Engineering and Construction

Tweetear

Ongoing and long term cleaning and maintenance will be required and may impact on the design. Ongoing work will require a simple safe means of access. It will be necessary to account for the loading and connection of gantries, walkways, harness points, etc., in the design of the primary structure.

Long term maintenance could be facilitated by independent access systems such as scaffolding. The main impact on the design of the structure will be the quality of the specification. The life to first maintenance of the finishes and components will increase with initial cost.This needs to be balanced by the costs of the maintenance activity.The client should be consulted on the options available and choose the most appropriate option. This will depend on his/her long term interest in the building fabric.

Tweetear

An atrium structure will have several complex interfaces:

• The cladding
• Cleaning equipment
• Supporting structure

The management of these interfaces is critical to the successful implementation of the design. If the structure, cladding, etc are procured through separate contracts it is even more critical to define the technical requirements and responsibilities at interfaces. The following issues should be addressed:

• Forces and movements to be accommodated
• Compatibility requirements of materials
• Design responsibility for components and fixings
• Definition of overall theoretical and as-built geometry.

It can be convenient to incorporate several packages of work into a single contract.

This can reduce risk associated with the management of interfaces. This route may increase cost due to the increased responsibilities carried by the contractor.

Tweetear

The engineer should always consider the erection method for a structure. In the case of atria this may be of particular concern for the following reasons;

• The site is constrained by surrounding buildings
• The application of load to existing structures to be carefully controlled
• The need to precamber the geometry
• Restrictions on the location, reach and capacity of craneage
• Limits on the size of components
• The formation of neat site connections
• Safe access for operatives for the erection of the primary structure, cladding and other items.

The erection methodology may preclude the use of some structural forms. It is, therefore, essential to consider this aspect at concept stage.

Tweetear

The detailing of individual members and their connections is crucial to achieving a satisfactory architectural treatment. It is commonly the fabricator’s responsibility to detail connections, and often this expertise can inform the design process.However, it is important to ensure that aesthetic considerations are properly defined and controlled. The following should be considered:

• Nature of the connection between components of the primary structure
• Connection of the cladding system
• Connection of sundry components such as access equipment, lighting, etc.

Connections can be expressed or ‘invisible’.The strategy for connecting components should take into account the practicality of achieving the desired outcome. Site welding can offer ‘invisible connections’ within certain limitations. The following should be considered:

• The type of welding technique and position required to form a connection
• Butt welding can be ground flush whereas fillet welds cannot
• The type of weld testing required to verify connection adequacy
• The effects of heat distortion on the structural strength and geometry
• The effect of welding on shop-applied finishes.

Where connections are aesthetically critical, sample workmanship should always be called for as a basis for selecting a contractor. Samples should be kept as a record against which to judge finished workmanship.

Steelwork will require a paint finish for reasons of aesthetics and durability. Paint technology provides an ever increasing number of options.However, the choice can be guided by consideration of the following criteria:

• The steelwork fabricator who delivers and erects the steel will inevitably damage any paint applied at the works no matter how carefully the protection measures are prepared.
• The contractor who has a very tight programme may like to see primer, under- coat and finish coat on the steelwork applied at the works since it may be very difficult to apply the finish in high-level steelwork without extensive temporary works.
• Very hard paint finishes are more difficult to damage although when the inevitable occurs they are usually more difficult to repair to a good standard.
• If the finish-coat is applied on site, some parts of the steelwork may be difficult to reach and the standard of finish will not be as good as that achieved in the paint shop.
• Unless touching up to damaged areas is very minor, the result is unlikely to  be satisfactory and therefore a full site applied decorative finish may prove  necessary.
• The choice of system will also depend on the fabricator’s paint shop and its relationship with the paint supplier.

Tweetear

Atria generally require transparent or translucent forms of cladding. Glazing is a traditional solution but is relatively heavy. If the space beneath is heated the thermal performance may dictate double-glazing, which is both heavy and  costly. Other forms of cladding such as ETFE foil cushions can offer a lightweight, thermally  efficient and cost-effective alternative to glazing.

Glazing systems may be framed or unframed.The allowable spans of the systems vary.The relationship between the layout of the primary structure and the cladding system will determine whether there is a need for a secondary system of support.

It is important to consider this relationship at concept stage to achieve the minimum  of elegantly arranged members. The following dimensions provide guidance:

• Planar glazing – spans in either direction around 2m
• Framed systems – spans 4–6m by 2m
• ETFE foil cushions – spans 6–8m in either direction.

Manufacturers should be consulted for more precise information.

Framing systems are generally made from aluminium extrusions, and therefore have thermal expansion and contraction characteristics which differ from steelwork.

This needs to be considered in the detailing of the connectivity between the two elements.

Tolerance between cladding and supporting steel framework also needs careful consideration. The accuracy with which steelwork can be fabricated and erected is generally lower than that of the cladding system. This incompatibility should also be accounted for in the detailing of connections.

Deflections of the atrium structure under loads can have significant bearing on the cladding system. Some glazing systems require a minimum slope to ensure their water tightness. The deflected shape of the supporting structure must satisfy these requirements under all relevant loading conditions. This may require the structure to incorporate a precamber from the theoretical final geometry.The cladding system must also be capable of accommodating any differential movements of the sup- porting structure.

Tweetear

For a new building there will be significant choice in the types of structural systems that can be adopted, as the perimeter supports can be designed accordingly.Where an enclosure is being added to an existing building, available support will depend on the capabilities of the existing structure and may determine the appropriate options.

The structures in Figs 5.36 to 5.40 illustrate a series of solutions in which there is a generally decreasing reliance on the perimeter structure.

Fig. 5.36 Arch

Fig. 5.37 Beam or truss

Fig. 5.38 Half portal

As a group the arch, the beam and the dome are potentially the most structurally efficient solutions but rely totally on the support provided by the perimeter structure. In the case of an existing building, the following should be considered:

• The strength and integrity of the existing structure if this is to be used for support.
• The percentage increase in vertical load imposed on the existing structure by the new.This will be smaller where new loads are introduced towards the lower levels of existing walls and columns than at the top.
• The horizontal forces imposed at the supports. Horizontal thrusts introduced  at lower levels will have smaller resultant overturning effect  than at a higher level.















Failure to investigate the implications of the choice of structural form could result in a costly and clumsy solution.

As town centres and streets are redeveloped to compete with out-of-town shopping, the need for enclosure is a common solution, often considered. In these cases, the adjacent buildings are typically of a wide variety of constructions of various heights and condition.The solutions defined in Fig. 5.39 may be appropriate to avoid the complexities of resolving the various boundary conditions.The perimeter structure is required to support a small proportion of the vertical  load and relatively small horizontal loads from wind and stabilizing forces for the new columns, which does not generally pose a problem since there is usually some redundant horizontal capacity due to the wind shielding effect of the new enclosure. The portal style solution (Fig. 5.40) is substantially self-supporting.

Temperature effects are often a governing design consideration  for an atrium structure due to exposure to heating from sunlight. Temperatures in steelwork painted black can be 30° higher compared with white where directly exposed to the sun’s radiation. It is important to allow for movement of the structure, to prevent locked-in stresses from building up.Releases of this sort are often incompatible with providing horizontal stability. The choice of structural form needs to reflect these opposing requirements. For example, a statically determinate structure such as a three pinned arch can accommodate temperature movements and differential  settlements without generating secondary internal forces (see Fig. 5.41).

Fig. 5.39 Beam and column

Fig. 5.40 Portal frame

(b) Fig. 5.41 Movements caused by temperature changes

Tweetear

An atrium is an enclosed courtyard space often rising through several storeys of a building. Its function is usually to provide daylighting within the plan of a building.

Often it provides covered circulation space between floor levels which readily orientates the building occupants. In common with all aspects of building design, there are a number of design considerations which come into play in the creation of an atrium space. These broad design requirements generate specific technical requirements for the design of the structure. These need to be taken into consideration at the developmental stages to ensure that an elegant design, consistent with all of the technical criteria, can be achieved.

The following design issues are common to the design of an atrium space:

• Structural form
• Type of cladding
• Tolerance and deflection
• Finishes and detailing
• Erection of structure, cladding and fit out
• Procurement
• Maintenance of the cladding, structure and fit out
• Internal and external cleaning
• Acoustic performance of the space
• Cost
• Fire safety of the atrium and surrounding spaces
• Provision of heating and cooling to achieve required environmental conditions.

The following sections look at some of the technical implications for the structural design and specification imposed by these design issues.

Tweetear

The galvanized strip steel from which the light steel framing is formed is usually designated as either grade S280GD or grade S350GD to BS EN 1014710 (formerly Fe E 280 G or Fe E 350 G). These designations indicate the yield strength (280 or 350N/mm2 ) and the fact that the material is galvanized with a minimum G275 coating. Cold formed steel sections are usually rolled from galvanized sheet steel that is typically 0.9–3.2mm thick. The normal thickness of zinc coating (275 g/m2 ) has excellent durability for internal applications. Heavier coatings are available for more aggressive external environments.

Tweetear

‘Stick-build’ or panel components may be assembled in either ‘platform’ or ‘balloon’ construction, as illustrated in Fig. 5.35. In platform construction, walls and floors are built sequentially one level at a time, so the walls are not structurally continuous.

In some forms of construction, loads from the walls above are transferred through the floor joists to the wall below.

In ‘balloon’ construction, the wall panels are often much larger and are continuous over more than one storey. Such panels are more difficult to erect than  single storey height panels and have to be temporarily braced whilst the floors are installed. The main advantage of this approach is that loads from the walls above are transferred directly to those below.

In both forms of construction, the external cladding or finishes  are generally installed and attached to the frames on-site.

Fig. 5.35 ‘Platform’ and ‘balloon’ forms of panel construction

Tweetear

In modular construction, units are completely prefabricated in the factory and may be delivered to site with all internal finishes, fixtures and fittings in place, as illustrated in Fig. 5.34. Units may be stacked side by side, or one above the other, to form the stable finished structure.

Modular construction is most cost-effective where large production runs are possible for the same basic configuration of modular unit. This is because prototyping and set-up costs, which are essentially independent of scale, can be shared across many units. Details of the use of modular construction in residential construction are given later.

Fig. 5.34 Modular construction using light steel framing

Tweetear

Wall panels, floor cassettes and roof trusses may be prefabricated in a factory and later assembled on site, as in Fig. 5.33. For accuracy, panels are manufactured in purpose-made jigs. Some of the finishing materials may be applied in the factory, to speed on-site construction. Panels can comprise the steel elements alone or the facing materials and insulation can be applied in the factory. The panels are connected on site using conventional techniques (bolts or self drilling screws).

The main advantages of panel or sub-frame construction are:

• speed of erection of the panels or sub-frames
• quality control in production
• reduced site labour costs
• scope for automation in factory production.

The geometrical accuracy and reliability of the panels and other components is better than with stick-build construction because panels are prefabricated in a factory environment. The accurate setting out and installation of foundations is a key factor to achieve rapid assembly of the panels and to obtain the maximum  efficiency of the construction process.

Fig. 5.33 Site assembly of light steel wall panels

Tweetear

In this method of construction (illustrated in Fig. 5.32), discrete members are assembled on site to form columns, walls, rafters, beams and bracing to which cladding, internal lining and other elements are attached. The elements are generally delivered cut to length, with pre-punched holes, but connections are made on site using self-drilling self-tapping screws, bolts, or other appropriate site techniques.

The main advantages of ‘stick-build’ construction are:

• construction tolerances and modifications can be accommodated on site
• connection techniques are relatively simple
• manufacturers do not require the workshop facilities associated with panel or modular construction
• large quantities of light steel members can be densely packed and transported in single loads
• components can be easily handled on site.

‘Stick-build’ construction is generally labour intensive on site compared with the other methods, but can be useful in complex construction, where prefabrication  is not feasible. This form of construction is widely used in North America and  Australia, where there is an infrastructure of contractors skilled in the technique.

This stems from a craft tradition of timber frame construction that now uses many power tools. In these countries, traditional timber contractors have changed to light steel framing with little difficulty.

Fig. 5.32 ‘Stick-build’ site construction using wall studs and floor joists

Tweetear

The basic building elements of light steel framing are cold formed sections which can be prefabricated into panels or modules, or assembled on site using various methods of connection.The different forms of construction are reviewed in the fol- lowing sections and are illustrated in Figs 5.29 to 5.31.

