2 2 2 Design Illustration

Design illustration of highway composite bridges
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  DC Iles page 1 D ESIGN I LLUSTRATION  –   C OMPOSITE H IGHWAY B RIDGES   DC Iles, The Steel Construction Institute, Ascot, UK   Abstract Design of a composite highway bridge to the Eurocodes will require reference to at least 14 separate Parts of the Eurocodes, each with its appropriate National Annex. To illustrate many of the aspects of applying the necessary documents to the design of typical multi-girder and ladder deck bridge configurations, SCI has published a book with two worked examples. This paper presents an overview of some of the design aspects revealed in preparation of those examples. Introduction For the last 20 years, SCI has provided guidance to the designers of composite highway  bridges. General guidance on best practice, based on the views of experienced senior designers, has been accompanied by guidance on the use of design standards, notably, in the  past, on the use of BS 5400. The guidance has been illustrated by worked examples,  presenting detailed calculations, with references to the relevant clauses of the standards. That guidance has now been updated with two new publications for design in accordance with the Eurocodes, one that offers general guidance (SCI publication P356) [1]  and one that presents two worked examples (SCI publication P357) [2] . In P357, one example is a multi-girder bridge and the other is a ladder deck bridge. The preparation of the examples has revealed many of the aspects where design practice differs, to a greater or lesser degree, from that in accordance with BS 5400. This paper presents an overview of those aspects. The Examples In SCI publication P357, Example 1 is a two-span multi-girder deck bridge with integral abutments. The arrangement of the bridge is shown in Figure 1. Example 2 is a three-span ladder deck bridge, with a curved soffit to the main girders in all three spans. The arrangement of the bridge is shown in Figure 2.  DC Iles page 2 Figure 1. General arrangement of Example 1 420002450024500   50050010001000marginalstripmarginalstrip25002500730011700varies1200 to 2200   Figure 2. General arrangement of Example 2 Preamble to Design Verification Documentation The division of the Eurocodes into numerous parts, each representing a separate subject, results in the requirement to refer to at least 14 Parts, possibly as many as 20 Parts, for the design of a composite highway bridge. In addition to EN 1990, for the basis of design, reference is needed at least to Eurocode 1 (Parts 1-1, 1-5, 1-6 and 2), Eurocode 2 (Parts -1- and 2), Eurocode 3 (Parts 1-1, 1-5, 1-8, 1-9, 1-10 and 2) and Eurocode 4 (Part 2). Foundation design will require reference to Eurocode 7. Other Parts will be needed for more complex structures, such as long-span cable-stayed bridges. 50050010001000370037003700marginalstripmarginalstrip200020007300 1100 2800028000  DC Iles page 3 For structures in the UK, the UK National Annexes must also be consulted. Some of these  National Annexes refer to BSI‟s „published documents‟ (PDs), which are one form of “non - contradictory complementary information” (  NCCI). NCCI has no special status according to the Eurocodes, it‟s  merely what it says it is: a text book may well be considered as NCCI; industry is free to produce documents that may be so termed NCCI (as long as they do not contradict). Design basis The categorisation of design situations and combinations of actions to be considered is well set out in EN 1990 and is readily applied for ordinary highway bridges. The terminology used is clear and the use of „ Ed ‟  and „ Rd ‟  as subscripts for design effects and design resistances contributes to clarity in many aspects of the verification procedure. One minor oversight in the UK NA to BS EN 1990 [3]  is the omission of values for partial factors during transient situations (notably construction), although the use of factors for persistent situations should be conservative. Material properties Properties of steel and concrete material properties are clearly defined in EN 1993-1-1 and EN 1992-1-1 respectively. For concrete, the determination of long-term modulus of elasticity and shrinkage depends on the project-specific parameters of relative humidity and age at loading. No doubt, a common practice will evolve on „normal‟ assumptions for these  parameters at opening to traffic and later in the working life. In the examples, where shrinkage effects are favourable they are neglected, rather than using the lesser values at the time of opening to traffic. The use of different coefficients of expansion for uniform temperature change and for temperature gradient will be puzzling to many but is easily dealt with. Design Effects Actions Values for the density of steel and concrete are given in EN 1991-1-1. It is reasonable to use the lower end of the range for steel self-weight (77 kN/m 3 ); the suggested addition of 1 kN/m 3  for the self weight of reinforcement may be rather low for typical reinforcement in deck slabs and the designer should adopt a suitable value. For surfacing, the allowance for an additional thickness of 55% of nominal value given by the UK NA to BS EN 1991-1-1, Table NA.1  is more onerous than the values given by BS 5400. Strictly, the traffic load UDL (in Load Model 1, see EN 1991-2:2003, 4.3.2 ) should only be applied to adverse areas of influence surface and these do not necessarily align with  boundaries between traffic lanes. A more practical approach of applying the UDL over full lane widths, rather than part widths, will make very little difference to design effects. The use of a fixed width of notional lane (rather than dividing the available width into an integral number of lanes) and „remainder‟ widths that are also loaded may seem a little odd but should cause little difficulty in practice. Thanks to the UK NA, the UDL has the same intensity over the full adverse area. Since the UDL does not vary with loaded length, it is simpler to apply than BS 5400 HA loading.  DC Iles page 4 The single vehicle fatigue load model defined in EN 1991-2:2003, 4.6.4  is used in the simplified assessment of fatigue in EN 1993-2:2006, 9.2.2 . Although EN 1993-2:2003, 9.5.2  implies that a spectrum of lorry weights is needed for the simple assessment, no appropriate spectrum is offered in EN 1991-2. The UK NA to BS EN 1993-2 avoids the need for defining a suitable spectrum  by giving an „average‟ weight  for UK traffic (the average is independent of the type of road). It is not entirely clear whether the fatigue load model should be applied in the same notional lanes as for Load Model 1 or in the marked lanes (as BS 5400 required). The reference in the UK NA to numbers of vehicles in „slow‟ and „fast‟ lanes (which ar  e not defined) would indicate the use of the marked lanes. The use of notional lanes would be conservative. Global Analysis According to both Eurocodes 3 and 4 elastic global analysis is to be used for bridge structures, although the UK NA to BS EN 1993-2:2006, NA.2.16  does allow particular projects to specify where plastic analysis would be acceptable (such as for accidental situations). No guidance is available on where plastic analysis might be appropriate. In the elastic global models, both EN 1993-2 and EN 1994-2 require the use of effective widths of wide flanges, allowing for shear lag. Simplified approaches are given. However, if FE models are used, with shell elements for the deck slab, shear lag will be automatically taken into account and no allowance should be made in the model. However, this means that moments on notional composite beams extracted from such a 3D model should be determined from the stresses in the gross flange width, for verification against the resistance of a cross section that includes only the relevant effective width. Traditionally, first-order (small deflection) analysis models have been used and buckling resistance has been determined by reference to empirical rules for „effective length‟. With the increasing power of modern software, elastic buckling analysis of a 3D model is now possible and this offers advantages in some situations. However, the interpretation of output from such software does require experience to ensure that appropriate buckling modes have been identified. The models used in the Examples were 3D FE models (first-order) using shell elements for the deck slab and beam elements for the flanges, stiffeners and bracing members. An illustration of the ladder deck model, for a construction stage with only part of the deck slab, is shown in Figure 3.
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks