The steel hollow section column as Architectural proposed in Schematic design is under suggestion to change to open web section to allow the flexible installation of anchor bolts of steel base column. The simple design approach was used as a design criterion. All load combinations were entered into the model, and the combined load effects were compared to the reduced nominal strengths of the members.
Chapter 31 Tolerances by COLIN TAYLOR 31.1 Introduction 31.1.1 Why set tolerances? Compared to other structural materials, steel (and aluminium) structures can be made economically to much closer tolerances Compared to mechanical parts, however, it is neither economic nor necessary to achieve extreme accuracy There are a number of distinct reasons why tolerances may need to be considered It is important to be quite clear which actually apply in any given case, particularly when deciding the values to be specified, or when deciding the actions to be taken in cases of non-compliance The various reasons for specifying tolerances are outlined in Table 31.1 In all cases no closer tolerances than are actually needed should normally be specified, because while additional accuracy may be achievable, it generally increases the costs disproportionately 31.1.2 Terminology ‘Tolerance’ as a general term means a permitted range of values Other terms which need definition are given in Table 31.2 31.1.3 Classes of tolerance Table 31.3 defines the three classes of tolerances which are recognized in Eurocode It is important to draw attention to any particular or special tolerances when calling for tenders, as they usually have cost implications Where nothing is stated, fabricators will automatically assume that only normal tolerances are required 917 918 Tolerances Table 31.1 Reasons for specifying tolerances Structural safety Dimensions (particularly of cross-sections, straightness, etc.) associated with structural resistance and safety of the structure Assembly requirements Tolerances necessary to enable fabricated parts to be put together Fit-up Requirements for fixing non-structural components, such as cladding panels, to the structure Interference Tolerances to ensure that the structure does not foul with walls, door or window openings or service runs, etc Clearances Clearances necessary between structures and moving parts, such as overhead travelling cranes, elevators, etc or for rail tracks, and also between the structure and fixed or moving plant items Site boundaries Boundaries of sites to be respected for legal reasons Besides plan position, this can include limits on the inclination of outer faces of tall buildings Serviceability Floors must be sufficiently flat and even, and crane gantry tracks etc must be accurately aligned, to enable the structure to fulfil its function Appearance The appearance of a building may impose limits on verticality, straightness, flatness and alignment, though generally the tolerance limits required for other reasons will already be sufficient Table 31.2 Definitions – deviations and tolerances Deviation The difference between a specified value and the actual measured value, expressed vectorially (i.e as a positive or negative value) Permitted deviation The vectorial limit specified for a particular deviation Tolerance range The sum of the absolute values of the permitted deviations each side of a specified value Tolerance limits The permitted deviations each side of a specified value, e.g ±3.5 mm or +5 mm -0 mm Table 31.3 Classes of tolerances Normal tolerances Those which are generally necessary for all buildings They include those normally required for structural safety, together with normal structural assembly tolerances Particular tolerances Tolerances which are closer than normal tolerances, but which apply only to certain components or only to certain dimensions They may be necessary in specific cases for reasons of fit-up or interference or in order to respect clearances or boundaries Special tolerances Tolerances which are closer than normal tolerances, and which apply to a complete structure or project They may be necessary in specific cases for reasons of serviceability or appearance, or possibly for special structural reasons (such as dynamic or cyclic loading or critical design criteria), or for special assembly requirements (such as interchangeability or speed of assembly) Standards 919 31.1.4 Types of tolerances For structural steel there are three types of dimensional tolerance: (1) Manufacturing tolerances, such as plate thickness and dimensions of sections (2) Fabrication tolerances, applicable in the workshops (3) Erection tolerances, relevant to work on site Manufacturing tolerances are specified in standards such as BS 4, BS 4848, BS EN 10024, BS EN 10029, BS EN 10034 and BS EN 10210 Only fabrication and erection tolerances will be covered here 31.2 Standards 31.2.1 Relevant documents The standards covering tolerances applicable to building steelwork are: (1) BS 5950 Structural use of steelwork in building Part 2: Specification for materials fabrication and erection: hot rolled sections Part 7: Specification for materials and workmanship: cold formed sections and sheeting (2) National structural steelwork specification for building construction NSSS, 4th edition (3) ENV 1090-1 Execution of steel structures: Part 1: General rules and rules for buildings (4) ISO 10721-2: 1999 Steel structures: Part 2: Fabrication and erection (5) BS 5606 Guide to accuracy in building 31.2.2 BS 5950 Structural use of steelwork in building The specification of tolerances for building steelwork was first introduced into British Standards in BS 5950: Part 2: 1985 The current edition was issued in 2001 This revision of the 1992 edition updates cross-references to other standards, many of which are now European Standards (BS EN standards) In addition the opportunity was taken to align the code