Design guide

INTRODUCTION

When prestressing was first introduced into bridgeworks in the 1930s it revolutionised concrete bridges. Prestressing made longer spans and slender decks possible. It is used during the construction and erection of the bridge, as well as in the permanent structure, increasing the load carrying capacity and durability of the concrete.

Today, prestressed concrete is used for simple supported spans a few meters long to cable-stayed bridges with spans of 500m. It has become the material of choice for medium and long-span bridges and viaducts around the world. The following sections describe the features of prestressed concrete bridges. The advantages and disadvantages are highlighted, and key design issues discussed.

OVERVIEW

Prestressed concrete is different from ordinary (non-prestressed) reinforced concrete because the tendons apply loads to the concrete as a result of their prestress force, whilst in reinforced concrete the stresses in the reinforcement result from the loads applied to the structure.

A proportion of the external loads is therefore resisted by applying a load in the opposite sense through the prestressing whilst the balance has to be resisted by ordinary reinforcement.

Prestressing tendons may be internal, i.e. within the concrete, either bonded to the concrete or unbonded, or external, i.e. outside the concrete but (generally) inside the envelope of the member, see Fig. 1

It is possible for external tendons to be outside the concrete envelope; as their eccentricity to the centroid of the concrete section increases, the section behaves more as an extradosed cable-stayed structure than a prestressed concrete member and different design rules are appropriate.

Prestressed members can be either pretensioned, i.e. the tendons are stressed before the concrete is cast around them and the force transferred to the concrete when it has obtained sufficient strength, or post-tensioned, i.e. sheathing is cast into the concrete to form ducts through which the tendons are threaded and then stressed after the concrete has gained sufficient strength. Table 1 compares the advantages and disadvantages of pre- and post-tensioning.

Precast members will generally be pretensioned with tendons bonded to the concrete or used in segmental construction where members are formed from precast segments, which are subsequently stressed together using post-tensioning. Precast segmental structures in the UK currently use external tendons (fib Bulletin 97 External tendons for bridges) because of doubts over the efficacy of the duct joints between segments, although details have been developed to overcome this. In-situ members will be post-tensioned with either bonded or unbonded tendons. Table 2 lists factors affecting the choice of bonded or unbonded tendons.

When a concrete member is prestressed it will deflect and shorten. If the tendon profile is such that the deflected shape of the isolated member is compatible with the restraints acting on the member, the profile is said to be concordant. This will always be the case for a statically determinate member. However, tendon profiles in statically indeterminate members will not generally be concordant.

Table 1. Comparison of pre- and post-tensioning

Type of construction	Advantages	Disadvantages
Pretensioned	No need for anchorages, tendons protected by concrete without the need for grouting or other protection, prestress is generally better distributed in transmission zones	Heavy stressing bed required, more difficult to incorporate deflected tendons
Post-tensioned	No external stressing bed required, more flexibility in tendon layout and profile draped tendons can be used easily	Tendons require a protective system, large concentrated forces in end blocks

Table 2. Comparison of bonded and unbonded construction

Type of construction	Advantages	Disadvantages
Bonded	Tendons are more effective at the ultimate limit state, does not depend on the anchorage after grouting localises the effect of damage	Tendons cannot be inspected or replaced, tendons cannot be re-stressed once grouted
Unbonded	Tendons can be removed for inspection and are replaceable if corroded, reduced friction losses generally faster, construction tendons can be re-stressed, thinner webs	Less efficient at ultimate limit state, relies on the integrity of the anchorages and deviators effects of any damage are more widespread, less efficient in controlling cracking

When the tendon profile is concordant, the only forces induced at any point in the member by the action of prestressing are an axial compression equal to the prestressing force and a moment equal to the product of the prestressing force and its eccentricity relative to the neutral axis of the member. These are the primary prestressing forces.

When the tendon profile is non-concordant, additional forces and moments will be induced in the member during prestressing by the restraints acting on it. These are known as the secondary or parasitic effects. Other common nomenclature associated with prestressed concrete is defined in Fig. 2.

Concrete is stronger in compression than in tension. Prestress is introduced to pre-compress the areas of concrete, which would otherwise be in tension under service loads. The concrete section is therefore stronger and behaves more as a homogeneous section, allowing elastic methods of analysis to be used, although the concrete compressive stresses can be high.

When the tendons are bonded to the concrete, their high ultimate strength can be mobilised, which generally means that the ultimate flexural capacity of a prestressed concrete member is much greater than the applied ultimate design moment. Therefore, limiting the maximum compressive stresses and crack widths under service loads is generally critical in the design of prestressed concrete members, although this may not be the case for members with unbonded tendons or members with bonded tendons and large allowable crack widths.

Therefore, flexural design is normally carried out at the serviceability limit state and then checked at the ultimate limit state. Shear design is carried out at the ultimate limit state.

The prestressing force applied to the concrete immediately after tensioning and anchoring (post-tensioning) or after transfer (pre-tensioning) will be less than the jacking load due to one or more of the following:

• elastic deformation of the concrete
• losses due to friction, and wedge-slip in the anchorages.

The value of the prestressing force will continue to reduce with time due to:

• relaxation of prestressing steel
• creep of the concrete
• shrinkage of the concrete.

It is normal to check the flexural stresses in a prestressed concrete member, both when the prestress force is initially transferred to the concrete (at transfer), taking account of the initial losses, and in service, after all losses have occurred. Prestress can be considered as a load or as a resistance. At the serviceability limit state, it is normally considered as a load whilst, at the ultimate limit state, it is considered as a combination of a load and a resistance.

When considered as a load, the effects of prestress can be determined by analysing the structure under a system of equivalent loads representing the forces from the prestressing tendons acting on the concrete. Such an analysis automatically accounts for primary and secondary effects.

When prestress has been considered as a load, the contribution of prestressing tendons to the resistance of a section is limited to their additional strength beyond prestressing. This can be calculated by assuming that the origin of the stress-strain relationship of the tendons is displaced by the effects of prestressing.

For bonded tendons, this is illustrated in Fig. 3. The origin of the stress-strain relationship is taken as being at point A, corresponding to a prestress force, Pt and the contribution of the tendons to the resistance of the section is Δf_pA_p. When the whole of the prestress is considered as a resistance, the origin is taken as point B.

For unbonded and external tendons, the additional strain in the tendons, Δε_p is determined by accounting for the deformations of the concrete member.

Partial prestressing can also be used in some circumstances where prestress is provided to counter some of the stresses in the concrete and reinforcement provided to limit crack widths. The approach in Eurocodes is to ensure that the concrete around the prestress tendon is always in compression while tension is allowed outside of that provided the cracks widths are kept within the allowable limits specified.

The design of prestressed concrete bridges is described in a number of standard texts, some of which are listed in the books and references section of the CBDG website. The following sections give some advice and information on particular aspects of the design of prestressed concrete bridges.

THE DESIGNERS PERSPECTIVE

For the bridge designer, prestressed concrete provides one of the most versatile materials to design and construct with. Unusual shapes and structural arrangements may be readily adopted with either in-situ or precast constructions, while using prestressed concrete allows the designer to adopt slender and longer spans than would be possible with just reinforced concrete construction.

Prestressing is used in many different ways in bridgeworks. It can be in the form of longitudinal tendons to cater for the longitudinal moments and shears in a deck or column, or as transverse tendons prestressing the top slab of a deck to cater for the local wheel loads. Occasionally, it is used as vertical prestress to enhance the deck web shear capacity or to distribute forces within the deck diaphragms.

Often contactors will use the prestressing in the temporary condition to support the bridge during the construction stages. For the innovative designer, its uses are endless, although it must be said that generally British designers do not use prestressing as often as some international designers.

The concept of prestressing is simple in that it is just an external load applied to a section to balance the tensile stresses that occur. However, in practice the design is more complex than for simple reinforced concrete and the interaction between the structural behaviour, the concrete properties and the prestressing tendons has to be carefully considered by the designer.

The prestressing force causes the concrete to creep while the creep and shrinkage in the concrete reduces the force in the prestress. The prestress secondary moments, sometimes referred to as parasitic moments, may also be redistributed by creep within the concrete and the designer has to combine all these effects into the design.

