VOLUME - 1 NOL Fad DESIGN OF STEEL STRUCTURES Third Edition Edwin H. Gaylord, Jr. Professor Emeritus of Civil Engineering University of Hinois at Urbana-Champaign Charles N. Gaylord Late Professor of Engineering of Virginia James E. Stallmeyer New York St. Louis Sa Lisbon London Madrid Pacis San JoanDESIGN OF STEEL STRUCTURES INTERNATIONAL EDITION 1992 78910 BE 20 ‘When ordering this tite use ISBN 0407-112623.6 eee ABOUT THE AUTHORS EER Edwin H. Gaylord, Jr. is professor emeritus of civil engineering at the Universi of Illinois at Urbana-Champaign. He is a graduate of Witenberg University, Case Western Reserve University, and the University of Michigan, He has been involved in teaching, research, and cons: gineering, particu- the area of steel structures, In addition to his coauthorship of “Design of Steel Structures” he is coauthor, with Charles N. Gaylord, of “Design of Steel Bins for Storage Bulk Solids,” coedi aylord, of * Struct ria and Loading struction Come tee on Specifications from 1959 to 1986 and in 1988 received the Ins Special standing contribution to the art of buil structural steel. He is a life member of the Structu James E. Stallmeyer is professor of civil engineering at the University of Illinois at Urbana-Champaign. He is a graduate of the University of Illinois. He has been involved in teaching, research and consulting in structural engineering; particu- fhway and Transportation Officials and In 1964 he was awarded the AdamsCONTENTS Preface Notation Loads on Structures 1-1 Engineered Structures ‘The Design Procedure Loads ive Loads on Building Floors Live Loads on Bridge Floors Impact Snow Loads Wind Loads Earthquake Loads Probabilistic Considerations of Safety Factor of Safety, Load Factors, and Resistance Factors Codes and Specifications Structures, Metals, and Fasteners Buildings Structural Bolts Bolted and Riveted Connections Welding Processes Welded Joints 2-12 Stresses in Weldsili conrests -13 Specifications for Welded Connections 14 Welding Quality Control 15 Fatigue 16 Brittle Fracture 17 Corrosion 2 2 3 Tension Members 3-1. Effect of Residual Stresses ‘Types of Tension Design Stresses for Base Material ‘ansmission in Connections and Splices 1 Long Joints Fasteners ting for Reduction in Cross-Sectional Area, 18 Secondary Stresses in Truss Members 5-19 Further Diseussion of Residual Stresses 4 Compression Members 41 Introduction 42. Elastic Buckling of Columns 43 Effect of Initial Crookedness. 44 Inelastic Buckling of Columns 45 Columns with Ends Rotationally Restrained 46 ffect of Residual Stresses 47 Determinants of Column Strength 48 Allowable-Stress Formulas for Steel Columns 49. AISC/LRFD Formulas for Steel Columns 4-10 Typical Sections for Compression Members 4411 Effective Length of Columns in Frames 4-12 Local Buckling 413 Local Buckling with Residual Stresses 4:14 Design Procedures for Local Buckling of Steel Columns 415 Design of Columns 4:16 Column Bases 417 Shear in Columns 4-18 Effect of Shear on Critical Load 419 Structural Members in Torsion Buckling of Compression Members £21 Columns with One Axis of Symmetry au 216 218 220 2 229 233 24 237 249 252 255 259 263 268 ow 4:22 Columns with No Axis of Symmetry 423 Aluminum Columns Beams Ss 32 Introduction Beam Cross Sections Behavior of Beams Fioles in Beam Flanges Shear in Beams 5-14 Shear Buckling of Beam Webs ‘5:15 Bend Buckling of Beam Webs 5:16 Combined Shear and Bending of Webs fing and Vertical Buckling of Webs 5-18 Beam Bearing Plates 519 Biaxial Bending 5-20 Built-Up Beams 5.21 The Shear Center 5.22 Composite Beams 5-23 Design of Composite Beams 5-24 Shear Connectors 5.25 Continuous Composite Beams Beam-Columns 1 62 63 ry 63 66 67 68 63 6 Beam-Columns Columns with Maximum Moment at One End Beam-Columns: General Case Buckling of Frames with Loaded Beams ‘Torsional Buckling of Beam-Columns ‘Design of Beam-Columns ress Formulas for Beam-Columns Formulas for Bearm-Colurmns |AISC/ASD) anid DP6-8-1 (AISC/LRED) nalyses of Frames Plate Girders MH 42 1 76 Introduet Bend Buckling of Plate-Girder Webs Propor Length of Flange Platesx coNTENTS 1-7 Shear Buckling of Plate-Girder Webs ath of Plate-Girder Webs Shear and Bending in Webs 7-1 Bearing Stiffeners 7-12 Hybrid Plate Gisders 7-13 Webs of Hybrid Girders ‘7-14 Flange Buckling in Hybrid Girders 7-15 Welding of Girder Components 7-16 Shop and Field Splices 8 Connections Introduction Bolted and Riveted Connections for Beams Unstitfened Beam Seats Stiffened Beam Seats Bolted or Riveted Framed Connections in Tension h Fasteners in Tension ion and Shear int Connections with Fasteners in Tension Prying Forces in Moment-Resistant Connections Rivets and A307 Bolts in Tension Resistant Connections 9 of Standard Connections Types of Construction Welded Framed Beam Conné Design of Welded Framed Connections -18 Unstiffened Welded Beam Seats 8:19 Stiffened Welded Beam Seats 8:20 Moment-Resistant Welded Beam Connections 8.21 Stiffeners in Beam-to-Column Connections 8-22 Eoventrically Loaded Bolted or Riveted Shear Connections — Blastic Analysis ly Loaded Bolted or Riveted Shear Connections — timate-Strength Analysis 8.24 Ecventrically Loaded Welded Shear Connections—Elastic Analysis 8-25 Eoventrically Loaded Welded Shear Connections— Ultimate-Sirength Analysis 8-26 Haunched Connections 8-27 Column’ 8-29 Base-Plate Details 8-30 Pinned Connections 9 Plastic Analysis and Design 841 nroduction 9.2. Development of Collapse Mechanises 9.3. Determination of Collapse Mechanism 94 Continuous Beams ism Analysis by Virtual Displacements 9-7 Analysis of Rectangular Frames 9.8 General Procedure for Mechanism Analysis of Frames 9.9 Two-Bay Frame—Cor ‘Member Strength Local and Lateral-Torsional Buckling, Beam-Columas 15 Shear Resistance 16 Defiections 17 Moment Balancing 18 Two-Bay Frame by Moment Balancing. 9:19 Frame Buckling 10 Stability and Strength of Flat Plates 10-1 Stability and Strength of Flat Plates 10-2 Postbuckling Strength of Flat Plates 10-3 Additional Formulas for Bifective Width 10-4 Specification Provisions for Postbuckli Plate Elements 10.5 Interaction of Local Buckling and Column Bend Buckling 10.6 Specification Provisions for Interaction of Local Buckling and Columa Bend Buckling 10-7 Discussion of Examples of Art, 10.6 1048 Postbuckling Strength of Beams 11 Steel Bridges ALT Introduction Economics of Simple-Span Bridges Bracing ‘Weight of Bridges Bridge Floors Design of Floor Slabs Design of Stringers and Floor Beams End Bearings DPI1-9-1: Eldersburg-Louisville Road Bridge Beam Bridges “Li Composite-Beam Bridges Length of Cover Plates -13 Shear-Connector Spacing 11-14 DPII-14-1: Eldersburg-Louisville Road Bridge 12. Buildings 12 Behavior of contents. xi 632 63Floor Construction Roof Systems ind Bracing for Tall Buildings ind-Bent An ‘The Cantilever Mothod 4 Limitations of Portal and Cantilever Methods