more closely with the industry standard document, the National structural steelwork specification for building construction 920 Tolerances 31.2.3 National structural steelwork specification (NSSS) The limitations of the tolerances specified in earlier versions of BS 5950: Part have been extended by an extensive coverage of tolerances in the National structural steelwork specification for building construction This is an industry standard based on established sound practice The widely accepted document, promoted by the British Constructional Steelwork Association (BCSA), is now in its 4th edition 31.2.4 ENV 1090-1 Execution of steel structures As part of the harmonization of construction standards in Europe, CEN has issued ENV 1090: Part 1: General rules and rules for buildings, which is available through BSI as DD ENV 1090-1: 1998 This document includes comprehensive recommendations for both erection and manufacturing tolerances To a large extent these recommendations are consistent with BS 5950: Part and the NSSS However, some of them are more detailed 31.2.5 ISO 1071-2 Steel structures: Part 2: Fabrication and erection This is very similar to ENV 1090-1 and BS 5950: Part It is unlikely to be issued as a BSI standard 31.2.6 BS 5606 Guide to accuracy in building BS 5606 is concerned with buildings generally and is not specific to steelwork The 1990 version has been rewritten as a guide, following difficulties due to incorrect application of the previous (1978) version, which was in the form of a code BS 5606 is not intended as a document to be simply called up in a contract specification It is primarily addressed to designers to explain the need for them to include means for adjustment, rather than to call for unattainable accuracy of construction Provided that this advice is heeded, its tables of ‘normal’ accuracy can then be included in specifications, except where they conflict with overriding structural requirements This can in fact happen, so it is important to remember that the requirements of BS 5950 must take precedence over BS 5606 BS 5606 introduces the idea of characteristic accuracy, the concept that any construction process will inevitably lead to deviations from the target dimensions, and its objective is to advise designers on how to avoid resulting problems on site by appropriate detailing The emphasis in BS 5606 is on the practical tolerances which will normally be achieved by good workmanship and proper site supervision This can only be improved upon by adopting intrinsically more accurate techniques, Implications of tolerances 921 which are likely to incur greater costs These affect the fit-up, the boundary dimensions, the finishes and the interference problems Data are given on the normal tolerances (to be expected and catered for in detailed design) under two headings: (1) Site construction (table of BS 5606) (2) Manufacture (table of BS 5606) Unfortunately many of the values for site construction of steelwork are only estimated No specific consideration is given in BS 5606 to dimensional tolerances necessary to comply with the assumptions inherent in structural design procedures, which may in fact be more stringent It does however recognize that special accuracy may be necessary for particular details, joints and interfaces Another important point mentioned in BS 5606 is the need to specify methods of monitoring compliance, including methods of measurement It has to be recognized that methods of measurement are also subject to deviations; for the methods necessary for monitoring site dimensions, these measurement deviations may in fact be quite significant compared to the permitted deviations of the structure itself 31.3 Implications of tolerances 31.3.1 Member sizes 31.3.1.1 Encasement The tolerances on cross-sectional dimensions have to be allowed for when encasing steel columns or other members, whether for appearance, fire resistance or structural reasons It should not be forgotten that the permitted deviations represent a further variation over and above the difference between the serial size and the nominal size For example, a 356 ¥ 406 ¥ 235 UC has a nominal size of 381 mm deep by 395 mm wide, but with tolerances to BS may actually measure 401 mm wide by 387 mm deep one side, and have a depth of 381 mm the other side The same is true of continental sections A 400 ¥ 400 ¥ 237 HD also has a nominal size of 381 mm deep by 395 mm wide, but with tolerances to Euronorm 34 may actually measure 398 mm wide by 389 mm deep one side, and have a depth of 380 mm the other side 31.3.1.2 Fabrication Variations of cross-sectional dimensions (with permitted deviations) may also need to be allowed for, either in detailing the workmanship drawings or in the fabrication process itself, if problems are to be avoided during erection on site 922 Tolerances The most obvious case is a splice between two components of the same nominal size, where packs may be needed before the flange splice plates fit properly, unless the components are carefully matched Similarly variations in the depths of adjacent crane girders or runway beams may necessitate the provision of packs, unless the members are carefully matched Less obviously, if the sizes of columns vary, the lengths of beams connected between them will need some form of adjustment, even if the columns are accurately located and the beams are exactly to length 31.3.2 Attachment of non-structural components It is good practice to ensure that all other items attached to the steel frame have adequate provision for adjustment in their fixings to cater for the effects of all steelwork tolerances, plus an allowance for deviations in their own dimensions Where necessary, further allowances may be needed to cater for structural movements under load and for differential expansion due to temperature changes Where possible, the number of fixing points should be limited