With prestressed concrete bridges, the designer must always consider the construction sequence and the way the bridge is going to be built. This quite often has more influence on the prestress tendon arrangement and profile than do the forces and moments applied to the sections. With balanced cantilever construction, the prestress tendons are arranged in groups of cantilever and continuity tendons. With span-by-span construction, the prestress tendons are usually anchored on the construction joints and often have couplers to extend them into the next span cast. If a bridge deck is fully cast before the prestress is applied then the prestress may consist of single tendons extending from one abutment to the other.

When choosing which type of prestressing to use, whether it be strand, wires, bars, internal or external, the designer's decision is usually based on economics, ease of construction and maintenance. Strands are usually the cheapest type of prestressing tendon when considered in terms of £ per kN of force, while prestressing bars can be simpler to install, especially for short lengths. External tendons can simplify concreting and reinforcement fixing as well as future inspection and maintenance, but internal tendons are usually less expensive and offer a greater level of protection against accidental damage.

The design process that the designer uses for concrete bridges depends on the type of prestressing used. For internal tendons, the design of the prestress is usually governed by the SLS conditions. The prestress quantity and layout need to ensure that the stresses within the structure are within the acceptable limits. In this case, the ultimate moment capacity is not usually critical while the shear design is usually a case of having sufficient concrete width and depth, and adequate reinforcement. With external tendons the longitudinal design may be governed by the ULS condition. The amount of prestress needed at any section is dictated by the ultimate moments applied with the SLS stresses usually less critical.

Designers have a range of software available to them for analysis of prestressed concrete bridges. To fully analysis all the aspects of prestressed concrete, the software package must be able to handle the stage-by-stage construction that inevitably occurs with this type of construction. Creep and shrinkage of the concrete and friction, draw-in and other prestress losses should also be modelled within the analysis.

Software programmes that combine all of these aspects include ADAPT, RM Bridge, Sofistik, LARSA, Midas, and others. As the power of computers increase, even the most complex of structures can be analysed in a short time. The designer should always be aware that all of these programmes have limitations and a certain amount of interpretation and approximation is still needed to end up with the right answer.

As with all forms of bridgeworks, prestressed concrete has its own set of detailing rules that need to be carefully followed.

Space within the concrete is needed for the anchorages and couplers and the extra reinforcement associated with these areas. The required duct spacing and cover is dependent on the force in the tendon and the tendon radius. The structure must also provide sufficient room to locate the jack and stress the tendons. Ducts clashing with reinforcement can challenge both the detailer and site team. There is usually a need to draw up the prestressing tendons and reinforcement in detail to ensure that it all fits within the concrete section, allowing room for the placing and compaction of the surrounding concrete.

The designer should not forget the need for future inspection and maintenance of the prestressing tendons. Where external tendons are employed, there should be good access for inspection and for moving the equipment for restressing and grouting around.

One of the main advantages of prestressing is the improved long-term durability of the structure by reducing or eliminating cracking within the concrete and a designer can make use of this in vulnerable parts of the bridge.

Where a project warrants the use of precast concrete, prestressing often simplifies the construction details across the connections. By prestressing the connection, the need for in-situ connections and complex reinforcing details can be minimised, or eliminated.

The longest free span in the world for a concrete box girder is currently Stolma Bridge in Norway, which has a main span of 301m. The Shibanpe Bridge in China has a main span of 330m which is a combination of concrete and steel boxes. This shows that prestressed concrete can rival steel and arch construction for spans up to this length, while for cable-stayed bridges it is common to find prestressed concrete being the material of choice for the deck with spans up to 500m.

INTRODUCTION

Reinforced concrete can be found in the majority of bridges in the form of foundations, piers, abutments or deck slabs. For short spans of less than 20m and where access is available to erect falsework, reinforced concrete provides an economic, efficient solution for any bridge deck.

The design of reinforced concrete involves the consideration of both the Serviceability Limit State and the Ultimate Limit State to ensure that the bridge satisfies its performance requirements and has adequate capacity to carry all the loads applied.

The following sections provide guidance on the design of reinforced concrete while essential references where the designer can find more extensive information on reinforced concrete bridges can be found in the references section of the website.

PURPOSE

The rationale behind reinforced concrete (RC) design is to provide sufficient information for production of drawings that will form the basis of the construction information of a structure. In so doing the design will satisfy the codified requirements in terms of both ultimate and serviceability limit states.

There are some aspects of RC design that are outside the scope of the main design standards and for these additional references are required. Notable aspects that fall into this category are deep beams and the end block design of complex post-tensioned concrete beams.

THE DESIGNER'S PERSPECTIVE

Reinforced concrete is probably the most common construction material in the world and the concepts of reinforced concrete design are well known to most designers. It is only the way that the different codes approach the derivation and detailing of the reinforcement required that need to be learnt. Most current design standards for concrete bridges, such as Eurocodes (superseding BS 5400), and AASHTO adopt a limit state approach. The loads (actions) acting on the structure are factored upwards to account for uncertainties in their magnitude due to dynamic effects, future increases or the possibility of overloading, whilst the strengths (resistances) of concrete and reinforcement are factored downwards to allow for variations in material properties within the ranges permitted by the various specifications and also the uncertainties inherent within the strength calculations. The codes generally consider Ultimate and Serviceability limit states in recognition of the need to produce both a safe structure and one that fulfils its function throughout its service life without the need for major maintenance or repair.

ULTIMATE LIMIT STATE (ULS)

At the ultimate limit state, the designer is concerned with the strength of the structure to prevent collapse under the factored loads, taking due account of the possibility of buckling or overturning. The codes aim to ensure the ductility of reinforced concrete members by preventing over-reinforcement of sections that would result in brittle failure via crushing of concrete, rather than yielding of reinforcement. The designer must ensure that sufficient flexural and shear reinforcement is provided at every section along a beam, slab or column to resist the ultimate applied loads. In order to produce an economic overall design, the designer must balance the opportunities to optimise the reinforcement provided at a section along with the desire to produce simple buildable details.

Achieving this balance is more a matter of experienced judgement than following a series of rules. The advent of computer software for the design of reinforced concrete sections allows the designer to quickly evaluate a large number of sections. Alternatively, most of the design formulae are readily incorporated within spreadsheets.

In order to determine the curtailment of reinforcement, it is usual for designers to overlay the ultimate bending moment and shear force diagrams with the design strengths provided by the reinforcement proposed at the different sections. This technique provides a pictorial comparison between design forces and member strengths, as well as assisting the designer in producing sketches for RC detailing. In determining the reinforcement requirements for sections over pier supports, it is usual to round off the peak values obtained from analysis in recognition that the support has a finite width. A number of techniques exist, but a common method is to calculate a reduction in the peak moment from analysis, M_r = (P x d)/8, where P is the reaction at the support and d is the pier or deck diaphragm width.

SERVICEABILITY LIMIT STATE (SLS)

The aim of the serviceability limit state is for the designer to produce a design that limits deflections of the structure to acceptable limits during normal loading and prevents deterioration over the service life of the structure that would necessitate major repair works.

For reinforced concrete in bridgeworks, the primary concern at SLS is the control of flexural crack widths. All reinforced concrete members crack at their tension face when under load as is necessary to permit strain in the reinforcement. However, when these cracks become excessively large, they permit the ingress of water into the cover zone of the concrete and thus to the steel reinforcement. This is a particular problem in temperate climates where de-icing salt is used on the roads during the winter months and thus the water run-off is particularly aggressive in corroding the reinforcement. In order to prevent this eventuality, the codes limit the crack widths permitted under SLS loads depending on the cover provided and the vulnerability of the section in terms of its location.

In designing reinforcement to control cracking at SLS, the designer should be aware that reinforcement spacing and distribution is an important factor in controlling crack widths. It is often more effective to reduce reinforcement spacing than to increase the bar size. The designer should include for the effects of coexistent axial forces in a section when evaluating crack widths. Where compression exists, e.g. in a bridge pier, this has a beneficial effect in reducing the amount of reinforcement required.

It should be noted that in accordance with the UK National Annex of Eurocodes, crack widths should be calculated at the nominal, rather than the actual cover provided. This is an important distinction, especially since the design includes for additional cover to be provided over the nominal values provided to cater for the construction and environmental conditions.

Some design software is able to explicitly calculate crack widths around the perimeter of a section; however, care is needed to ensure that the results are correctly interpreted, particularly with respect to the difference between nominal and actual cover at corners.

EARLY THERMAL AND SHRINKAGE EFFECTS

In addition to the strength and durability requirements at ULS and SLS, the designer must also consider early thermal and shrinkage effects in determining the member reinforcement requirements. Early thermal cracking arises from the restraint provided internally within the concrete, or by a previously cast section of concrete or blinding. As with flexural cracks, it is not possible to always prevent thermal cracks from occurring only to control their size and distribution. National Highways (was Highways England) guidance document CD350 The design of highway structures refers to CIRIA publication C766 Control of cracking caused by restrained deformation in concrete which can be used for the design of reinforcement to control early thermal cracking

STRUT AND TIE METHODS

There are a number of instances where it may be appropriate and economic to employ 'strut and tie' methods of design. These are well covered within the concrete structures design section of Eurocodes and can also be found in other international codes, including the American highway bridge code, AASHTO. The 'strut and tie' method is suitable where a concentrated force is being distributed to a support that is within 2d of the point of application. The force is transferred by means of an inclined concrete strut between that point of application and support in conjunction with a tie force provided by the reinforcement. A typical example of this type of system is for the design of a simple pile cap, where the bottom mat reinforcement in both directions acts as the tie to permit transmission of the vertical loading through an inclined concrete strut from the column to the piles.

REINFORCEMENT DETAILING

The importance of good reinforcement detailing in producing a buildable bridge structure should not be underestimated. It is not satisfactory for the designer to simply leave this important aspect to draughtsmen. The two must work together to produce RC details that implement the design intent in the most efficient manner possible.

The most common problems with detailing are associated with over-congestion of reinforcement not permitting the proper placing and compaction of concrete. In this situation lack of thought at the design stage can result in the construction (and demolition!) of defective members on site. Over congestion can be avoided by staggering of laps, layering of reinforcement and the use of more efficient bar shapes. Where over congestion is possible, it is often worthwhile drawing the details to scale, including allowances for the ribs on deformed reinforcement. In this manner it is possible to identify whether there is sufficient clearance between bars to comply with the minimum code requirements.

In detailing the RC member, the designer should work 'bottom-up' as corresponds with the actual fixing sequence on-site.

The designer should also consider the method by which the reinforcement is likely to be fixed on-site, to ensure that the bar shapes proposed do not unnecessarily hinder the fixer's work. For example, it is often better to use pairs of lapped 'U-bars' rather than closed links for shear reinforcement in beams as it allows easy access from the top to place the longitudinal reinforcement. It is worth bearing in mind that since labour forms a significant proportion of the contractor's costs, a more buildable bridge is often a cheaper one.

Reinforced concrete is an inexpensive readily available construction material. When used properly, it produces durable structures capable of carrying a range of loads, from cars and trains to pedestrians, cyclists and equestrian traffic. It seldom experiences vibration or fatigue problems and is easy to design and detail.

INTRODUCTION

Bridges, like most structures, start in the ground; there are very few bridges that do not have a substructure to support them. Often described as everything below deck level, the substructure is required to support heavy loads from above with long-term durability and minimal maintenance, making concrete the ideal material for these parts of the bridge.

Cast in-situ or precast the concrete can be constructed to any required shape, texture or colour although in the past designers and clients have not always made the most of this attribute.

The substructure located below the deck soffit or bearings usually includes the piers, abutments and foundations. These elements transfer the load from the deck down to the ground and, in the case of abutments, provide a transition from the supported deck structure to the approaches. Arches and the pylons and towers of cable-stayed and suspension bridges are an integral part of the superstructure of a bridge and are not considered in this section, although they have many aspects in common with substructures in general. See Figure 1 for terminology.

Piers are used to support the superstructure along the length of the deck and may be built integral with the deck or incorporate a bearing, to allow relative movement between the pier and deck to occur. The piers support the deck during the construction and in service. Most bridges are subjected to high vertical loads combined with significant horizontal forces and moments.

The abutments may also be integral with the deck where anticipated movements are accommodated within the structural design.

Foundations are usually pad footings or piled arrangements where the ground conditions are poor and suitable founding strata does not occur near the surface.

This makes concrete an ideal material for these parts of the bridge as it combines inexpensive construction with high capacity, robustness and relatively maintenance free attributes.

The different parts of the substructure and their design are discussed in more detail in the following sections.

PIERS AND COLUMNS

Piers and columns occur in many different shapes and arrangements, which are often due to the choice of the designer and governed by their appearance, as well as their function to support the deck and suit the at-grade constraints. Simple circular columns often provide efficient supports, while wall piers give a high resistance to impact loading and good lateral stability. The following images illustrate some of the different forms that piers often take:

Where the deck uses precast beams the piers usually include a cross girder or pierhead to support the beams. With this arrangement the pier may be constructed integral with the deck or with bearings under each of the beams. An example of this type of arrangement is illustrated in the following image.

Foundations often have to be arranged to suit the at-grade restraints, avoiding existing services or other obstruction below the bridge. This can lead to unusual arrangements with the substructure offset from the deck, or with portals to span over the obstruction. The following photographs show two different arrangements frequently used to avoid at-grade obstructions.

The examples below show piers and columns both integral with the deck or with bearings to allow the deck to move relative to the substructure.

Integral construction has become fashionable in recent years and offers the great advantage in eliminating the maintenance associated with bearings. However, using bearings can reduce the forces imposed on the substructure and simplify construction in certain circumstances.

ABUTMENTS

The abutments provide the transition from the elevated structure to the approach embankment or approach structure. As well as supporting the end of the deck, they usually retain the fill of the embankment and provide means of access into the deck where applicable. There are many different types of abutments, including:

• wall abutments
• bank seat abutments
• piled abutments
• mass abutments
• counterfort abutments
• reinforced earth abutments.

The characteristics of these different types of abutments are described in the BRE report Bridge foundations and substructures, which provides guidance on their use and arrangement.

For decks lengths of up to 60m, it is now usual practice to make the abutments integral with the deck, although with skew alignments or box girder decks, bearings are often provided to simplify the overall structural behaviour. For longer lengths of deck, the expansion and contraction movements may dictate the need for a non-integral arrangement with bearings to minimise the loading imposed on the abutment. Where bearings and expansion joints occur at the abutments, a suitable size access chamber is required to facilitate the inspection and future replacement of these items. The following images and sketches show some typical abutment arrangements.

Back-fill to abutments is traditional a granular material, or 6N as it is classified in the UK. The granular material allows a reasonable compaction to be achieved, while maintaining good drainage characteristics. A positive drainage system should be provided to the back of an abutment to remove any water as quickly as possible. Where an abutment has bearings and a bearing shelf or inspection chamber, this must also be well-drained to prevent any build-up of standing water which could be detrimental to the durability and appearance of the structure.

RUN-ON (OR APPROACH) SLABS

Should run-on slabs be provided? This is one of the most contentious issues for experienced bridge engineers in the UK and the debate is likely to continue with run-on slabs generally not used in the UK, but commonly used internationally. A run-on slab provided behind an abutment gives a transition between the rigid abutment wall and the softer embankment behind, spanning over any poorly compacted fill next to the abutment, although they have been subjected to problems in the past where poorly designed.

Without run-on slabs, settlement of the backfill behind an abutment leaves a bump in the road until the road is resurfaced next. With run-on slabs any settlement is taken out over the length of the slab, however if there are any problems with them or the embankment below, the remedial works can be a lot more expensive.

The use of run-on slabs is the choice of the bridge owners and their designers and will depend on many factors, including the maintenance regime, construction quality and design requirements. Designed properly, run-on slabs have proven successful on numerous bridges in the UK and around the world, while many bridges without run-on slabs are currently in use in the UK.

PILED FOOTINGS

Where the founding strata is at depth below the ground surface, piles are often used to transfer the load from the substructure into the ground. A pile cap connects the top of the piles and distributes the forces and moments from the piers or abutments to the individual piles.

There are many types and sizes of piles used in bridgeworks, from large diameter bored concrete piles with diameters of up to 3m, to small mini-piles of a few mm diameter. They can be precast and driven or cast in-situ concrete, and the choice depends largely on the ground conditions, loads imposed, access and equipment available.

Tomlinson and Woodward's Pile design and construction practice provides comprehensive guidance on the different piling systems available and their design and construction requirements.

Pile caps are designed using a truss analogy or by simple bending theory to derive the tensile reinforcement and shear requirements. Guidance on the design of pile caps is given in the Eurocodes BS EN 1992-1-1 and BS 5400 Part 4 clause 5.7.3 Design of bases.

PAD FOOTINGS

These are used where good founding material is near the ground surface. Simple pad or spread footings are usually the most cost effective and simplest foundations to construct. The concrete pad is used to spread out the load from the piers or abutments above so that the bearing pressures on the soil beneath are within acceptable limits.

Guidance for deriving the presumed allowable bearing values for different soils and rocks are given in BS 8004, the British Standard Code of practice for foundations, which can provide an initial guide for the design of the pad footings. Boreholes and testing of the soil at a particular site are needed to determine the allowable pressures for the detailed design of a foundation, and guidance on the site investigation techniques available is given in CIRIA's publication, Site Investigation Manual.

CAISSONS

Although not a common form of foundation, for larger bridges and when working over water, the use of a caisson can simplify the construction process. Caissons are more popular in countries where large piling equipment is not always readily available, such as in India where they are called wells.

Where the foundations are over water with sufficient depth, the caissons may be precast and floated or lifted into position. This technique has been used on a number of major bridges such as the approach viaducts for the Second Severn Crossing.

Alternatively, the caissons are cast in stages above their final location and sunk into the ground through a combination of excavating inside the void and adding additional weights.

DESIGN ASPECTS

The design of the elements making up the substructure of a bridge is similar to any reinforced or prestressed concrete member and more information on this is provided under the appropriate sections elsewhere in this guide. Discussed below are some additional topics specific to the design of bridge substructures.

STANDARDS

Design of the substructure for a bridge in the UK is generally carried out to Eurocodes, the UK National Annex and the associated Highways England standards and guidance notes or Network Rail standards. These standards provide rules for the design of the reinforced concrete sections with additional rules and guidance for the specific structural element being considered. Further British Standards and Highways England standards provide additional guidance for the different structural elements and types of structure and for general serviceability and durability requirements.

LOADING

Piers, columns and abutments are generally subjected to loads from the deck, plus loads applied directly to the element being considered, such as vehicular impact, earth pressures, wind pressure or water flow if in a river. Where the substructure is in a seismic region, additional forces are generated through soil movements and dynamic effects. The seismic design for substructures is a specialist subject which should only be undertaken by engineers with appropriate experience in this field.

The loading for highway bridges in the UK is defined in Highways England's departmental standard CD350 in conjunction with the Eurocodes. Within these documents the loads applied to both the deck and the substructure for UK bridges are defined.

For rail bridges in the UK the design and loading requirements are outlined in the Network Rail Standard NR/L3/CIV/020.

Scour must be considered where the substructure is located in a river or on a seabed and accounted for in the design. Scour and designing for the effect is dealt with in CIRIA Report C551 Manual on scour at bridges and other hydraulic structures.

In general, either steps are taken to prevent the scour, such as the provision of rock mattresses, or the structure is designed for the condition after the predicted scour has occurred.

Thin columns and walls may attract additional moments due to slenderness effects and construction tolerances. A thin tall column needs to consider buckling with the moments adjusted accordingly. Short columns do not need to consider buckling but the construction tolerance needs to be taken into account by considering an additional moment equal to the axial load acting eccentrically on the cross-section depth.

ANALYSIS

Where the loads are applied to the deck, a structural analysis is usually required to determine the distribution of the load between the different substructure elements. This is particularly the case where the substructure is integral with the deck and the load distribution between the deck, piers and abutments depends on the structural arrangement and relative stiffness of the different elements. Columns and piers are frequently subjected to biaxial bending in combination with a range of axial loads.

For most spread or pad footings for bridgeworks, the load distribution is usually based on assuming the footing is rigid with a linear distribution of pressure beneath. In these circumstances the pressure distribution may be derived using the formulae given Reynolds et al Reinforced Concrete Designer's Handbook, Table 192.

With piled foundations the derivation of the forces and moments in the individual piles is usually carried out using proprietary software such as Repute, Mpile or similar. Simple bolt-group theory can be used as an initial estimate of the vertical load distribution between the piles, but a more sophisticated analysis is usually required for the detailed design and to quantify the moments down the pile.

CONCRETE

Concrete mix design for substructures depends on the location, soil conditions and the design strengths required. Typical concrete strength classes for substructure elements are:

• Piers; C32/40 or C40/50
• Abutments and wing walls; C32/40 or C40/50
• Foundations and pile caps; C32/40
• Bored concrete piles; C32/40

The concrete cover to the reinforcement depends on the exposure of the concrete and the environmental conditions. Guidance on the cover required is given in Eurocodes, however the cover used is often greater than this to account for construction tolerances and durability requirements.

Typical minimum concrete cover used for substructure elements is:

• Piers = 40 or 50mm
• Abutments and wing walls = 40 or 50mm
• Foundations and pile caps = 50mm
• Bored concrete piles = 75mm

Early thermal cracking in the concrete is sometimes a problem for substructures where the large concrete sections require special measures to control the initial temperature rise and limit cracking. Additional reinforcement is often required to control early thermal cracking and guidance on this is given in CIRIA publication C766. Other measures are also frequently employed to reduce the early thermal effects, including adjustments to the concrete mix, use of fly ash (fa), ground granulated blast furnace slag (ggbs) or similar secondary cementitious material, reduction to the initial temperature of the concrete and control of the subsequent temperature increase by cooling or staged concreting.

Foundations and parts of the substructure below ground are usually waterproofed or protected in some way. The degree of protection will depend on the soil conditions and vulnerability of the concrete. This protection may consist of simple bitumen painting or extend to a full waterproofing membrane where full protection is needed.

THE DESIGNER'S PERSPECTIVE

The substructure of a bridge often represents between 50% and 70% of the total structure cost. It is also the area presenting the greatest risk to the client or contractor due to the unknowns associated with the foundation material. The bridge designer can contribute significantly to producing an economic design and mitigating these risks through careful planning, design and detailing. There are a number of key factors which the designer should consider:

Adequacy of site investigation

The foundation design will only be as reliable as the information on which it is based, namely the soil parameters that are derived from the site investigation. Where the bridge designer is responsible for specifying the site investigation, the minimum requirements for the design should carefully consider. This will largely depend on the consistency of the ground, but as a minimum would include one borehole at each pier or abutment location. Where ground conditions are variable it may be necessary to specify several boreholes at each foundation location or even at each pile location.

The depth of any boreholes should also extend to a level sufficiently below the anticipated founding depth of the piles or spread footings. The in-situ and laboratory testing undertaken should be sufficient for the designer to understand both the physical and chemical properties of the founding material.

It is often tempting for clients or contractors to seek to cut back on the site investigation to be undertaken in order to reduce costs. This can be a false economy where the up-front savings made are outweighed by uneconomic foundation designs or the costs associated with redesigns during construction as more information becomes available.

Upper and lower bound design

On completion of the site investigation, the designer will assess the soil or rock strength measures that have been derived from in-situ or laboratory testing in order to derive a set of design parameters. A common method would be to plot SPT-N values against depth. In determining a 'design line' from this data, the designer must decide whether a 'best-fit' approach is adequate, or whether the design of the structure will be sensitive to the variability of soil strength or stiffness. Where the latter is true, it will be better for the designer to adopt upper and lower bound design lines in order to represent the potential variability of the soil-based on the scatter of data derived from the site investigation.

This approach is recommended when designing substructures that are integral with the deck and where the soil stiffness contributes to the distribution of forces within the structure. In this case any structural analysis would be undertaken on the basis of both upper and lower foundation stiffnesses and the worst case taken. Using this method, the sensitivity of the design to the soil parameters can be investigated, understood and accommodated in the final design.

Stiffness versus strength

It is increasingly commonplace for substructures, both piers and abutments, to be constructed integrally with bridge decks. As discussed in the other parts of this section, this has the advantage of eliminating bearings and expansion joints and thus the associated long-term maintenance liability.

Integral bridges are also more efficient in the distribution of external forces between the superstructure and substructure, thus usually resulting in a more economic overall design. However, the main difficulty with integral bridges is the accommodation of internal strains due to creep, shrinkage and temperature effects. These cause contraction and expansion of the bridge deck and thus impose lateral movement on the piers and abutments. The magnitude of forces generated by these movements is governed by the global substructure stiffness, including contributions from both the soil and structure itself.

As substructure stiffness increases, so do the forces generated by these internal strains. It is therefore important for the designer to balance stiffness and strength in order to achieve an economic design. Experience has shown that the natural inclination of the designer is to increase structure stiffness during design iterations in an attempt to increase the strength of a particular member.

The designer should consider whether a better balance might be obtained by reducing substructure stiffness in some manner and thus reduce the forces generated.

A considerable amount of work has been undertaken on the effect of cyclic movement of abutment walls into the retained soil. This cyclic movement, due to thermal expansion and contraction, generates significant pressures on the abutment wall. In this situation a balance is required between minimising the soil angle of friction that gives rise to the soil pressures and controlling backfill settlement.

Ground improvement

There are a large number of very effective ground improvement measures available, including soil replacement, vibro-compaction, stone columns etc. The designer should therefore consider whether ground improvement would be an economic solution as compared with a more expensive foundation solution. For example, where a surface layer of weak soil is encountered, it may be more economic to improve or replace this material and adopt a spread footing than to use piles. Clearly any evaluation of alternative options must include for the disposal costs of unsuitable material as these are becoming very significant in many countries.

Wingwalls

The wingwalls required to retain the soil around bridge abutments can be a significant proportion of the cost of the substructure, particularly if they have to be supported on separate foundations. The designer should therefore give considerable thought to the means of reducing the length and hence cost of abutment wingwalls. The size and orientation of wingwalls also has significant influence on the aesthetic impact of the bridge (See CBDG report The Aesthetics of Concrete Bridges). It may be appropriate to consider the use of steepened backfill slopes, either using soil nailing or reinforced soil in order to reduce the length of wingwalls. Both these methods have been found to produce overall economies on bridges in the past.

Irrespective of the amount of site investigation undertaken, the construction of bridge foundations carries inherent risks due to the possibility of unforeseen ground conditions or buried services. Within this context the bridge designer should produce the most economic substructure design that the site constraints and desire to mitigate construction risk will permit.

INTRODUCTION

Bridges are subjected to some of the harshest conditions of any structure. They are battered by the environment, sprayed with salty water and pounded by traffic. It is therefore imperative that the design, construction and specification for concrete bridges take durability into account from the start.

Concrete bridges were first built about 100 years ago, but have evolved a great deal since then. Modern concrete bridges cover a wide range of types and arrangements. The current standards, codes, specifications and design guides encapsulate the growing experience gained through the maintenance of our current bridge stock.

The following sections provide guidance on how durable concrete bridges are achieved and in particular the requirements for specifying concrete. It is recommended that you are familiar with the relevant standards e.g. BS 8500 Concrete-Complementary British Standard to BS EN 206

DESIGN/INTENDED WORKING LIFE

In BS EN 1990 Eurocode-Basis of structural design, the term used for the "intended working life" is the "design working life". The terms may be treated as being synonymous (BS 8500-1, note to A.4.1). The theoretical design life for a bridge designed in the UK is 120 years.

BS EN 1990 gives indicative design working lives (in Table 2.1) for design purposes for various types of structures, as follows:

Category 1 - Temporary structures, not including structures or parts of structures that can be dismantled with a view to being re-used - 10 years
Category 2 - Replaceable structural parts, e.g. gantry girders, bearings - 10 to 25 years
Category 3 - Agricultural and similar buildings - 15 to 30 years
Category 4 - Building structures and other common structures - 50 years
Category 5 - Monumental building structures, bridges and other civil engineering structures - 100 years

The UK National Annex to BS EN 1990:2002 gives modified indicative design working lives for some of the Categories as follows:

Category 2 - 10 to 30 years
Category 3 - 15 to 25 years
Category 5 - 120 years

Presumably the Category 5 change to 120 years was to bring the recommendations into line with the requirements of the Highways England. This was covered in IAN 95/07, although this has now been withdrawn.

Highways England now refer to BS 8500 Concrete for limiting constituents for durability, i.e. based on at least 50 or 100 years intended working life.

Prior to the introduction of the Eurocode a design life of 60 years was required for buildings, though this period was never included in any of the structural design Codes. For concrete bridges up until 1978, it had always been a 100-year design life. 100 years does not appear to be explicitly included in prior codes, it just seems to be an expectation for that type of structure. The design life figure went up to 120 years with the introduction of BS 5400 Steel, concrete and composite bridges Part 2 Specification for loads in 1978 and Part 4 Code of practice for the design of concrete bridges in 1984.

For most highway structures, 120 years is considered appropriate by Highways England as an intended working life. In practical terms it is not possible to distinguish between concrete quality for an intended working life of 100 years and 120 years.

REINFORCED AND PRESTRESSED CONCRETE

Specifying concrete

European Standard BS EN 206 Concrete applies to concrete for structures cast in-situ, precast structures, and structural precast products for buildings and civil engineering structures. Other European Standards for specific products or for processes within the field of the scope of this standard may require or permit deviations e.g. BS 6349-1-4:2021 Maritime works. General - code of practice for materials.

BS EN 206 defines the term "specification of concrete" as the "final compilation of documented technical requirements given to the producer in terms of performance or composition". It also recognizes that this specification can contain requirements for a number of different persons or bodies, e.g. architects, structural designers, contractors. The architect and/or the structural designer compile the majority of the specification, but there are important aspects that have to be added by the user, e.g. a consistence that is suitable for the intended method of placing and finishing the concrete. The term "specifier" is reserved for the "person or body establishing the specification for the fresh and hardened concrete" i.e. the person or body who gives the specification to the producer.

Durability

In the UK, BS 8500 is the complimentary standard to BS EN 206, giving the National provisions. Part 1 of BS 8500 describes methods of specifying concrete and gives guidance for the specifier. Annex A (informative) provides guidance on the concrete quality to be specified for selected exposure classes, intended working life and nominal cover to normal reinforcement. It does not give guidance on stainless steel and non-metallic reinforcement. Guidance on nominal cover to reinforcement for structural (prestressed) and fire consideration is available in other publications, e.g. structural design codes of practice.

Relevant exposure classes need to be identified; there will always be one and frequently more than one exposure class for each element face. The classes are;

X0 No risk of corrosion or attack
XC Corrosion induced by carbonation
XD Corrosion induced by chlorides other than from sea water
XS Corrosion induced by chlorides from sea water
XF Resistance to freeze-thaw attack
XA Resistance to chemical attack

BS EN 206 gives description of classes in Table 1 and the limiting values of composition are given in Annex F, Table F.1. This gives no indication of cement type, intended working life and limited information in Table 2 on sulfates (XA). Hence the UK complimentary standard BS 8500 and BRE in Special Digest 1 Concrete in aggressive ground.

The guidance in BS 8500-1 Annex A gives the limiting values of composition and properties of concrete, when using a particular maximum size of aggregate to provide acceptable durability for each identified exposure class. The limits relate to cement type and content, water cement ratio, and cover to reinforcement commensurate to the environmental action (exposure class) for an intended working life of 50 or 100 years.

Cement type

There is a plethora of cement types and combinations which are covered in BS 8500-1 Table A.6 and BS 8500-2 Table 1. Determining the limiting factors of w/c ratio, cement content and cover etc. for durability is dependent on the cement type.

The nomenclature relates to whether the cement is factory produced (CEM) or a combination of powers batched at the concrete batching plant (C) followed by the composition i.e percentage range (II to IV) and type of addition (L, V, S, D, P, Q) to the Portland cement (CEMI) or Portland limestone cement (CEM II/A-L(LL)). These are also known as secondary cementitious materials (SCM). For example, CIIA-S is a combination of Portland cement with 6 to 20% ggbs. Certain composite and combination are considered sulfate resisting (SR). The following is a list of designations.

L Limestone powder
L-L Limestone powder (low total organics content)
V Fly ash
S Ground granulated blastfurnace slag (ggbs)
D Silica fume
P Natural pozzolana
Q Natural calcined pozzolana or high reactivity Natural calcined pozzolana
M CEMII/A-L or LL with 6 to 29% addition
SR Sulfate resisting

Alkali-activated cementitious materials

There are also geopolymers or alkali-activated cementitious materials (AACMs). These are not currently covered in BS 8500, but that can be specified using PAS 8820 Construction materials. Alkali-activated cementitious material and concrete. Specification. This specification is produced by BSI. AACM's are normally based on ggbs which is activated by a chemical rather than Portland cement.

Designation

Concrete can be specified as either designed, designated, standardised prescribed, prescribed, proprietary or nominally proportioned. For structural concrete the first two are appropriate.

For designed concretes the purchaser/specifier is responsible for the required performance e.g. C28/35, CIIIA, S2, DC-2 and the producer is responsible for selecting the concrete proportions to satisfy the performance, where the compliance is by strength.

For Designated concretes, an alpha-numeric reference system is used to 'designate' these concretes for particular purposes. The concrete is chosen from a list of designated concretes (GEN, FND, PAV, RC etc.) in accordance with BS 8500 and the producer must hold a current "accredited production control certification". The purchaser is responsible for specifying the designation, e.g. PAV2, and the producer is responsible for selecting the concrete proportions to satisfy the performance. Again, compliance is by strength. Specifying a designated concrete may not necessarily define the prevailing exposure conditions.

Execution

BS EN 13670 Execution of concrete structures, is part of the Eurocode suite of standards and covers workmanship requirements previously covered in design standards, including precast and prestressed concrete. This European Standard applies to the execution of concrete structures to achieve the intended level of safety and serviceability during its service life, as given by EN 1990, Eurocode - Basis of structural design, EN 1992, Eurocode 2 - Design of concrete structures and EN 1994, Eurocode 4 - Design of composite steel and concrete structures.

The 10 sections plus Informative Annexes A to H cover;
Scope, normative references and definitions
Execution management
Falsework and formwork
Reinforcement
Prestressing
Concreting
Execution with precast concrete elements
Geometrical tolerances

Fixing tolerances for reinforcement

To avoid confusion, BS 8500-1 Tables A.5 and A.6 give cover as a minimum dimension plus a tolerance Δc. This tolerance is covered in BS EN 13670 and the deviation found in the NA to BS EN 1992-1-1 clause 4.4.1. Unless otherwise specified it is taken as 10mm. Note that structural drawings show nominal cover (clause 4.4.1.1).

THE DESIGNER'S PERSPECTIVE

For a designer, durability is a topic that extends through the design, specification, construction and maintenance phases of a project. For a bridge to be durable it must be well designed, suitably specified, well-constructed and fully maintained. If any one of these falls below the required standards, the durability of a bridge can be adversely affected. The design and specifications come first in the building process and the designer has the prime responsibility in achieving a durable bridge.

The choice of a bridge form and type, whether reinforced concrete, prestressed concrete or steel, is the start of the many design decisions that influence the durability of a structure. As well as the design, the detailing is also paramount in achieving a durable structure. Details should be adopted that are proven in their effectiveness and durability, and guidance on this may be obtained from the section on detailing within these design guides.

With all concrete structures, it is essential that the final product has dense, well compacted, homogenous concrete. As well as specifying the correct properties and type of concrete to be used, the designer should also make sure that there is adequate room around the reinforcement and prestressing ducts to place and vibrate the concrete. A concrete with defects which is subsequently repaired is seldom as sound as well-placed concrete from the outset.

The designer may also obtain guidance on durability from Highways England's Standards. These documents have been promoted by Highways England to achieve the lowest whole-life cost of a bridge, rather than the least construction cost.

A cornerstone of Highways England's standards is the need to eliminate expansion joints and bearings. These have proven to be problematic in their maintenance and wherever possible they should be designed out of a structure. For deck lengths of less than 60m it is now common in the UK to use integral bridges with the deck built into the piers and abutment.

Access for inspection and maintenance is another key issue for the designer. It is preferable for all concrete surfaces, except those against the soil, to be easily reached within arm's length so that future inspectors can closely view and monitor the structure. Critical items such as bearing and expansion joints, if the bridge has them, should have sufficient room for them to be removed and replaced as this is likely to happen several times in the life of the bridge.

Drainage of any surface water away from the bridge is another problematic area. Expansion joints, drains and pipes tend to leak, especially after years of use, and in the UK the damage done by chloride laden water on reinforced concrete is one of the most serious problems confronting maintenance engineers. The designer should make provisions in the design to ensure that any water is effectively taken away from the concrete and not allowed to 'pond' or peculate to the reinforcement or bearings etc.

The design codes and standards usually specify a minimum level of durability within the structure, but it is often the designer's prerogative to take more measures to improve the long-term performance of the bridge. The designer may specify greater cover to the concrete, admixtures that can be added to improve the durability or coatings that increase the protection. Concrete can be enclosed within waterproof membrane when below ground or sprayed with impregnable coatings above. These additional measures may increase the initial cost of a structure, but can result in savings from future maintenance and repair.

If concrete is being used in a particularly vulnerable location where repairs would be difficult and costly, such as under water, protection to the reinforcement can be achieved with the use of cathodic protection systems, although these are usually better if installed during the construction, rather than later after the problems have started.

Reinforced concrete has been used in bridgeworks for over 100 years, while prestressed concrete was first used on a bridge 75 years ago. Many of the early concrete bridges are still standing and in use, proving that concrete is a durable robust material.

INTRODUCTION

The major ancillary items of a bridge include the bearings, parapets, waterproofing and expansion joints. Other items, such as lighting, drainage and access provisions, are also often required. Without these, a bridge would not function as efficiently or last so long.

There are many proprietary products that are available on the market from which the designer can choose his preferred solution. Most of these products are tried and tested and can be customized to suit a particular bridge requirement.

The following sections give guidance to the designer on the different ancillary items that complete a bridge structure.

BEARINGS

Most common bearing types
Elastomeric, Roller, Pot, Rocker, Knuckle, and Spherical

Manufacturers
There are many, some of which are:
Ekspan, Freyssinet, FIP, Mageba, Maurer.

BS EN 1337 Structural bearings
Part 1: 2000: General Design Rules
Part 2: 2004: Sliding Elements
Part 3: 2005 Elastomeric Bearings
Part 4: 2004 Roller Bearings*
Part 5: 2005 Pot Bearings*
Part 6: 2004 Rocker Bearings*
Part 7: 2004 Spherical and Cylindrical PTFE Bearings*
Part 8:2007 Guide Bearings and Restrain Bearings*
Part 9: 1998 Protection
Part 10: 2003 Inspection and Maintenance
Part 11: 1998: Transport, Storage and Installation
*under review Aug 2021

Limitations on Performance
Load Capacity - vertical & horizontal
Allowable movement - translation & rotation
Restrictions on material properties
Construction tolerances
Detailing
Protective Coatings.

WATERPROOFING

Type
Sheet and Liquid applied

Manufacturers
Bridge Authority Registered Systems:
BDW, Pitchmastic, GCP Applied Technologies (acquired Stirling Lloyd), Universal Sealants

Code
Design Manual for Roads and Bridges CD 358 Waterproofing and Surfacing of Concrete Bridge Decks

Limitations on Performance
BBA Registration
Adhesion
Age of substrate
Restrictions on material properties
Detailing
Compatibility with protective coatings

MOVEMENT/EXPANSION JOINTS

Introduction

The performance of bridge joints is affected by temperature and traffic volume, together with the nature of the traffic. Both must be determined prior to selection of a type of joint. Different joints will have advantages and disadvantages. When selecting a joint type, the user or specifier should take into reasonable account the whole life costing of a joint system and the necessity of regular or long-term maintenance.

Bridge expansion joints have to function as "riding plates" to carry the imposed traffic loads and also accommodate the thermal movement, shrinkage, pre-stress creep and rotation of the deck. These joints can be simple flexibilised asphalts or complex mechanical or elastomeric elements, according to the range of movements to be accommodated. The expansion joint should give good riding characteristics without generating excessive noise from traffic, especially in urban areas where adjacent residential property may need careful consideration. It must also be functional for the road-user whilst having good skid resistance and be suitable for the road curvature and alignment. If pedestrians, animals and cyclists use a bridge the expansion joint should be of a design which does not cause safety problems. Footpaths may need cover plates slightly recessed below the surface to provide safe access.

Durability

The effectiveness of bridge expansion joints may be very significant in the life expectancy of bridges. Recognition of the need for high standards of waterproofing to prevent the ingress of water, chlorides and other waterborne contaminants into the bridge structure has focused attention on the importance of effective bridge expansion joints.

It is therefore essential to use materials which are durable and offer a maintenance-free operation. Any elements subject to wear must be replaceable using simple techniques since traffic management schemes and lane closures are costly and need special authorisation, as well as causing public irritation. Therefore, it may be expedient to replace bridge expansion joints prematurely while other maintenance work, such as re-surfacing, is carried out so that future road closures are minimised.

Design

The bridge designer must clearly set out the desired operating standards and define the total movements related to the imposed loadings, temperature range, deck shortening and rotation. In this way, the manufacturer or supplier can provide the correct technical solution which can be incorporated into the working drawings.

It is essential to embrace the particular design features of various bridge expansion joints so that box-outs and plinths can be formed without resorting to changes in reinforcement, etc, at a later state. It is vital that the bridge expansion joint is formed continuously from parapet to parapet, taking into account footpaths, kerbs, central reserves and skew angles. The aggregated longitudinal movement and skew movement should be used to select the correct size and performance of the joint. The fixing and bonding of different types of joints should not cause or propagate damage to the road surfacing or the supporting bridge structure.

Installation

Installation of the joints should only be carried out by approved bridge joint installation contractors. Installation should be delayed for as long as possible to allow for shrinkage and creep of the deck and settlement of the supports. The gap widths should be formed to suit the bridge deck temperature in relation to the mean deck temperature. Further information on this relationship can be obtained from TRL Report SR479 Bridge temperature for setting bearings and expansion joints by Mary Emerson.

Drainage

Where segmental types, elastomeric elements in metal runners or cantilever comb/tooth joints are used, a separate drainage membrane should be used to collect and discharge any infiltrated water into outlets to increase the security of the joints.

Generic expansion joint types

The information provided below is a simplified information guide based on the Design Manual for Roads and Bridges CD 357 Bridge expansion joints. This document replaced BD 33/94, BA 26/94, IAN 168/12 and IAN 169/12 in 2020.

The various 'families' of expansion joint type covered by the constituent parts of ETAG 032 Guideline for European Technical Approval of Expansion Joints for Road Bridges.

Buried expansion joint (ETAG 032 Part 2)
One or more components may be used to form continuity of the waterproofing and surfacing. These can comprise of proprietary flashings and straps with bridging plates or similar. Movement ranges generally 5-20mm.

Flexible plug joint (ETAG 032 Part 3)
Proprietary systems comprising of layers of specially modified binders and aggregate to provide a homogeneous expansion medium and smooth running surface. Movement ranges for standard grade modified binders 5-40mm.

Nosing expansion joint (ETAG 032 Part 4)
In-situ resins or cementitious mixtures placed either side of the bridge deck air gap to produce firm edges and protect the surfacing. Complete with watertight extruded compression seal or sealant. Movement ranges 5-40mm with preformed seals 5-12mm with poured sealant.

Precision extruded metal rails set between special resins forming combined nosings and bed, stuck down to the structural concrete and surfacing. Incorporating various sizes of watertight seals fitted between the rails. Movement range up to 150mm. Seldom used today.

Mat expansion joint (ETAG 032 Part 5)
A joint prefabricated to exact lengths and widths, comprising of rubber surrounding metal elements, bearing plates and reinforcement. Placed onto flat beds with resin strips either side as a protection and to provide a smooth running surface. Bolted directly to the structural concrete or to a prefabricated metal cradle set into the structure during the construction of the deck and ballast walls. Movement range: up to 350mm.

Cantilever expansion Joint (ETAG 032 Part 6)
A prefabricated precision made joint consisting broadly of two sets of finger plates set or fixed across the joint gap with a separate flexible waterproof membrane sheet clamped beneath the plates. Movement range: standard 440mm, non-standard 1000mm.

Modular expansion joint (ETAG 032 Part 8)
Modular expansion joints are more complex, and can accommodate a larger range of movement. In multi-element form, one or more intermediate runners are provided (in addition to the runners either side of the deck joint gap), into which the elastomeric seals are fitted, and these are supported on support beams placed transversely to the joint gap. The support beams are, in turn, supported on sliding bearings, which facilitate the movement of the deck.

Supported expansion joint (ETAG 032 Part7)
A supported expansion joint consists of one sub-component flush with the running surface, which is fixed by hinges on one side and sliding supports on the other side and which spans the deck joint gap. The expected structure movement is allowed through sliding on the non-fixed side of the hinged sub-component. Designed to accommodate large movements of 150mm upwards, with good riding quality.

THE DESIGNER'S PERSPECTIVE

There are many different types of ancillary items available and a range of proprietary products marketed. It can be difficult for the designer to know which the best solution for a particular bridge. One of the major problems faced by the designer is finding out how the proprietary products have performed in the past. Past problems are soon forgotten and the successes, by their nature, are not usually noticed.

Although guidance can be found in these design guides and the references given, the experience and knowledge of the designer is paramount.

Items such as bearings and expansion joints are usually dictated by their functional requirement and will depend on the forces and movements being imposed by the bridge. A designer usually specifies these functional requirements which the contractor uses these to source the cheapest available product. The designer must take care in preparing the specifications and drawings to provide clear guidance on the type of expansion joint or bearing he wants, especially if he wants more than the minimum standard.

Bearings must be designed for removal and replacement and the structures above and below arranged and detailed so that temporary jacks can be positioned to carry the deck and relieve the load on the bearing. Using rubber bearings can provide inexpensive solutions when only limited loads and movement are imposed, while pot or mechanical bearings can cater for higher loads and higher movements but are more complex to install and replace.

Expansion joints must also be accessible and replaceable, but often it is only the component that spans the gap that needs to be replaced. Expansion joints on road bridges should have sub-surface drainage provided across the width of the deck behind the joint to prevent a build-up of water, otherwise hydraulic action caused by the traffic passing may lead to breaking up of the asphalt. Drainage facilities should be provided across the gap under the expansion joint where possible, to carry away any water that may leak through.

Parapets down the sides of the bridge have a great impact on the appearance of the structure and should be chosen by the designer with that in mind. Standard highway and pedestrian parapets can be used where the barrier is primarily functional. Where visual impact is important, the designer should consider developing an arrangement more pleasing to the eye. Concrete parapets can be shaped and featured on the outside to give a pleasing visual impact although they increase the overall apparent depth of the structure and need to be developed along with the overall deck profile. Steel or aluminium parapets have less impact, on the general view of the observer, but tend to be unattractive when viewed up close.

Bridge designers in the UK have learnt new rules for the design of highway parapets and barriers. The current design for Road Restraint Systems is governed by BS EN 1317 Road Restraint Systems and DMRB CD 377 Requirements for road restraint systems. BS EN 1317 is a performance-based standard and provides requirements on the layout and performance of the restraint system. Parapets and barriers are specified in terms of its Normal Containment Level, Impact Severity Level and Working Width Class, which any proprietary product must comply with.

Historically in the UK, parapets and barriers have comprised of steel components mounted on concrete upstands or plinths. In recent years, concrete barriers have been installed between motorway carriageways. Overseas, the use of concrete barriers for roads and bridges is much more common. They are safer, more robust and require less maintenance than their steel counterpart. With the introduction of the new standards there is an opportunity for bridge designers to utilise the advantages of concrete parapets and barriers for more diverse projects.

Waterproofing the concrete deck top surface using an impermeable membrane is essential to keep water away from the concrete, for both highway and rail bridges. Highway bridges in particular are vulnerable due to the de-icing salts used on the roads during the winter. Designers should ensure that durable and effective waterproofing is used. Proprietary systems that have a higher capital cost can be cost saving later, by offering better protection through the life of the bridge.

Deck drainage is an area which is sometimes not given enough attention at the design stage. Drainage gulleys and pipes can have a significant impact on the detailing and construction of a bridge. They are usually easily incorporated into concrete bridges with the reinforcement and concrete arranged to suit the drainage system used, but too often they are only considered at the end of the design when it becomes more difficult to squeeze them in. In the UK, kerb drains are becoming more popular even for longer bridges. These eliminate the need for large gulleys and the pipes can usually be embedded within the kerb upstand concrete.

With concrete bridges it is usually easy to incorporate the lighting requirements of the structure. Highway or pedestrian lighting may be required for the carriageway over the bridge, or as soffit lighting for roads or paths beneath. The conduits and anchorages for the fittings can be cast into the concrete to give protection and hide them from view. Underbridge lighting for roads below can be embedded in the bottom concrete surface, or left proud of the surface where headroom and vandalism is not an issue.