Appendix Table Al Approximate Radii of Gyration Table A2 SI Conversion Factors Indexes Name Index Subject Index 781 785 ee PREFACE — OE This book deals with the design of steel structural members, and their connec- tions, with emphasis on their use in bridges and buildings. Discussion of theory and behavior under the various combinations of loads such members must resist is followed by a discussion of applications according to standard specifications for load-and-resistance factor design and allowable-stress design. Discussions of se two types of specifications are presented in separately numbered articles lowed by worked-out examples. This will te matters for those who may ‘wish to cover only load-and-resistance factor design, only allowable-stress design, or both. In addition to a number of worked-out examples there are 32 design prob- Jems (labeled DP), which was a feature of previous editions. These examples help to show structural members as components of real-world structures rather than ‘ed elements. Practically all DPs and worked-out examples are presented in both load-and-resistance factor design and allowable-stress design. DPs are especially useful in helping the student prepare flow diagrams and computer mentals of the design procedure and the sequence of the calculations involved much than if they are presented in the form of flow charts and computer Programs. It has been the authors’ experience that students who rely on flow diagrams and computer programs prepared by others do not develop a capacity for critical evalu of the procedures and their resulting output. Another feature is the pr mn and discussion of the design calculations for the truss spans and composite-beam approach spans of a highway bridge in Maryland. Discussion and applications AISC specifications are based on the Specification for Structural Stee! Buildings: Allowable Stress Design and Plastic Design, Jun '9 and the Load and Resistance Factor Design Specification for Structural Steel Buildings, September 1, 1991, Discussion and applications of the AASHTO Standard Specification for Highway Bridges are based on the 1989xiv raerace References to AREA Specifications for Stee! Railway Bridges are to the ition authors suggest that the first course in steel structures be based on Chapters 1 to 3; Ai Chapter 5; and 6-3 and 6-6 to 6.9 of Chapter 6 or S-1 to 8-5 and 8-16 to 8-19 of Chapter 8. Special thanks are extended to Prof. W. J. Hall, for his support and encour- agement during the preparation of this edition. Suggestions by Profs. Hall and Scott Schiff, who used parts of the manuscript for their classes, and the assistance of Scott Greer in the preparation of examples in Chapter 6, are grat acknowledged. We are also grateful to Nestor Iwankiw of the American In of Steel Construction for alerting us to forthcoming changes in the Ins specifications. McGraw-Hill would like to thank Terry Hand, US. Military Academy and Ralph Mozingo, Penn State University for their many helpful comments and suggestions in reviewing the manuscript. Edwin H. Gaylord, Jr. James B, NOTATION Abbreviations ket ips per cubic foot Kit kips per lineal foot Kli_kips per lineal inch ksf_kips per square foot ksi kips per square inch mph miles per hour pef pounds per cubic foot pif pounds per lineal foot pli pounds per lineal inch psf pounds per square foot psi pounds per square inch AASHTO American Association of State Highway and Transportation Officials AISC_ American Institute of Steel Construction AISI American Iron and Stee! Institute ANSI American National Standards Institut AREA American Railway Engineering Association ASCE American Society of Civil Engineers ASTM American Society for Testing and Materials ding Officials and Code Administrators National Building Code NIST National Institute of Standards and Technology RCSC_ Research Council for Structural Connections SSBC Southern Standard Building Code SSRC_ Structural Stability Research Councilee eer Nomenclature Bae wa PO p > > AB UBC Uniform Building Code WRC Welding Research Couneil area area of bolt effective net area area of beam flange gross area net area of tension member area of steel section in composite beam; arca of longitudinal stiffener; shear area of bolts in connection cross-sectional area of stud shear connector area of transverse stiffener area of beam or plate-girder web length of plate; distance between vertical stiffeners bending factor (area of cross section divided by section modulus) amplification factor spacing of stringers in bridge floor; width of flange; width of plate; length of bearing on beam web; effective width of con- crete flange of composite beam effective width of plate or web for deter strength compressive fo drag coefficient lift coefficient lateral-torsional buckling coefficients moment-gradient coefficient for lateral-torsional buckling of beams column slenderness ra ing postbuckling ; fundamental period coefficient viding elastic and inelastic buckling coefficient of bending term in beam-column interaction formula shape factor determining effective wind pressure ratio of critical shear stress to yield shear stress warping constant of eross section distance from neutral axis to extreme fiber of beam dead load diameter of hok diameter of bolt column-web depth between fillets; web depth between fillets diameter of rocker; depth of cross section; NoTATION xvii diameter of stud shear connector (Young's modulus) of concrete double modulus (also called reduced modulus) E, strain-hardening modulus E, tangent modulus ty of load; natural logarithm base allowable Euler column stress weld electrode strengt axial force in stiffener range of stress for fatigue loading stress clastic critical shear stress inelastic critical shear stress. calculated stress f calculated axial compressive stress cd bending stress J. calculated compressive stress Le Specified 28-day compressive strength of concrete J, postbuckling edge stress in plate J, calculated bearing stress Jo calculated shear stress G modulus of elasticity in shear (modulus of rigidity); center of gravity; ratio of sum of column stiffnesses to sum of beam iffnesses at joint in frame G, tangent modulus in shear9 sage (transverse spacing of fastener lines); distance from shear ion of load; acceleration depth of web of beam or plate girder; depth of aoe impact ratio; moment of inertia; occupancy importance factor effective moment of inert polar moment of inertia polar moment of inertia with respect to centroid polar moment of inertia with respect to shear center moment a of stiffener product of inertia torsion constant of exoss section spring constant; effective-length coefficient cffective-length coefficient of laced or battened column net-section efficiency coefficients distance from flange outside face to toc of web fillet; coefficient in