of the shell but also connected to adjacent parts of the building - and some of these had a rectangular layout, e. g. a system of beams in a floor slab.

this is a common problem - let us see how other people have solved it:
i have mentioned the bmw pavillion before. its beams, seen from above, resembled a rectangular grid. they were a system of planar curves cut out of sheet metal and then bolted together with rectangular connectors. please mind that, although the connectors are rectangular, the cladding is not because the curves will meet in all possible angles. a similar example for this approach are the domes that have been developed by the german engineer jörg schlaich. he is also using a rectangular system of interconnecting beams. schlaich's domes have a very even and gentle curvature and their apex is not very high compared to the footprint. geometrically they are a flat rectangular grid with a circular outline that has been compressed and therefore the structure bends away and subsequently moves up. it is very crucial that the mesh is rectangular and not triangular because the single fields in the net need to be skewed. they will be stiffened by diagonals (usually a cable that runs continuosly from end to end) after the deformation. what i want to point out is that the skewing of the fields has an influence on the cladding since it will not be rectangular anymore. this may be almost negligible for schlaich's domes, but when the apex becomes higher or the structure starts to overhang (like the bmw pavillion) the shortcomings of this system become evident because you can not resolve 'corner situations' any more. of course buckminster fuller has found the classic solution for this and applied it to spherical structures - but not for strange shells like the kunsthaus.