Fig. 5.30 Light steel framing using prefabricated panels

Fig. 5.31 Modular construction using light steel framing

Tweetear

Light steel framing uses galvanized cold-formed steel sections as the main structural components.These sections are widely used in the building industry and are part of a proven technology. Light steel framing extends the range of steel framed options into residential construction, which has traditionally been in timber and masonry.

The Egan Task Force report called for improved quality, increased use of off-site manufacture, and reduced waste in construction, and the Egan principles have  been adopted by The Housing Corporation and other major clients in the residential sector. Light steel framing satisfies these Egan principles and it combines the benefits of a reliable quality controlled product with speed of construction on site and the ability to create existing structural solutions.

Light steel framing is generally based on the use of standard C or Z shaped  steel sections produced by cold rolling from strip steel. Cold formed sections are generically different from hot rolled steel sections, such as Universal Beams, which are used in fabricated steelwork.The steel used in cold formed sections is relatively thin, typically 0.9–3.2mm, and is galvanized for corrosion protection.

Cold formed steel sections are widely used in many sectors of construction, including mezzanine floors, industrial buildings, commercial buildings and hotels, and are gaining greater acceptance in the residential sector. Light steel framing is already well established in residential construction in North America,Australia and Japan.

Tweetear

Wire rope cables are spun from high tensile wire. For structural work the cables  are multi-strand, typically 6 x 19 or 6 x37 with independent wire rope core and  galvanized to Class A. For increased corrosion-resistance, the largest diameter  wire should be used, and cables can be filled with zinc powder in a slow-setting polyurethane varnish during the spinning process. For even greater corrosion-resistance, filled strand or locked-coil strand can be used to which a shrunk-on polyurethane or polypropylene sleeve can be fitted. Stainless steel, although appar- ently highly corrosion-resistant, is affected by some aggressive atmospheres if air is excluded; the resulting corrosion can be more severe than with mild steel.

The simplest and cheapest type of termination is a swaged Talurit Eye made round a thimble (Fig. 5.28(a)) and connected into a clevis type connection or on to the pin of a shackle.The neatest and most streamlined fitting is a swaged eye or jaw end termination (Fig. 5.28(b)). Hot-poured zinc terminations have to be used for very heavy cables of greater diameter than 50mm (Fig. 5.28(c)).

Epoxy resin with steel balls can be used as a filler in place of zinc, offering an improvement in fatigue life at the termination.

On-site connections can be made with bulldog clips but they are ugly and damage the rope. For cable net construction the standard detail is a three-part forged steel clamp, of which the two outer parts are identical (see Fig. 5.28(d)). Forging is expen-

sive for small numbers and so for smaller structures machined aluminium components may be preferred. Double cables can have a swaged aluminium extrusion  prefixed to each pair of cables, which can then be connected with a single bolt. For the attachment of net cables to edge cables, forged steel clamps are generally used (Fig. 5.28(e)).Lower cost alternatives are bent plate or machined aluminium clamps.

Cable life is reduced by corrosion and fatigue. Galvanized cables under cover suffer very little corrosion; external cables properly protected should have a life of 50 years. Plastic sheathing has the great disadvantage of making inspection of the cable impossible.

Fatigue investigations have shown that it is wise to limit the maximum tension in a cable to 40% of its ultimate strength for long-life structures. For structures with  a design life of up to ten years a limit of 50% is acceptable. Flexing of the cables at clamps or end termination will cause rapid fatigue damage.

Fig. 5.28 Cable fittings: (a) swaged eye, (b) swaged terminator, (c) hot-poured white metal eye, (d) cross clamp, (e) net and boundary connection

Tweetear

The analysis of space frames results in the production of large sets of linear simultaneous equations which must inevitably require the use of a computer for their solution. For these equations to be formulated it is necessary to input into the  computer significant amounts of information relating to the topology of the structure and properties of the individual members. This operation can be very time- consuming unless modern pre-processing techniques which allow rapid data  generation are adopted.