to three or four, only one of which should be positive with all the others having slotted holes or other means of adjustment 31.3.3 Building envelope It must be appreciated that erection tolerances, including variation in the position of the site grid lines, will affect the exact location of the external building envelope relative to other buildings or to site boundaries, and there may be legal constraints to be respected which will have to be taken into account at the planning and preliminary stages of design These effects also need to be taken into account where a building is intended to have provision for future extension or where the project is an extension of an existing building, in which case deviations in the actual dimensions have to be catered for at the interface In the case of tall multi-storey buildings, the building envelope deviates increasingly with height compared to the location at ground level, even though permitted deviations for column lean generally reduce with height Unless there are step-backs or other features with a similar effect, it may be necessary to impose particular tolerance limits on the outward deviations of the columns Fabrication tolerances 923 31.3.4 Lift shafts for elevators The deviations from verticality that can be tolerated in the construction of guides for ‘lifts’ or elevators are commonly more stringent than those for the construction of the building in which they operate In low-rise buildings sufficient adjustment can be provided in association with the clearances, but in tall buildings it becomes necessary either to impose ‘special’ tolerances on column verticality or else to impose ‘particular’ tolerances on those columns bounding the lift shaft In agreeing the limits to be observed with the lift supplier, it should not be overlooked that the horizontal deflections of the building due to wind load also have implications for the verticality of the lift shafts 31.4 Fabrication tolerances 31.4.1 Scope of fabrication tolerances The description ‘fabrication tolerances’ is used here to include tolerances for all normal workshop operations except welding It thus covers tolerances for: (1) (2) (3) (4) (5) (6) cross sections, other than rolled sections, member length, straightness and squareness, webs, stiffened plates and stiffeners, holes, edges and notches, bolted joints and splices, column baseplates and cap plates However, tolerances for cross sections of rolled sections and for thicknesses of plates and flats are treated as manufacturing tolerances Welding tolerances (including tolerances on weld preparations and fit-up and sizes of permitted weld defects) are treated elsewhere 31.4.2 Relation to erection tolerances An overriding requirement for accuracy of fabrication must always be to ensure that it is possible to erect the steelwork within the specified erection tolerances Due to the wide variety of steel structures and the even wider variety of their components, any recommended tolerances must always be specified in a very general way Even if it were possible to specify fabrication tolerances in such a way that their cumulative effect would always permit the specified erection tolerances to be satisfied, the resulting permitted deviations would be so small as to be unreasonably expensive, if not impossible, to achieve 924 Tolerances Fortunately in most cases it is possible to rely on the inherent improbability of all unfavourable extreme deviations occurring together Also the usually accepted values for fabrication tolerances make some limited allowances for the need to avoid cumulative effects developing on site They are tolerances that have been shown by experience to be workable, provided that simple means of adjustment are incorporated where the effects of a number of deviations could otherwise become cumulative For example, beams with bolted end cleats usually have sufficient adjustment available due to hole clearances, but where a line of beams all have end plate connections, provision for packing at intervals may be advisable, unless other measures are taken to ensure that the beams are not all systematically over-length or under-length by the normal permitted deviation Other possible means for adjustment include threaded rods and slotted holes Where it can be seen from the drawings that the fabrication tolerances could easily accumulate in such a way as to create a serious problem in erection, either closer tolerances or means of adjustment should be considered; however, the coincident occurrence of all extreme deviations is highly improbable, and judgement should be exercised both on the need for providing means of adjustment and on the range of adjustment to be incorporated 31.4.3 Full contact bearing 31.4.3.1 Application The requirements for contact surfaces in joints which are required to transmit compression by ‘full contact bearing’ probably cause more trouble than any other item in a fabrication specification, largely due to misapprehension of what is actually intended to be achieved First it is necessary to be clear about the kind of joint to which the requirements for full contact bearing should be applied Figure 31.1(a) shows the normal case, where the profile of a member is required to be in full contact bearing on a baseplate or cap plate or division plate The stress on the contact area equals the stress in the member: thus full contact is needed to transmit this stress from the member into the plate Only that part of the plate in contact with the member need satisfy the full contact bearing criteria, though it may be easier to prepare the whole plate Figure 31.1(b) shows two end plates in simple bearing The potential contact area is substantially larger than the cross-sectional area of the member: thus full contact bearing is not necessary All that is needed is for the end plates to be square to the axis of the member Another common case