plate-buckling formula; //P/ET: weld geometry parameter flange local-buckling parameter length; story height basic live load length of channel shear connector spacing of braces center-to-center spacing of battens length moment smaller of end moments on unbraced length larger of ead moments on unbraced length critical moment moment due to lateral trans! critical moment in pl nominal resis moment with no lateral translation moment due to transverse loads on beam column plastic moment limiting elastic-buckling moment simple-beam moment ultimate moment; factored moment yield moment bolt pretension; cycles to failure; normal force; length of bearing factor of safety; number of fasteners; shape factor for ctoss tion in shear; E/E, worarion xix load ‘column critical load rag component of wind Euler column load lift component of wind that an event will be exceeded at least once in column twist-buckling load jamn tangent-modulus load ate load ; factored axial load ‘imate load plastic axial load pressure per unit of area; allowable bearing pressure of rocker on bearing plate shear; prying force; first moment of area in shear-stress formula coefficients used in evaluating postbuckling allowable toads static strength of shear connector velocity pressure (also called dynamic pressure or stagnation pressure); allowable load on shear connector; shear per lineal inch strength of shear connector reaction; shear force on fastener; return period (also called recurrence interval) of snow, Wind, ete.; reduction in live load on building floors, percent; ratio of maximum stress to mini- reduction factor for post-bend buckling strength of fastener; strength of weld element structural system coefficient radius of gyration; rate of reduction of live load; radius of hole equivalent radius of gyration for lateral-torsional buckling radius of gyration of chord of battened column polar radius of gyration with respect to shear center radius of gyration of section consisting of compression flange and one-third the compression web area equivalent radius of gyration for torsional buckling equivalent radius of gyration for twist-bend buckling ion modulus; shear center; effective span of bridge fatigue strengthNorarion xxi XX Noraniow strain at onset of strain hardening Sse compression-flange section modulus «yield strain Sqq_tension-fiange section modulus @ rotation of joint in frame or of beam at support; angle of twist 5 pitch (distance, in the direction of the gage lines, between two pet unit length; angle of plate-girder panel diagonal with horic successive holes) zontal; mechanism angle; acute angle between shear force and | T tensile force; torsional moment; fundamental period of vibra- ied aig | tion 4. slendemess parameter specified pretension of high-strength bol 4. column slenderness parameter | effective throat of fillet weld or partial-penetration groove weld 4, slenderness parameter to attain M, St. Venant torsional resistance slenderness paramete in M, at torsional resistance due to nonuniform warping Poisson's ratio; coefficient of friction Paine radius of curvature; mass density of air babtaderenti standard deviation web thickness ratio of tangent modulus to Young's modulus shear strength inclination of tension field in plate-girder web; rotation at end reduction coefficient shear; velocity; coefficient of variation critical shear in plate-girder web shear force in composite concrete slab factored shear force shear component of plate-girder web tension field shear strength of beam or plate-girder web velocity y y A > i W load on beam, kips (| wm W, L Sl 4aye resistance factor for bending resistance factor for columns ae external work during virtwal displacement internal work during virtual displacement wind load distributed load on beam, kips per foot; unit weight of concrete W, gross width of part in tension ion Xo distance from shear center to centroid Yo deflection duc to transverse loads on beam columns; distance from shear center to centroid Plastic section modulus; zone factor fastener deformation; weld deformation ry translational deflection displacement; deffection of column at midlength idlength displacement due t € unit str €, residual strain crookednessCHAPTER oa LOADS AND STRUCTURES 1-1 ENGINEERED STRUCTURES Engineered structures are of such variety that they defy any attempt to enumer- ate them except in a general way. The countless problems which arise in their design have prompted engineers to specialize in the design of particular struc- tures or groups of related structures, and it is profitable to study design some- what according to customary areas of specialization. Although the complete design of many structures is the result of the coordinated efforts of se branches of engineering, we refer sometimes to the design of a structure having rind only that part of the design which comes wi province of one of the branches. ‘Among the structures that are designed by civi ings, transmission towers, hway pavements, and aire up of structures is large for convenient study as a unit. In this book we will limit ourselves to a study of structural members in metal and the methods by which connected, together with applications to the design of bridges and angineers are bridges, 16 walls, docks,2. esiow oF sree sraucrunes to be bui he designer must create a plan whi a suitable arrangement of rooms, corridors, s ete.; which will be aesthetically acceptable; and which can be b «is prepared to pay. This phase of design, som imctional planning, calls for a designer with a high order Wand ink of its development as extent to which the scheme lanning stage depends upon the struc- n of the columns in a building usually must be functional plan, and sufficient space must be anticipated between finished ceiling and finished floor of adjacent stories to accommodate the floor construction, The functional plan and structural scheme of a highway bridge usually are not so interdependent. The roadway grade and highway bridge are influenced pr by clearance requiremen respect, to whatever he necessity of providing adequate approaches ch depends upon the number of lanes . Many different types of bridge can be adapted to a given functional plan. is usually necessary to make tentative cost estimates for several prelimi- nary structural layouts. may have to be done while the Fun: plan is being developed; sometimes it can be done later. Sel materials must be based upon consideration of a and the corresponding skilled labor, relative costs, and wage scal } materials for the structure, Successful development of an efficient scheme hinges on the engineer's familiarity with the many types of i structure which have been developed in the past. On the other hand, the designer who leans too heavily on tradition may fail to see the possi new and better solutions. ‘The third stage