The members forming double-layer space frames are principally stressed in  either axial tension or axial compression. Consequently it is usual in the analysis  of double-layer space trusses to assume that the members in the structure are  pin-jointed, irrespective of the actual joint rigidity.This assumption may lead to the overestimation of node deflections, but because only three degrees of freedom are permitted at each node this approach minimizes computer storage requirements and processing time.

Single-layer grid and barrel vault structures carry the imposed load by flexure of the members so it is important to include in the analysis the flexural, torsional and shear rigidity of the members. Linear analysis only is required  for the majority of skeletal space structures but to ensure stability of shallow domes it is essential to undertake a non-linear analysis of these structures.Large-span space structures may benefit from a full collapse analysis where the ultimate load-carrying capacity and collapse behaviour of the structure are determined.

The behaviour of both single-layer and double-layer space trusses is influenced to a great extent by the support positions of the structure. The effects of joint and overall frame rigidity also have a commanding influence and affect the buckling behaviour of the compression members within the structure. Great care must be taken in assessing the effective lengths of compression members and it is unlikely that internal and edge compression members will exhibit similar critical buckling loads.

Tweetear

The inherent characteristics of steel skeletal space frames facilitate their ease of  fabrication, transportation and erection on site. There are two main groups into which the majority of space frames may be classified for assembly purposes: the  particular structure may be assembled from a number of individual members  connected together by purpose-made nodes or alternatively may be constructed by joining together modular units which have been accurately fabricated in a factory before transportation to site.

There are numerous examples of ‘chord and joint’ space frame systems available for immediate construction. These systems offer full flexibility of member lengths and intersecting angles required in the construction of skeletal dome structures.

Many jointing systems are available; Figs 5.15 and 5.16 show a typical spherical node used in the MERO system and a cast steel node used in the NODUS system. 

Fig. 5.15 MERO node connector

Fig. 5.16 NODUS node connector

The ‘modular’ systems are usually based on pyramidal units which are prefabricated from channel, angle, circular hollow section or solid bars. The individual  units are designed to nest together to facilitate storage and transportation by  road or sea. Most manufacturers of modular systems hold standard units in stock, which greatly enhances the speed of erection. Figure 5.17 shows  typical details of the prefabricated steel modular inverted pyramidal units used in the Space Deck System.

Fig. 5.17 Exploded view of 2.4m square section of Space Deck

Steel space frames are generally erected rapidly without the use of falsework.

Double-layer grids of substantial span can be constructed entirely at ground level including services and cladding and subsequently lifted or jacked up into the final position.Dome structures can be assembled from the top downwards using a central climbing column or tower. A novel approach adopted for the erection of a dome with a major axis of 110m and a minor axis of 70m involved fabrication of the dome on the ground in five sections, which were temporarily pinned to each other.

The central section was then lifted and the remaining segments of the dome locked into position as shown in Fig. 5.18.

Fig. 5.18 Novel method of dome erection

Tweetear

Space frames are classified as single-, double- or multi-layered  structures, which  may be flat, resulting in grid structures, or may be curved in one or two directions, forming barrel vaults and dome structures. Grid structures can be further categorized into lattice and space grids in which the members may run  in two, three or four principal directions. In double-layer lattice grids the top and bottom grids are identical, with the top layer positioned directly over the bottom layer.Double-layer space grids are usually formed from pyramidal units with triangular or square bases resulting in either identical parallel top and bottom grids offset horizontally to  each other, or parallel top and bottom grids each with a different configuration  interconnected at the node points by inclined web members to form a regular stable structure.

Single-layer grids are primarily subject to flexural moments,whereas the members in double- and triple-layer grids are almost entirely subject to axial tensile or com- pressive forces. These characteristics of single-, double- and triple-layer grids  determine to a very large extent their structural performance. Single-layer grids, developing high flexural stresses, are suitable for clear spans up to 15m while double-layer grids have proved to be economical for clear spans in excess of 100m.

The main types of double-layer grids in common use are shown in Fig. 5.12.

Fig. 5.12 Lattice and space grids

Skeletal space frames curved in one direction forming single- or double-layer barrel vaults also provide elegant structures capable of covering large clear spans.

Single-layer vaults are suitable for column-free spans of up to 40m, which may be substantially increased by incorporating selected areas of double-layer structure forming stiffening rings. Double-layer barrel vaults are normally capable of clear spans in excess of 120m. Figure 5.13 shows the main types of bracing used for single- layer barrel vaults.