of simple bearing is shown in Fig 31.1(c) By contrast, the case shown in Fig 31.1(d) is one where, if full contact bearing is needed, it is also necessary to take special measures to ensure that the profiles of the two members align accurately, otherwise the area in contact may be significantly less than the area required to transmit the load Particular tolerances should be specified in such cases, based on the maximum local reduction of area that can be Fabrication tolerances 925 Fig 31.1 Types of member-to-member bearing: (a) profile to plate, (b) plate to plate, (c) flange to flange, (d) profile to profile (accurate alignment necessary) accepted according to the design calculations Alternatively a division plate could be introduced; if the stresses are high this may well prove to be the most practical solution 31.4.3.2 Requirements Where full contact bearing is required, there are in fact three different criteria involved: (1) Squareness (2) Flatness (3) Smoothness 926 Tolerances 31.4.3.3 Squareness If the ends of a length of column are not square to its axis, then after erection either the column will not be vertical or else there may be tapered gaps at the joints, depending on the extent to which surrounding parts of the structure prevent the column from tilting Under load any such gap will try to close, exerting extra forces on the surrounding members In addition, both a gap or a tilt will induce a local eccentricity in the column A practical erection criterion is that the column should not lean more than in x (where x is 600 in NSSS and 500 in ENV 1090-1) This slope is measured relative to a line joining the centres of each end of the column length, referred to as the overall centreline The column is also allowed a lack of straightness tolerance of (length/1000), which corresponds to end slopes of about 1/300 (see Fig 31.2(a)) It is thus necessary to specify end squareness criteria relative to the overall centreline, rather than to the local centreline adjacent to the end (see Fig 31.2(b)) There is generally a design assumption that the line of action of the force in the column does not change direction at a braced joint by more than 1/250, requiring an end squareness in a simple bearing connection (relative to the overall axis of the member) of 1/500 (see Fig 31.2(c)) However, full contact bearing generally arises at column splices which are not at braced points, so an end squareness tolerance of 1/1000 is usually specified, producing a maximum change of slope of 1/500 (see Fig 31.2(d)) Once a column has been erected, it is more practical to measure the remaining gaps in a joint These gaps are affected not only by the squareness of the ends but also by the second criterion, flatness 31.4.3.4 Flatness Ends have to be reasonably flat (as distinct from curved or grossly uneven) to enable the load to be transferred properly Following a history of arguments over appropriate specifications, the American Institute of Steel Construction (AISC) commissioned some tests, which are the basis for their current specifications It was found that a surprisingly high tolerance was quite acceptable, and that beyond its limit (or to compensate for end squareness deviations) the use of localized packs or shims was acceptable Basically similar rules are now beginning to appear in other specifications including the CEN standard (see section 31.5.6 in relation to erection tolerances) This is an essentially simple and effective method of correcting excessive gaps on site (see also section 31.5.6) However, inserting shims into column joints is not a matter to be undertaken lightly It is normally more economic to avoid the need for shimming by working to close fabrication tolerances in joints where full contact bearing is required 1056 The Eurocodes countries were the various Recommendations published by ECCS From these were developed the initial draft Eurocode published by the European Commission, followed by the various parts of ENV 1993 issued by CEN The best known is Part 1.1 General rules and rules for buildings; this is what is often thought of as EC3 but is only the first of a total of 18 documents issued as ENVs However, attention must now turn to the forthcoming EN version, which will have full status as British Standard BS EN 1993 This is also being developed in a number of parts Part will probably be called Generic rules and will be sub-divided into 11 sub-parts dealing with different types of steel components Each of the other parts will then cover the application of the relevant generic rules to various types of structure The various parts are listed in Tables 36.1 and 36.2 At present only a minority of drafts are nearing completion, so the titles should be understood as indicating the intended scope, rather than precisely defining the final titles The position may change, but it seems probable that the total length of the nine application documents comprising Parts to will be far less than that of Part Table 36.1 Components of EC3 Design of Steel Structures Part – generic rules Part no Generic rules 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 1.11 Common rules Fire design Cold-formed thin gauge members and sheeting Stainless steels Stiffened plating subject to in-plane loading Steel shells Stiffened plating subject to out-of-plane bending Joints Fatigue Fracture toughness and through-thickness properties Cables Table 36.2 Application of the generic rules from EC3 to various structure types Parts 4.1 4.2 4.3 7.1 7.2 Application Bridges Buildings Tanks Silos Pipelines Piling Crane supporting structures Towers and masts Chimneys EC3 Design of steel structures 1057 36.3.3 Design rules Material resistance The design resistances of steel members and other steel components is generally related to the yield strength of the material However, in the case of both bolted connections and welded connections, the design strength is related to the tensile strength (i.e the UTS) This is also relevant when considering net cross-section resistance at bolt holes Thus generally the design resistance Rd of a cross-section needs to be related to a characteristic value of the yield strength fy using: Rd = Afy /g M in which A is the cross-section area For convenience, the characteristic yield strength fy is taken as equal to the specified ‘guaranteed’ minimum value for yield strength ReH given in the relevant product standard for structural steel, generally EN 10025 This is not the ‘true’ characteristic strength (based on 95% or any other statistically determined level of probability), which clearly will be higher More importantly, the value of g M needs to cover the variability of the section properties represented by A as well as that of the yield strength fy An ECSC funded study of the variability of section properties, yield strength and cross-section resistance using large numbers of samples of current production at a variety of European (including British) steel mills found that the mean value of A is approximately equal to the nominal value, but that the mean value of fy is about 1.2 times the nominal value In addition, the shape of the distribution curve showing its variability is markedly skewed, being noticeably truncated on the negative side, presumably due to the process control procedure Reliability analysis related to the EN 1990 target reliability b, and taking account of the actual distribution of yield strength and section properties, leads to a value of Rd related to the EN 1990 target reliability b = 3.8 that corresponds to a required value of g M of slightly less than 1.0 In simplified terms, the mean yield strength is about 1.2 times the nominal value: thus the mean resistance is also about 1.2 times the nominal resistance Afy (see Fig 36.1) The precise value of the characteristic resistance is not important, because the design resistance has been determined directly from the data in conformity with EN 1990 This design resistance turns out to be equal to the nominal resistance, hence g M = 1.0 It is important to note that this is only possible because this value of g M = 1.0 is used exclusively with the nominal ‘guaranteed minimum’ value of the yield strength For steel reinforcement the yield strength used in design is a characteristic value, and a larger value of g M is needed to obtain the design value 1058 The Eurocodes Table 36.3 The differences in axes between BS 5950-1 and the Eurocodes Axes Along the member Major axis Minor axis BS 5950-1 Eurocodes X Y X Y Z Table 36.4 Comparison of frequently used symbols in BS 5950-1 and the Eurocodes Area Elastic modulus Plastic modulus Inertia about major axis Inertia about minor axis Warping constant Torsion constant Radius of gyration Applied axial force Resistance to axial force Bending moment Applied shear force Shear resistance Yield stress Bending strength Compressive strength BS 5950-1 Eurocodes A Z S Ix Iy H J r F P M Fv Pv py pb pc A Wel Wpl Iy Iz Iw It i N NRd M V VRd fy cLTfy cfy Axes and notation Tables 36.3 and 36.4 summarize the differences in axes and notation that exist between BS 5950 and the Eurocodes Compression members As with BS5950-1, multiple strut curves are defined, depending on the type of cross-section and the axes of buckling The formulations for these differ only slightly from current UK practice However, the design process differs in detail.The member slenderness is first converted to a relative slenderness, as in BS 5400-3, thereby eliminating the need for the numerous pages of tables used in BS 5950-1 This leads to a reduction factor c which is applied to the cross-section resistance to obtain the member buckling resistance EC3 Design of steel structures 1059 Beams For restrained beams, the cross-section resistance is based on the plastic modulus, provided that the cross-section is classified as class (plastic) or class (composite), just as in BS 5950-1, except for detailed differences in the limiting values for the different classes For laterally unrestrained beams, there are alternative methods A method based on the column curve may be used for all members; this comprehensive approach gives lower resistances than those in BS 5950-1 for rolled sections There is a specific approach for rolled beams that gives higher resistances than those in BS 5950-1 The procedure is more similar to that of BS 5400-3 than to BS 5950-1 It also introduces the intermediate step of using the relative slenderness rather than the resistance to cover design cases with varying moments (In terms familiar to users of the 1990 version of BS 5950-1, it uses n-factors not m-factors.) Members subject to combined axial force and bending moments There are two methods A general method examines the overall lateral slenderness of the frame and then determines the resistance as a function of the ultimate resistance without buckling (e.g the plastic collapse load) reduced by a buckling factor A more complex interaction method is also available This considers the design of individual members separately It differentiates between members that are not susceptible to torsional deformations and those that are For each of these classes there are less complicated and more complicated methods Even for the simplest case, i.e members not susceptible to torsional deformation and treated more simply, the resulting equations are too complex for manual design and will require carefully prepared design aids, or software, before they can be implemented in practice without a significant increase in design time Connections Connections are treated in a separate part of EC3 and in considerably more detail than is given in BS 5950-1 In addition to conventional information on basic strengths, connection analysis is treated in considerable detail, at least for certain classes of connection There