of the design is a structural analysis. Although design spe- cifications and building codes usually prescribe the nature and magnitude of the loads to which the structure may be subjected, at times the engineer must make loads are defined, a structural analysis must be made to 8 which will be produce various members of the framework. Although this is a fairly routine procedure, simplifying assump- tions must invariably be made before the principles of mechanics can be applied. The designer must be conscious of his or her assum structure as designed can be expected to behave accordingl fh phase of the design the engineer prop the structural system. They must be chosen so that they will be able to resist, th an appropriate margin of safety, the forces which the structural analysis has sed. Familiarity with the methods and processes of fabrication (and their ions) and with the techniques of construction (and their limitations) is indispensable. tosos ano staucrures 3 The four steps in structural design discussed above are seldom, if ever, dis- tinct, and in many cases they must be carried along more or less simultaneously. Furthermore, degrees of importance relative to one another. Design is necessarily a trial-and-ertor procedure, Most structures are stati- cally indeterminate and require that member properties be specified before the is for the load effects can be carried out, After the member forces have been determined, the val mn must be evaluated. If changes in member properties be cattied out. The procedure must be repeated u ‘ant member forces are in acceptable agreement. The development of the computer has greatly facilitated this phase of the design process, but the judgment and experience of the designer ace impossible to build into a completely logical system as required by the com- pater. Development of procedures for translating design specifications into com- Puter programs continues to receive the attention of engineers in many sp Some sp er the enging fanding of the require- ments of the specifications, the method of analysis employed in the program, and the behavior of the many types of structural members, 1.3 LOADS ‘The weight of a structure is called dead foad. It can be determined with a high degree of precision, although not until after the structure has been designed, For this reason it is necessary to estimate dead load before a made 0 that the part of :xperienced designers can ofte arts with good accuracy. The actual weight should be determined and compared mn made if the difference is significant. The designers of the ill-fated first Quebec bridge across the St. Lawrence River neglected this point. After the collapse of the bridge during erection, with the resulting death of more than 100 workmen, investigation disclosed dead loads 20 to 30 percent larger than those assumed for the design. Although this discrepancy would have handicapped the bridge had it sdy or unsteady; they may be fixed, movable, or slowly or suddenly; and they may vary consider~ ably in magnitude, The history of some live loads—notably the weights of the heaviest highway trucks—is one of more or less continual increase in magnitude The following loads usually must be considered 1. The weight of the structure 2. The weight of people, furniture, ma 3. The weight of traffic on a bridge 4. The weight of snow ery, and goods in ab4 Eston oF sree srrucTURES 5. Dynamic forces resuiting from moving loads 6 Dynamic forces induced by wind and earthquakes 7. The pressure of liquids in storage vessels & Forces resulting from temperature change if expansion and contraction are impeded 9. The pressure of earth, as on retaining walls and column footings ‘The primary effect of gravity loads on structures is calculated from their weight; ie, they are considered to be static loads. However, live loads in motion may produce forces that are considerably greater than those resulting from the same loads at rest. These are the dynamic forces mentioned in category 5 above. i led impact if the effect is equivalent to additional gravity load and lateral ot longitudinal force (depending upon its diree. tion relative to the path of the vehicle) when the r equivalent to load in the horizontal plane. Lateral force may result from motion in a curved path (centrifugal force) or from the nosing of a ft track. Longitudinal forces are caused by acceleration and deceleration of moving vehicles The determination of the loads for which a given structure or elass of struc- {ure should be proportioned is one of the most difficult problems in design be answered: What loads may the structure be called its lifetime? In what combinations may these loads da possible but highly improbable load or com: bination of loads be allowed to dictate the design? The probal live load will be exceeded at some time dur depends on the period of exposure (life) ofthe design load. For example, locality only once in so many years in the long run. This period is called mean return period or mean recurrence interval. Such periods can be determined by statistical analysis of snowfall records. Of course, extremes of other natural Phenomena such as wind or flood also occur infrequently, and return petiods for specific extremes can be determi an extreme snowfall, wind veloc- lity that the extreme value will be exceeded in any one year. i period R of a wind speed of, say, 90 mph at a certain local 100 years, the probability that there will be @ wind speed greater than 90 mp! any one year is 1/R = 1/100 = 0.01. However, the probability which in choosing a wind velocity for design purpes design speed will be exceeded in any one year but r: will be exceeded during the life of the structure. This probability can be deter. yy that the specified Loans ano sraucrunes 5 It not be exceeded. Then rn years, where n is the life of y P, that it will be exceeded ~ RY. Therefore, the probal the m years is p=t-(1-4 J =t—-(1- As an example, suppose a structure which is expected 10 have a life of 50 years is to be built in the locality mentioned above, where the mean recurrence interval of a wind speed of 90 mph is 100 years. The probability that the structure will encounter a wind speed exceeding 90 mph du at least once Po = 1 ~ (1 — 0.01)" = 1 — 0.60 = 0.40 ‘That is, there is a 40 percent chance that the structure will be exposed to a wind exceeding 90 mph. If this is an acceptable risk, it is sufficient to design the build. ing to resist the pressures from a 90 mph wind. There will be a margin of safety the design, of course, and so the structure would not be expected to collapse under a wind of this velocity 14 LIVE LOADS ON BUILDING FLOORS gs serve such diverse purposes and present such random arrangements of | equipment and persons as to make it extremely dilfic able design loads, Although a number of systemat there is still a lack of adequate data. Many muni of buildings through building codes id requirements as well as other factors pertaining to design, ions have issued codes for national or regional use. Among lowing: Uniform Building Code, sponsored since 1972 by the International Con- ference of Building Officials, Pasadena, California. This code is used widely in the western states, Sor ing Code, sponsored since 1945 by the Southern Building irmingham, Alabama. This code is used in the southern and southeastern states, BOCA National Building Code, sponsored since 1950 by the B Off. sials and Code Administrators International, Chicago. This code is used in the eastern and north central states. Live-load recommendations are also published by the American Society of Building Code Requirements for Minimum Design ings and Other Structures, ASCE 7-88 (formerly American National Standards Institute ANSI A58,1-1982).6 bestow oF stest ermucrones Buildings may be classified according to occupancy as follows: ring, fabrication, assembly) 7. Storage (warehouses) Except for studies of combustible contents made in connection with fire- live loads which are an actual weighing of Materials and Structures Report 92, Fire-R: Construct tion which served as a library. These figures are probably good approx- ‘mations of total contents excluding persons, A residential room containing 1 person to every 6 ft’—surely adequate allowance for a crowd—would average, say, 25 psf. At the most, then, we might expect resident ‘occupancy. Several surveys of crowded hospital wards have been made, and even those wards which contained 1 bed for every 30 ? supported average live loads of only 9 psf, Building codes are in substantial agreement on 40 psf as a minimum live load for ional private rooms. There have been a number of investigations of live loads due to crowds of people, At the University of Iowa, students packed for the purpose of testing dynamic loads on balcony construc tions of normal loading conditions on the elevators of Grand Cent New York City showed live loads of about 100 psf. In a test by the Milwaukee Board of Education 0, a room normally intended for 48 pupils was crowded i ‘he resulting live load, including furniture, was 41.7 psf. The range of minimum- live-load requirements for assembly occupancy, as specified by various building codes, is given in Table 1-1. The table is not intended to be complete but rather ington? In the former building, 20 psf or less of actual average found on 70 percent of its area, 40 psf or less on 88 percent, and 60 psf of less on 96.5 percent. The maximum average live load of 106 psf occupied 825 ft? (0.5 Percent of the total area), In the latter building, 95 percent of the floor area Loans AND sraucroRes 7 TABLE [1 Live loads for ty Range in values requited by building codes ‘Type of space Live load, pst Schoo! classrooms (xed 40-60 (movable seats) 40-100 ‘Assembly halls (xed sets) 50-60 ‘Assembly halls (movable seats) 10 necessarily balconies) 50-60 Dance halls 400-120 supported an average load of 20 psf or less, 97.8 percent supported 40 psf or less, and 99,5 percent supported 60 psf or less. The maximum average live load of 90, psf was found on 1176 ft? of the tenth floor (0.5 percent of the total area). Some- ‘what similar prior studies of the Equitable Building in New York City disclosed average loads of 11.6 psf over three selected floors. The maximum load was 78.3, psf, the minimum 0.87, The National Institute of Standards and Technology (omerly the National Bureau of Standards) surveyed their Ad ing and the U.S. Civil Service Commission Building. The largest load intensity found was 72.5 psf. Building-code requirements for office-space build- ings range from 50 to 80 psf. Studies, similar to those mentioned above, of mercantile, industrial, and storage occupancies are reported in Ref. 4. The studies included two department stores, two mattress factories, one men’s clothing factory, one dress factory, two furniture factories, one newspaper plant, one printing plant, and two warehouses, ‘As might be expected, wide variations in live load were found, For exam ries was only 41 psf on 0.3 percent it was 101 psf on 11.1 percent of the area, The latter load is somewhat misleading, because it was in a cotton storage area, Maximums for the furniture factories were 98 and 120 psf. The largest load in the printing plant was 168 psf on 1 percent of its area. The range of loads in ‘one of the department stores is shown in Fig, 1-1. The ordinates of this diagram give the percentage of area occupied by loads varying in increments of 5 psf. The load was 21 psf, and the heaviest 61 psf. In one warehouse, 0.4 percent of 8 area carried a load of 257 psf; in the other, 0.7 percent of the area supported TT sj A Percent of total oreo d it 20 WW 8 ad 45 90 38 GO OD et Toad, FIGURE 1-1TABLE 12 Live loads for various occupancies Range in values requited by building codes Occupancy Live load, pat Residential Libraces(eading rooms) 120-125 250 miimaume 304 psf. The usual range of building-code requirements for the occupancies discussed in this paragraph is given in Table 1-2, which is not intended to be The load surveys mentioned above involve observations of load intensity at an instant of time, ie, at the time of the observation. However, floor loads in @ particular structure vary randomly with time, a the peak lifetime loads may be. Very little informati is ki The tendency of average building live loads to decrease the floor area considered poses a question with respect to the Kinds of supporting members. For example, it would be unrealistic to assume all the floor areas supported by a column to have loads of the same intensity as supported by a j larly, because the maximum expected load is not likely to be realized on all floors simultaneously (except perhaps for some types of warehouse), it would be unrealistic to assume the same uniform load on all the floors supported by a lower column as on the floors supported by a column higher in the building. Building codes make varying provisions for this contin. gency. The Uniform Building Code and the BOCA National Building Code permit 2 reduction in the basic design live load on any member supporting more than 150 f2, except for floors in places of public assembly, in accordance with the formula Re 14 ~ 150) where R = reduction, percent = rate of reduction = 0.08 percent for floors A = area supported by floor or member (contributory area) ‘The reduction cannot exceed 40 percent fot members receiving load ftom one level only, 60 percent for other members, nor R as determined from ; neas(s2) where D = dead load, pst L = basic live load, pst Loans AND stmuctoREs 9 The maximum reduction according to prescribed so that the structure designed for I be safe ifthe total area were t0 bbe subjected to the full live load. Should this happen the maximum load would be 30 percent greater than the load for which the structure was designed. For example, if D/L = the maximum reduction according to Eq, (1-3) is 308 percent, The contributory area corresponding to this reduction can be obtained from Eq.(1-2) as, 30.8 = 0.08(4 ~- 150) which yields 4 = 535 ft®. The structure would be designed for 535D + $35(1 — 0.308) ‘Substituting D = 0,331 yields 535 x 0.33L + 535 — 0.308)L = 548.5. Should the structure receive full live load over the entire area, the total load would be 535 x 0.33L + 53SL = 7133L ‘The maximum load, 713.3L, is 1.3 times the design load $48.5L. For allowable stress design (Art. 1-12), thi specifications is at least 1.6 or 1.67. But for load-factor design and load-and- resistance-factor design (Art. 1-12), the ratio of mai il design load is not constant. In these cases values of dead load and live load and the case illustrated above (D/L = 4, A = 535 follows. The design load is S3S[L.2 x O.33L + 1.6 x (1 — 0.308)L] = 806.41, 2 function of the relative 18 applied to them. For the , R= 308 percent) the result is as If the entire area should experience the full live load, the total load would be 935(0.33L + LOL) = 713.3L ‘The ratio of the design load to maximum possible load is 806.4/713,3 = 1.13, For the extreme cases the results are as follows. If D/L = 0, the ratio of design load to maximum possible load is 1.23. However, for the case where D/L produces the maximum reduction of 40 percent from E ‘maximum possible load is only 1.06. (The maximus responds to a ratio of dead load to live load of 0.73) load-factor design and load-and-resistance-factor desi maximum loads. Consequently, for structures designed according to specifications of this type, the designer might10 bestow oF steat sreucruRes want to consider in some cases somewhat smaller load-reduction factors than hazard which is often posted in commercial should be held responsible for keeping in the specified limit. Failures due to overload are not uncommon. Many such failures are the result of the conversion of buildings oF portions of them to purposes for which they were not designed. 15 LIVE LOADS ON BRIDGE FLOORS Highway traffic is made up of four principal kinds of vehicles: the tru with semitrailer, the truck, the bus, and the passenger car. These units vary widely in weight, and their average weights are considerably less than the weights of the heaviest units. The heaviest truck which is used to any considerable extent at present weighs about 50,000 Ib when fully loaded, and since it is about 40 ft long, it represents an average load of, say, 1250 pif of traffic lane. An average the heaviest car may passenger car weighs perhaps 4000 tb full ‘weigh something less than 10,000 Ib, with correspondi pif. The distance between vehicles in the same traffic ‘obviously important 1m 25 to 50 ft center to by the American Association of State Highway and Transportation Off (AASHTO). These loads consist of a system of concentrated loads to repres truck or of a load distributed uniformly along the traffic lane, together concentrated load, at a long line of medium-weight taffic with a heavy vo systems are necessary because of the be of such length that it can support only one oF les of a truck, and since the effect of one or two concentrated loads on & beam is quite different from the effect of an equal amount of load distributed uniformly, the actual concentrations must be considered. On the other hand, the force in a chord member of a simple truss span will be largest when the full length of the bridge is loaded. A relatively small error results from the substitu- tion of a uniform load for a large number of concentrated loads; hence the lane Joad may be used, with considerable simplification in calculations. Since excep- tionally heavy trucks usually operate at respectable distances from one another, @ Loabs ano stRuctuRES IL L@=— OO 000-1 are 32000-0b axle 32,000-% axle Lg yy v= variable seating, 8'to 30% Use valve mich eroduces mosimum stress © 18.000 10s tor moment |, 000 0s for sacar Uniform tod 540 pt per lone — Concentrated ood w FIGURE 1.2 1820-44 load line of traffic is considered to have only one such vehicle (two in the case of continuovs spans) represented by the concentrated load mentioned above. ‘The heaviest loading of the AASHTO specifications is the HS20-44, pictured in Fig, 1-2. Although it i not an accurate representation of modern locomotives, the system of loads used in the design of most railroad bridges is one whose makeup ‘was proposed a few years prior to 1890 by Cooper. The Cooper load intended to represent two locomotives followed by a uniformly Although from time to time load systems intended to be closer ap ns to the steam locor ical of the era preceding the advent of the diesel have been proposed, they have never been adopted. For comparison with the Cooper 40 8 8.80.80 8252 5252 40 BpA0 GOGO 52 52 5252 ax 40,0090 00.06 83836 66 85 mim POPP P QOD fo, 3 SST Fe a F BR 812. bestow oF sree srauerunes 6 shows the axle loads and spacing for two units of one of the idge specifications of the American Railway Engineering Association recommend that main-line bridges be designed for the Cooper E-80 train Fig. 1-3a. In his original specifications, Cooper recommended three classes of load, which he called A, B, and C. The locomotive driving-axle loads for these thres classes were 24, 22, and 25 kips, respectively, as compared with the 80-kip axles of the E-80 load. 1-6 IMPACT ‘The meaning of the word impact as it is used in structural design may b trated by describing two different ways in which a spring may be loaded. attaches 20-Ib weight to a suspended spring and supports the weight whi being lowered to the px force in the spring is 20 weight is released afte fastened to the spring, is double that required to and the corresponding for ea) t 40 Ib, The 20 Ib of force is customary to express is case, therefore, the impact ing live loads is a much more complex phenomenon than the one just described. The speed of a moving vehicle, its mass relative to the mass of a bridge, and irregula the track or floor and in the wheels of the vehicle are significant factors. Pulsating loads are particularly critical if the frequency of the pulses happens to be coincident, or nearly so, with the period of a fundamental mode of vibration of a structure, Design-specification provisions for impact are frankly empirical and do not attempt to account for al the variables. The AASHTO specifications require impact allowance given by the equation 50 Ty but not to exceed 03. In this formula, I is the ratio of i is the length in feet of the part of the span which cations make similar but more severe provision for impact. Impact allowance for moving loads such as elevators, traveling cranes, and reciprocating machinery is usually specified as a fixed percentage of the load. ‘The effect of impact is taken into accor effects, For example building-floor the values dis- -4 are considered to be sufficiently on the safe side to cover any impact likely to occur, One exception to this statement is an AISC stipulation Loxos avp staocrunes 13 that the live load on hangers supporting floor and balcony construction be inereased by one-third for impact. 