Fig. 5.13 Bracing of single-layer barrel vaults

Dome structures present a particularly efficient and graceful way of providing cover to large areas. Single-layer steel domes have been constructed from tubular members with spans in excess of 50m while double-layer dome structures have been constructed with clear spans slightly greater than 200m.

Skeletal dome structures can be classified into several categories depending on the orientation and position of the principal members. The four most popular types usually constructed in steel are ribbed domes, Schwedler domes, three-way grid domes and parallel lamella domes.

Ribbed domes, as the name suggests, are formed from a number of identical rib members, which follow the meridian line of the dome and span from the foundations up to the top of the structure. The individual rib members may be of tubular lattice construction and are usually interconnected at the crown of the dome using a small diameter ring beam.

The Schwedler dome is also formed from a series of meridional ribs but, unlike the ribbed dome, these members are interconnected along their length by a series of horizontal rings. In order to resist unsymmetric loads the structure is braced by diagonal members positioned on the surface of the dome bisecting each trapezium formed by the meridional ribs and horizontal rings.

Three-way grid domes are formed from three principal sets of members arranged to form a triangular space lattice. This member topology is ideally suited to both single-layer and double-layer domes, and numerous beautiful large-span steel three- way domes have been constructed throughout the world.

The steel lamella dome is formed from a number of ‘lozenge’-shaped lamella units which are interconnected together to form a diamond or rhombus arrangement.The spectacular Houston Astrodome is an excellent example of this type of construction. This impressive steel double-layer dome was constructed from lamella  units 1.52m deep and has an outside diameter of 217m with an overall height of 63.4m. Figure 5.14 shows the four main dome configurations now in prominent use worldwide.

Fig. 5.14 Dome configurations

Tweetear

Steel skeletal space frames are three-dimensional structures capable of very large column-free spans.These structures, constructed from either individual elements or prefabricated modules, process a high strength-to-weight ratio and stiffness. Steel space frames may be used efficiently to form roofs,walls and floors for projects such as shopping arcades, but their real supremacy is in providing roof cover for sports stadia, exhibition halls, aircraft hangars and similar major structures.

Fig. 5.11 Outrigger stability: (a) Vierendeel, (b) plan bracing, (c) diagonal bracing

Tweetear

Consideration of a masted solution arises from the need to provide a greater flexibility in the plan or layout of the building coupled with its aesthetic value to the project as a whole. At the same time it offers the opportunity to utilize structural materials in their most economic and effective tensile condition.The towers or masts can also provide high-level access for maintenance and plant support for services.

The plan form resulting from a mast structure eliminates the need for either internal support or a deeper structure to accommodate the clear span.By providing span assistance via suspension systems the overall structural depth is minimized giving a reduction in the clad area of the building perimeter.The concentration of structural loads to the mast or towers can also benefit substructure particularly in poor ground conditions where it is cost effective to limit the extent of substructures (Figs 5.6 and 5.7). However, differential settlement can have a significant effect on the structure by relaxing ties on suspension systems. The consequent load redistribution must be considered.

Fig. 5.6 Traditional long-span structure

Fig. 5.7 Suspension structure

Most tension structure building forms consist of either central support or perimeter support, or a mixture of the two.Any other solutions are invariably a variation on a theme. Plan form tends to be either linear or a series of repetitive squares.


The forces and loads experienced by towers and masts are illustrated in Figs 5.8 and 5.9. In all cases it is advantageous but not essential that  forces are balanced about the mast. Out-of-balance loads will obviously generate variable horizontal and vertical forces, which require resolution in the assessment of suitable structural sections.

Fig. 5.8 Types of perimeter support.

Fig. 5.9 Central support

Suspension ties must be designed to resist not only tension but also the effects of vibration, ice build-up and catenary sag. Ties induce additional compressive forces in the members they assist. These forces require careful consideration, often necessitating additional restraints in the roof plane in either the  open sections or top chord of any truss.


Longitudinal stability is created either by twinning the masts and creating a vertical truss or by cross bracing preferably to ground (Fig. 5.10).

Fig. 5.10 Longitudinal stability