is explicit consideration of both the strength and stiffness of the connections and also a recognition of their effect on overall structural behaviour Frame stability Frame stability is checked by considering the effects of imperfection on both the global analysis and the performance of any bracing system For the former, the sway 1060 The Eurocodes imperfections at height/200 (or an equivalent horizontal force) are similar to BS 5950-1 Imperfections of the bracing system are both more rigorous and require more design effort than current UK practice Frame analysis The analysis is required to take account of the effects of the deformed geometry on the structure First order analysis may be used where the increases in internal forces and moments are less than 10% This implies a critical load ratio (lcr), under factored loading, of more than 10; it is equivalent to the BS 5950-1: 2000 ‘nonsway’ frame approach A range of second order approaches may be used for structures for which gcr is less than 10 These methods include: • • • • effective length approaches amplification factors energy methods formal second order analysis 36.3.4 Supporting standards Relevant supporting standards include those for: • • • • • • Steel properties Tolerances on steel sections and plates Ordinary bolts HSFG bolts Welding electrodes Fabrication and erection Steel properties Properties of structural steels are currently given in EN 10025, EN 10113, EN 10137, EN 10155, EN 10210 and EN 10219, the last two covering structural hollow sections A new edition of EN 10025 is planned that will combine all of these except steel for hollow sections (and may include these also) All of these have been adopted as BS EN standards Further BS EN standards are available covering steel sheets and strip, both plain and galvanized, for use in cold-formed sections and sheeting They are already in use in connection with BS 5950 Parts 5, 6, and EC4 Design of composite steel and concrete structures 1061 Tolerances on steel sections and plates Tolerances for plates and for each type of section are similarly given in numerous EN standards adopted as BS ENs Details are given in BS 5950 Part Ordinary bolts Bolts, nuts and washers are covered in ISO standards in three series commencing with ISO 4014, ISO 4032 and ISO 7089, that have been adopted as BS EN ISO standards HSFG bolts A new series of standards for high-strength bolts suitable for using as preloaded fasteners in friction grip applications has been under development for some years It is hoped these may be finalized and agreed by 2002, but this remains to be seen Welding electrodes A range of EN standards have been developed covering all types of steel welded electrodes, using common strength grades, and adopted as BS ENs Details are given in BS 5950-2 Fabrication and erection The rules for execution (fabrication and erection) of steel structures originally developed as part of EC3 under the aegis of the European Commission were separated from the ENV version of the design rules and became ENV 1090-1 Parts to of ENV 1090 were subsequently added to cover cold-formed thin gauge sections and sheeting, higher strength steels, structural hollow sections, bridges and stainless steels These are now being converted to EN status as a single EN 1090 36.4 EC4 Design of composite steel and concrete structures 36.4.1 Scope and contents Eurocode will come in three parts: 1.1 General rules for buildings, 1.2 Structural fire design and Bridges This section covers Part 1.1 and compares it with BS 5950 Parts 3.1 and It is based on the latest available draft so some details may change, 1062 The Eurocodes but the overall concepts are now well established For certain parameters recommended values will be given, and it will be left to national bodies to decide whether to accept these or choose something different Where necessary, recommended values have been used in this section The plan is for Parts 1.1 and 1.2 to be made available by CEN in the first half of 2003, with Part following a year later Eurocode covers more than BS 5950 Part 3.1 It includes rules for partially encased sections and, more importantly, includes guidance on composite columns As well as individual elements, the EC3 rules for moment connections are extended to composite joints The basis for the work was the ENV version published in 1994, but there have been significant changes since then Guidance on composite slabs equivalent to that in BS 5950 Part is included in EC4 Part 1.1 Taking an overall view EC4 will give very similar results to BS 5950 for composite beams and slabs For composite columns the resistance will be larger than that determined by current UK guidance One area where EC4 is more conservative than BS 5950 is in the capacity of stud shear connections 36.4.2 Design rules Materials The Eurocode covers a wider range of properties than BS 5950 Steel grades S420 and S460 are included, as are concrete grades up to a cube strength of 75 MPa For concrete, the limiting stress used in EC4 is 85% of the design strength According to EC2 the concrete design strength (fcd) is afck/gc, where fck is the cylinder strength The recommended value for the material safety factor gc is 1.5 To compare EC4 and BS 5950 you also need to account for the relationship between cube and cylinder strength Typical values, given in Table 36.5, show that the limiting stresses are very similar For steel, both for sections and reinforcement, the limiting stress in EC4 is the design strength For structural steel, EC3 recommends that this is equal to the nominal yield stress, but for reinforcement EC2 recommends a partial factor of 1.15 Table 36.5 Comparison of concrete strength in BS 5950 and EC4 Cube strength (MPa) 25 30 35 40 BS 5950 limiting stress 11.25 13.50 15.75 18.00 EC4 Normal weight concrete Lightweight concrete Cylinder strength Limiting stress EC4/BS 5950 Cylinder strength