General Design Issue::Design::Knowledgebase::SAFAS

Design of spatial structures is generally conducted through collaboration among the structural engineer, the architect, and the contractor. The type of grid configuration selected can be influenced by the site erection requirements, especially if the structure has to be erected in parts. (Click here for more information on the assembly and erection of spatial structures.)

Since the connections are usually the most expensive part of spatial structures, designing the structure with the least number of connectors (nodes) possible is usually the most economical alternative. In most cases the total costs of DLG spatial structures are up to 50% more than other conventional one-way or two-way systems, however, they are generally lighter by 20%-30%. In the past, use of the same members and type of connections throughout the structure was one of the main advantages of prefabricated spatial structural systems. However, use of the new computer-controlled cutting and drilling equipment have now made it possible to design structures comprised of different member lengths, node types, and connection angles with reasonable fabrication and assembly costs. In addition, computer aided analysis and design have greatly simplified the design process, facilitating the use of spatial structures, and in particular, the application of non-standard configurations. It has to be noted that typically the total costs of spatial structures are divided almost equally amongst engineering, fabrication, assembly, and erection.

For spatial structures with single or double curvatures such as barrel vaults, domes, hyperbolic-paraboloids only nodular systems such as Mero, Triodetic, etc. can be used.

For preliminary design purposes, members of spatial structures are generally assumed to be pin-ended.

Flat double-layer grid (FDLG) spatial structures are efficient forms for two-way structural systems. Clear spans of 100 ft x 100 ft can easily be achieved with relatively small size structural members. Due to the load distribution capabilities, these structures can carry overhead loads suspended underneath, in addition to their dead and live loads.

At the preliminary stage of design, once the overall dimensions of the building are established, the FDLG designer needs to select the following parameters: (a) the topology (configuration); (b) the grid size; (c) the structural depth and (d) support locations and conditions to minimize the stresses due to load and temperature variations. For the preliminary analysis and the design of DLGs, all members are often assumed to be of the same size, However, this results in a heavier structure, in particular due to the over-sizing of the diagonal members.

The most efficient (least weight) spatial structure configuration (topology) is usually based on several parameters including the total loads, the span and support conditions. The square-on-square offset and diagonal-on-diagonal offset provide the most dense and structurally rigid configurations which are particularly suitable for large loadings and long spans. However, the less dense configurations of square-on-diagonal, diagonal-on-square and square-on-larger square are usually used for light loads and moderate spans (less than 120 ft for FDLGs).

To create the most efficient square-on-square offset FDLG spatial structure, the ratio of the cross sectional areas of the bracing members to the top/bottom layer members should range between 0.2 and 0.4 (based on the condition that the allowable stress for all the members is the same). However, this does not have any effect on the maximum deflection. The self-weight of FDLG spatial structures with spans up to 120 ft is about 3 psf with the weight of cladding about 3 psf. In general, when compared to traditional truss systems the FDLGs are 30%-50% lighter. FDLGs have been typically used for spans within 75 ft to 400 ft range. For larger spans, one may use multi-layer grids or barrel vaults or domes with single, double, or multi-layer configurations.

When selecting spatial structures as a roof system, the soil condition is an important factor that has to be particularly checked. If the roof is supported by only four corner columns, they will be subjected to large loads which may require very good soil capacities, otherwise large foundation systems may be required.

In most cases, spatial structures are made of interconnected tubular members. The main reasons for this selection are:

Tubular members have concentric cross-sections and their moments of inertia and radius of gyration are the same about any axis (i.e. the member buckling load is the same about any direction).
It is possible to keep the outside diameters of all members the same and only change the inside diameters if different member sizes (based on forces to be carried) are required. This can simplify the construction of spatial structures.