Low-level lighting and electrical power supply may be required for inspections and maintenance at key locations of the bridge, especially inside box girders. These can usually be easily built-in to the concrete elements.

Fixing of ancillary items to the concrete is relatively easy, with anchorages or other parts being embedded in the concrete during casting, or with drilled in fixings afterwards.

INTRODUCTION

It is common for bridge engineers to become involved with the design of bridges in other countries, using a variety of international codes and satisfying local requirements.

Whichever code or specification is used, concrete is similar the world over and the fundamental principles of designing and constructing concrete bridges remain the same. However, the approach taken by the different codes can vary greatly and each needs to be learned when first used.

The following sections provide guidance on international codes and the differences with their UK counterparts. The references include links to other websites where more information can be found on the major international codes likely to be encountered.

INTERNATIONAL DESIGN CODES

In the past we have been fortunate in the UK that BS 5400 was recognised internationally as providing sound guidance for the design of bridges. Some countries still employ it as their primary standard or have used it for the design of longer span or more complex structures.

Over the past decade the UK has adopted the Eurocode suite of standards which is also becoming more common around the world as other countries adopt it. AASHTO is the USA standard for bridge design and is often adopted in countries with strong links to the USA.

Basic code model

The main concrete model in use are the Eurocodes which superseded BS 5400 Steel, concrete and composite bridges. Both use the CEB-fib approach towards the design of reinforced concrete sections. An alternative model is produced by ACI which underpins American, Australian and New Zealand codes.

Eurocodes

The Eurocodes present both a major challenges and opportunities for the bridge engineer. Much has been written about Eurocodes and only the briefest of summary is given here.

After a period of coexistence of five years when either the British Standard or the Eurocode may be used, Eurocodes are now well-established. Designers must now use Eurocodes.

BS 5400 was a comprehensive, bridge specific code that provided all the design information in one document whereas there are multiple documents that need to be referred to with Eurocodes to carry out a full design.

The main Eurocode is BS EN 1992-1-1 Eurocode 1 Actions on structures. General actions - Densities, self-weight and imposed loads for buildings. The bridge designer will need to also obtain a copy of BS EN 1992-2 BS EN 1992-2: Eurocode 2: Design of concrete structures. Concrete bridges - Design and detailing rules, which contains bridge specific details. The basis for design is given in BS EN 1990 Eurocode. Basis of structural design, while loading is provided in BS EN 1991-2 Eurocode 1. Actions on structures. Traffic loads on bridges.

The designer will also need to obtain the National Annexes to see what Nationally Defined Parameters are specified (the National Annexes are documents that are specific to designs carried out for each Member State and take into account local design information such as environmental load effects, safety requirements and other sensitive parameters).

There are a number of guides for Eurocodes, including the CBDG Technical Guide TG13 Integral concrete bridges to Eurocode 2. Both the Institution of Civil Engineers and the Institution of Structural Engineers have produced design guides.

Stress Models

One of the fundamental differences between concrete codes of practice is the calculation of the stress block in the concrete. BS 8110 (withdrawn and superseded) and BS EN 1992-1-1, use similar models, but with slight difference in the parameters. The shape of the stress block in the ACI model is arguably more realistic in that it takes into account the reduction in strength as the material approaches compression failure.

Perhaps a more significant effect is that BS 5400 and Eurocodes included material factors within the stress/strain relationship of both materials, whereas the ACI model applies a capacity reduction factor, dependent on the aspect of the section being considered. Thus, sections in bending (which for an under reinforced section is predominantly dependent on steel strength) a higher factor is used, reflecting the greater consistency of the steel material strengths. Lower factors are used for columns and for sections in shear, in keeping with the greater variation in concrete strengths.

One other significant area of difference is the design limits for shear. BS 5400 and Eurocodes are more penal on shear forces than other codes. Paradoxically, the longitudinal shear requirements are higher than in ACI.

Limit States

The UK has used limit state design since the 1970s and designers are very familiar with these requirements. The same is not true internationally and limit state codes may not be used in some countries. Indian bridge codes are available in both limit state and working stress forms, as are AASHTOs (AASHTO is the American Association of State Highway Transportation Officials). It is worth closely checking which standard has been adopted by the client for projects in such countries, although experience shows that the limit state approach is generally used for major international projects.

Load Factors

Another area of fundamental difference is the approach taken in the determination of load factors and load combinations. National design requirements are implicit in the choice of factors and it is as well for users to acquaint themselves with the differences. Cases which may be governing for some bridge types under Eurocodes and BS 5400 are not necessarily governing in the AASHTO codes.

Mixing and matching

Great care is needed if provisions from one code are to be used in conjunction with another. There are basic incompatibilities between the ACI and CEB-fib methods of design and whereas some of the data is underpinned by the similar research, the codification may be very different. If it is required to draw on the requirements in the ACI provisions for whatever reason, care is required to determine the full application of the provisions used.

The ACI commentary is very thorough in that it provides lists of references of the source data used in assembling the standard. It is also possible to get good advice directly from the code committees by following the links through the ACI's website.

THE DESIGNER'S PERSPECTIVE

Many engineers have now become part of an itinerant workforce that moves from project to project around the world. Often this is on a virtual basis where the designer sits in his office in his home country and services the client's needs via the internet and e-mail, but just as common is for engineer to travel to where the projects are being done. This requires bridge engineers to be able to design bridges using any of the international design codes commonly used.

From a designers view point this creates challenges in adopting different codes and standards to those they are more familiar with, but it also gives the opportunity for designers to broaden their knowledge and learn different ways of doing things. It requires engineers to learn the basic principles behind the codes so that they can apply them correctly and understand how the different parts coexist.

When using international design codes, the designer's first challenge is to determine the local requirements as well as understand the design standards being used. Many countries have developed their own interpretation of the international standards and often publish their own accompanying rules to be followed. Often these local requirements are only published in the local language or are only available through the local authorities and obtaining this information might be a lengthy process. Fortunately, it is common for international consultants to work in association with local consultants who are usually able to provide guidance on the local rules and regulations.

It can sometimes take a while to get used to the symbols, notation and formulae used in the different codes. Most codes contain lengthy explanations to help the designer.

One of the most common confusions when using the different codes is in the use of the cube strength or the cylinder strength of the concrete. Eurocodes and AASHTO use cylinder strengths in their design codes and specifications. However, quality control testing may be undertaken on cubes, as in the UK, hence the dual notation of characteristic compressive strength of, for example, C32/40 (cylinder strength/cube strength) adopted by European standards. This relationship is covered in the Eurocodes and specification standards e.g. BS EN 206 and BS 8500.

Wherever a bridge is built in the world it uses similar materials and has the same function, irrespective of the design code. The codes are just a means to an end to design a safe, durable bridge that carries the road, rail or pedestrian traffic over some obstruction below.

Much of the designer's work is the same for all bridges and does not depend on the code is being used. The behaviour and analysis of a bridge, the generation of bending moments and shears, and drafting of the structural elements are common to all bridges. It is usually in the application of the loading, the design formulae and the detailing where the codes differ.

The differences can be significant in some respects. It is generally considered that bridges designed to AASHTO will produce a structure with less concrete, reinforcing and prestressing when compared to one designed to Eurocodes.

A big difference between the two is the applied loading, but the detailing rules and general design approach also make a difference. When a designer is working in a country where they have a choice of which code to use, they will often choose AASHTO, especially if they are designing for a contractor where cost is paramount.

Many aspects of UK bridge design are unheard of overseas. The following lists some of the differences that the designer will come across:

• Run-on or approach slabs are still used in many countries, but are less common in the UK
• Internal tendons are used with precast segmental construction in most counties, apart from the UK
• Steel ducts for prestressing tendons are more common elsewhere than the plastic ducts favoured in the UK.
• Partial prestressing is common in many countries, although almost unheard of in the UK.

While local requirements sometimes seem onerous or unreasonable when compared to what is common in the UK, the local practice should usually be adhered to as engineers have many years of experience of local bridge design practice.

Concrete, whether reinforced or prestressed, is the same basic material wherever it is used. Differences in aggregate and cement will produce local variants, but if a designer understands the fundamental properties and behaviour of concrete and its use in bridge structures, they should have little difficulty in designing and constructing concrete bridges anywhere in the world.