Limiting stress EC4/BS 5950 20.0 25.0 28.6 31.9 11.33 14.17 16.21 18.10 1.01 1.05 1.03 1.00 22.5 27.0 32.0 36.7 10.84 13.00 15.41 17.68 0.96 0.96 0.98 0.98 EC4 Design of composite steel and concrete structures 1063 on the characteristic strength The limiting stresses are therefore identical to those in BS 5950 Beam design For simply-supported secondary beams in buildings, the effective width limit of span/8 is the same as BS 5950 For primary beams, EC4 does not include the ‘0.8b’ limit in the BS For continuous beams the equivalent spans are slightly different from those in the BS, and there is no simplified method for calculating moments Redistribution of moments in continuous beams is allowed in EC4, with similar limits depending on the class of steel section The plastic capacity of beams is based on rectangular stress blocks with the limiting stresses given above For high-strength steels, i.e S420 and S460, there are additional requirements The capacities will therefore be very similar to those calculated to BS 5950 There is no requirement to check stresses under working loads for beams in buildings When calculating deflections, partial interaction only needs to be taken into account if it is less than 50% However, shrinkage must be considered, unless the ratio of the span to overall depth is less than 20 and the free shrinkage strain of the Fig 36.1 Rectangular stress blocks for simply-supported composite beams 1064 The Eurocodes concrete is less than 400 ¥ 16-6 This will affect a lot of beams, particularly those with lightweight concrete Shear connection The ENV version and initial drafts included many types of shear connections It was decided to limit the standard to headed studs Other connectors are allowed but no application rules are given Expressions are given for the capacity of headed studs The governing expression is usually that based on the concrete but there is a limiting value based on the strength of the steel in the stud For lightweight concrete, the capacity is reduced in the ratio of the density of the concrete compared with the 10% reduction given in BS 5950 For the typical 19/100 mm stud, characteristic capacities in kN are shown in Table 36.6 To calculate the design capacity, the recommended partial safety factor is 1.25 The relationship between design strengths to EC4 and BS 5950 will therefore be exactly the same for positive moment regions For negative moment regions, the 0.6 factor in BS 5950 means that the design capacity to EC4 will be relatively higher Like BS 5950, EC4 includes reduction factors for the stud capacity based on the geometry of the decking For decking parallel to the beam the factor is identical to that in the BS For transverse decking the factor is generally similar but there are two differences Instead of the coefficients 0.85, 0.6 and 0.5 for one, two and three or more studs, the coefficients work out as 0.7 for one stud and 0.5 for two or more There is also a limiting value of reduction factor for decks with a thickness less than or equal to 1.0 mm This limiting value is 0.85 for one stud and 0.7 for two or more studs This additional limit is based on tests in the UK For re-entrant profiles the main difference is likely to be the limiting value due to thickness For certain trapezoidal decks there will be a smaller reduction factor, especially for two studs per trough Transverse shear The ENV version of EC4 had a similar expression for transverse shear to that in BS 5950 For the EN version it has been decided to use the expressions for concrete Table 36.6 Comparison of shear connector capacities Cube strength 25 30 35 40 Normal weight concrete Lightweight concrete BS 5950 EC4 EC4/BS 5950 BS 5950 EC4 EC4/BS 5950 95 100 104 109 81.0 92.9 100.9 102.1 0.853 0.929 0.970 0.936 85.5 90.0 93.6 98.1 71.2 79.7 88.5 96.3 0.833 0.885 0.945 0.982 EC4 Design of composite steel and concrete structures 1065 tee-beams in EC2 The transverse shear capacity is typically dependent on the amount of reinforcement, though there is a cut off value that varies with concrete strength The decking can contribute to the transverse shear capacity when it is perpendicular to the beam The contribution is similar to that in BS 5950 but it only reduces the amount of reinforcement required and does not increase the cutoff strength based on the concrete strength The variation of the transverse shear capacity with percentage reinforcement for normal and lightweight concrete is shown in Figures 36.2 and 36.3 It can be seen that the transverse shear capacity to EC4 is typically greater than that to BS 5950 Composite slabs EC4 Part 1.1 also covers composite slabs with profiled sheeting for buildings The guidance is very similar to that in BS 5950 Part but there are some differences in the details There is no specific mention of testing as an alternative to the application rules as a design procedure The bearing distance at end supports for the slab should not be less than 70 mm for bearings on steel or concrete and 100 mm for other materials This is for the slab; the bearing for the decking can be less, i.e 70 and 50 mm respectively The bending capacity of the slab is based on similar limiting stresses to those used for composite beams, 0.85fed for concrete and fyp,d for the decking A continuous slab may be designed as a series of simply supported spans In this case the minimum reinforcement is 0.2% of the concrete area above the ribs rather Fig 36.2 Treatment of transverse shear for normal weight concrete to EC4 and BS 5950 Part 3.1 1066 The Eurocodes Fig 36.3 Treatment of transverse shear for lightweight concrete to EC4 and BS 5950 Part 3.1 than 0.1% of the gross concrete area as given in BS 5950 Part Where the decking is propped, the minimum reinforcement is 0.4% To calculate the shear bond capacity the m–k method can be used in an identical manner to BS 5950 but no increase in capacity is allowed due to end anchorage An alternative ‘partial connection method’ is given which is similar to the design of beams with partial interaction This method relies on a design shear bond strength tu,Rd, which is determined by test End anchorage can be used to increase the capacity with the partial connection method The effective width for point loads is identical to that given in BS 5950 Part However, it is stated that, for concentrated characteristic loads greater than 7.5 kN, the slab must be specifically designed to spread the load with the addition of appropriate reinforcement For vertical shear, reference is made to EC2 The critical perimeter for concentrated loads has the same overall size as that in BS 5950 but has rounded corners Composite columns The guidance on composite columns given in EC4 Part 1.1 should be welcomed as it will give a significant increase in capacity compared with the method in BS 5950 Part For columns subject to axial load only, the effect of slenderness is calculated by using a reduction factor from EC3, i.e it is designed as an equivalent steel column This is a similar method to BS 5950 The calculation of the appropriate slenderness is more accurate but more complicated Implications of the Eurocodes for practice in the UK Fig 36.4 1067 Strength of composite column cross-sections to EC4 For members in combined compression and bending, a ‘local capacity’ check is carried out for axial force and moment where the moment includes effects due to slenderness as shown in Fig 36.4 The local check is based on rectangular stress blocks with the normal limiting stresses and produces an interaction curve similar to that for reinforced concrete columns Composite joints The section on joints in EC4 Part 1.1 covers moment connections where there is some composite action This can be by the use of reinforcement in the slab to take tension or by encasement of the column to increase the shear and compression capacity of the column web Rules are given for the strength and stiffness of these joint components The stiffness values are in an ‘informative’ annex to reflect the fact that these are estimates and not hard and fast values 36.5 Implications of the Eurocodes for practice in the UK Extensive calibrations of the design rules in the various parts of EC3 and EC4 have been carried out in connection with their drafting, as part of the process of determining the values of gM needed to provide the uniform level of reliability represented by the target reliability b given in EN 1990 1068 The Eurocodes In addition, the UK authorities carried out extensive comparisons with current UK design practice in connection with the National Application Documents at the ENV stage These studies have shown that the rules provide a suitable uniform level of reliability As far as design resistances are concerned, although there are naturally some variations, on average the EC3 and EC4 design resistances are comparable with those of the relevant BSs and the other design procedures identified as representing current UK practice for forms of structure not yet covered by a BS At the ENV stage, the partial factors gF to be applied to loads (‘actions’) were established and each country was invited to fix its own values for gM For the EN stage, both the gF and the gM factors are to be determined on a national basis At some point a decision will have to be made about which level of reliability all the structural Eurocodes will be required to achieve in the UK The relative economy of each Eurocode compared with current UK practice will depend upon this decision 36.6 Conclusions The scopes of Eurocode and are very wide in terms of types of structure, forms of construction, methods of design and materials Much of their content has not been covered previously, particularly in Eurocode In relation to common types of building using rolled steel sections, the differences in results are only minor, though the design procedures are sometimes more tedious The real differences compared with existing British Standards lie in the new methods For buildings, detailed methods for semi-rigid joints are given For cold-formed steelwork, more advanced methods of design are included Rules for stainless steels appear for the first time in a standard The rules for shells and for the design of piles, sheet piling and silos are all new By presenting all the rules for as wide a variety of steel and composite designs as possible in a consistent format, the specialist designer will be given a greater versatility than ever before Appendix Steel technology Elastic properties of steel European standards for structural steels 1071 1072 Design theory Bending moment, shear and deflection Bending moment and reaction Influence lines Second moments of area Geometrical properties of plane sections Plastic moduli Formulae for rigid frames 1077 1102 1105 1116 1124 1127 1130 Element design Explanatory notes on section dimensions and properties, bolts and welds Tables of dimensions and gross section properties Extracts from BS 5950: Part 1148 1166 1220 Connection design Bolt data Weld data 1236 1266 Other elements Piling information Floor plates 1274 1280 Construction Fire resistance Corrosion resistance 1282 1308 Miscellaneous British and European Standards for steelwork 1311 1069 ... erection: hot rolled sections Part 7: Specification for materials and workmanship: cold formed sections and sheeting (2) National structural steelwork specification for building construction NSSS,... may actually measure 401 mm wide by 3 87 mm deep one side, and have a depth of 381 mm the other side The same is true of continental sections A 400 ¥ 400 ¥ 2 37 HD also has a nominal size of 381... detailed design) under two headings: (1) Site construction (table of BS 5606) (2) Manufacture (table of BS 5606) Unfortunately many of the values for site construction of steelwork are only estimated