1-7 SNOW LOADS Freshly fallen dry snow weighs 5 to 6 pef, packed snow about 10 pet. A number of studies of records of the National Weather Service have been made in an attempt to set reasonable snow loads for various sections of the country. In 1939 the Service published data from 166 weather stations on the great snow in sheltered areas such as clearings in forests. The resul loads ranging from zero in the southeramost pa through about 5 psf for the remainder of the south; 10 psf for the cen states; 15 psf through southern New York, the southern Great Lakes region, and the northern Plains states; and up to 20 to 25 psf for the Lake Superior and New England areas, Exceptions to the: jountainous regions, where of the same where snow rarely falls. ier Service gives ground snow loads having a 50-year mean recurr ig. 1-4). These loads were based on the maximum annual water equi of snow on the ground, Except for the southern states, these loads tend to be higher than those discussed in the preced- ing paragraph, They range from 5 psf in the southern coastal states to 40 to 50 ‘A'map published by the 2 120 8 10 105 100 95 $0 35 69 75 ee FIGURE 1-4 load on the ground, 0-year mean recurtence interval. (ited States Weather Bureau Map January 1969),14 bestow oF sree srrucrunes pof in the northern Gre Data on the Rocky Moun The snow load on a roof is usually less than that on the ground. Based on a ‘comprehensive study of roof loads relative to ground loads, the National Build- ing Code of Canada specifies for flat roofs a basic snow load of 80 percent of the sum of the weights of the 30-year ground snowpack and the maximum 1-day rainfall in late winter or early spring. The weight of snowpack plus rain rune to as much as 120 psf in the Canadian Rocky Mountains? ASCE 7-88 specifies a basic snow load of 80 percent of the 50-year return ground snow, A further reduction (to 60 percent) is allowed in both codes for roofs that have a clear exposure to winds of sufficient intensity to remove snow. Still further reductions are allowed for sloping roofs, since they accumulate less snow than flat roofs. On the other hand, valleys of adjoining gabled roofs of multibay buildings may accu- ulate more than normal amounts of snow. Projections, such as penthouses on fat roofs, may also cause drifting, The Canadian Building Code recommends that aakes region (0 70 to 80 psf in northeastern Maine. but not to exceed 15 ft is of seasonal snowpacks having a mean of 10 years, the loads in Table 1-3 were suggested for the design of roofs.® ld be noted, however, that the mountain areas defined in this table were not included in the study because of extreme local variations in depths of snow. Therefore, the suggested 40 psf for flat roofs will be excessive for many localities in these areas and too low in others. For example, the 10-year- return snowpacks in Reno and Salt Lake City were 20 and 30 psf, respect but these cities are in an ex. e other hand, measurements of the snowpack on the roof of a fountains in Colorado, where the winter- n was 7 ft deep at the valley and 5 ft TABLE 13 ‘Suggested minimum snow or other live load for roofs* On horizomal projection, pat —— Slope of roof Region Sin 2orless Sin 12 9in12 12 in 2 or more Southern states 20 5 2 0 2 5 10 25 0 10 2 2» Loans an structures 15 deep at the caves of the V-shaped roof, disclosed densities of 8, 16, and 25 pef at depths of 3, 5, and 7 ft. Thus, the average density was about 16 pch and the roof load at the 7-f depth was about 100 psf (Ref. 7) The live load suggested hem states (20 psf) exceeds the 10 psf snowpack. This load was chosen because it conformed with the value commonly specified for this region by tended to provide for live loads other than snow, such as those incidental to construction and maintenance. In many cases, of course, the minimum live load prescribed by a governing building code may differ ftom the value in Table 1-3. Snow load need not be considered in the design of bridges, since a fall heavy enough to be of consequence would make the bridge impassable or else compel traffic to move at such a pace as to reduce the dynamic effect, 1-8 WIND LOADS The evaluation of the effects of wind on an object in its path is 2 complex problem in aerodynamics, If we consider air to be nonviscous and incompress- le assumption for velocities of the magnitudes for which equation for streamline flow can be used to determine the local pressure at the stagnation point as a column of air strikes (at 90°) an immovable body. Thus, q= 4p @ where q = pressure p= mass density of air v= velocity of air , OF stagnation pressure, important to note that this equation is based on steady flow and does not account for the dynamic effects of gusts or the dynamic response of the body. ‘The resultant wind pressure on a body depends upon the pattern of flow around it, Pressures vary from poi on the surface, depending upon the local changes in velocity, which depend in turn upon the shape and size of the body. The resultant pressure P is expressed in terms of the drag component Pp ‘and the lift component P;: Py= CoA P=c,a® ) 2 The drag coefficient Cy and the lift coefficient C, depend on the shape of the body and its orientation with respect to the wind. 4 is a character body, usually the projection of the body's surface on a plane. The terms drag and ‘on buildings, bridges, and16 estan oF stest sraucruREs surface, is expressed in terms of a shape factor C, (also called pressure ent): p=Ca=c oe © Air at a temperature of 15°C (59°F) at sea level weighs 0.0765 pet. Substituting the corresponding mass density 0.0765/32.2 into Eq. (@) gives p = 0,001 19C, ‘With the symbol ¥ to denote velocity in miles per hou, this gives p= 0,00256C, ¥? 5) It will be noted that a velocity of 100 mph will induce a pressure of 26 pst when C, = I. Shape factors are discussed later in this article. Measured wind velocities are necessarily averages of the fluctuating veloc- which ate encountered during a finite interval of time. The usual reported value in the United States is the average of the velocities which are recorded during th Kes a horizontal column of air 1 mi long to pass a fixed point (the measuring anemometer). For example, if a 1-mi columa of air is moving at an average velocity of 60 mph, it passes a fixed point in 60 s; the reported veloc- ity is the average of the velocities recorded during these 60 s. The fast the highest velocity in 1 day. The annual extreme mile is the largest of maximums, Furthermore, since the annual extreme ii wind pressures to be used in design should be based. specific mean recurrence interval. Charts of annual-extreme-mi mean recurrence intervals of 2, 10, 25, 50, and 100 yea ‘These are based on statistical record covering a period of 21 years. rval has been suggested fi permanent” structure a 10-year recurrence e smaller risk with short 75 percent of the 50-year velocities are res, such as those used during construction.’ The exceptions subject to the judg- mn) by ASCE 7-88, The velocities shown in ‘open-country” velocities; ie, they ‘obtain where surface friction is rel form for a fetch of about 25 mi. If the exposure is elevated, s ‘0 channeling of the wind, etc, the map values must be adjusted accordingly. Locations with unobstructed exposure to large bodies of water may experience extreme winds 30 mpi (or more) greater than for locations a short distance inland. However, this effect has been taken into erage, the 100-year velocity in the United States is about 8 percent higher then the so-sear velocity, Loans aND svaverunss 17 7s 7 ar FIGURE 15 Annual exireme festestumile wind speed 30 above ground, $0-year mean recurtence intersl, (From Ref 8) The National Building Code of Canada recommends pressures based on y wind speed (average of the velocities measured during 1 bh). thc only speed measured at most weather stations in Canada. on the ground exponential variation ‘exponents have been suggested. Hi 4}-power law of Fig. 1-6 is generally accepted for flat open country. Velocity at 30 & above ground is used as the basic vak Purposes, and increases with height are provided for by speci ressures) for various height zones. Table suggested velocities for various height zones for Intermediate values can be i veral basic wind vel ereases for inland areas are fe average valu necessary to con (gusts). The response of a structure to such fluctuations is a dynamic one and depends on the size of the structure, its natural period of vibration, and its damping characteristics. The dynamic effect is usually accounted for by multiplying the wind velocity by a gust factor and com- puting the corresponding pressure by Eq. (1-5), which is to say that the response18 oEsIGN oF sreeL stmucTURES 1209, 1600 va00F- 8 1000}- t ‘Flot, open country, ‘Cente of lrge oped ior coosea Beles ep FIGURE 1-6 ic one. The gust factor depends on the wind vel size of the structure, because the wind pressures are not fully developed until the loped in the moving mass of air. For this reason, a massive ly insensitive to gusts of short duration, wh I account for a 1-5 gust in a 90 mph basi ind fength of 130 ft and would be adequate for signs and small structures.'' A gust factor of 1.1 will account for a 10-s gust in a 90 mph factor of 1.3 ile of wind for various zones above ground* Design velocity, Design velocity, mosben ae finland aces, mph Zone height Sor olhrsource Loans ano staucrunes 19) basic wind, which would have a downwind length of about 1300 ft. This gust factor has been suggested for structures on the order of 125 ft wide transverse to the wind.'* A method for evaluating the gust factor as a function of wind velocity and the characteristics of the structure has been developed. that the gust factors discussed in this paragraph do n effects such as flutter or vortex shedding. These effects ar Design wind pressures can b Eq, (1-5). The shape factor C, varies c the proportions of the structure and the horizontal angle of incidence of the wind. The shape factor for the windward face of a flat-roofed rectangular building is about 0.9, regardless of the proportions of the building. There is negative pressure (suction) on the rear face, for which the shape f 6, depending on the proportions of the building. Thus, the resultant pressure on such a b can be determined by using a shape factor ranging from 1.2 to LS in Eq, ‘The value 1.3 is commonly pressure and suction when de wind bracing of buildings. Sidewalls experience suction, for which C, ranges from about ~0.4 to —0.8. The roof also experiences suction, for which C, ranges from about 0.5 to 08 for the average over the roof. However, the suction is larger on the windward side, and the average coefficient for the windward half may be as much as twice that on the leeward half The preceding discussion applies to an airtight bh through small openings around doors, windows, etc, gives nal pres- as large as 0.25 if the openings are chiefly on the windward face, fernal suctions, with C, as large as ~0.35 if the openings are predomi. leeward. The following internal wind pressures were suggested in Ref, 14 ig8 which are nominally airtight, a pressure or suction of 4.5 psf normal to the walls and roof 2. For buildings which have 30 percent or more of the wall surfaces open or subject to being opened or broken open, a pressure of 12 psf or a suction of 9 pst 3. For buildings which have wall openings between 0 and 30 percent of the wall area, pressures or suctions varying linearly between the values recommended in Land 2 The pressures in items 1 to 3 above are for a velocity which produces a resultant Pressure (sum of windward-wall pressure and leeward-wall suction) of 20 psf, would be consistent to increase (or decrease) them prop larger (or smaller) basic pressures. When openings are large, as in hangars, internal pressures may be quite large. Wind pressures on sloping roofs depend on the exposu Proportions of the building. For wind normal to a side parallel to the ridge, the leeward roof surface is always subjected to suction. There is suction on the wind- ward surface for slopes less than about 30° and pressure for larger slopes. These20. oesiox of srest staUCTURES pressures are not uniform but have maximum values at the eaves, The following pressures for single-ridged roofs are given in Ref. 9 (074 Osa” a — 2.1) P4530" For the windward surface =a oo eee (0.03 -09jg 30° 3. REFERENCES 1 Thom, H. C.S.: Distibutions of Extreme Winds in the Uni 1560. 2 Dunbar, J. W.: Design Live Loads in Bi 3. arysoo, J-O, and D, Grose: Techniques States, J. Strut. Dis, ASCE, April 1947 of Live Fleor Loads and Ser. 16, 1967. Thompson: Live Loads on Floors in Buildings, Nat. December 1982. Res, Coune. Oxtawa, Suppl. to Net ing Code of Canada, Load Studies, Housing Home Finance Agency Washington, Housing Res. Pop. 19, May 1952 7. Andersen, A, E. Saow Loads for Rocfe, Cin. Eng, July 1968, ‘Thom, H.C: $. New Distr 9. Wind Force on to on Loads snd Stresses, St 10. Davenport, Res, Tech Pap 8, 1960. 4H, Sherlock, R. H.: Gust Factors forthe Design of Buildings, Int. Assoc, Bridge Struct. Eng, Zurich, vol 8 1987 2p. 12 Freudenthal, A. M, J. M. Gamels, and M. Shinozuka: The Anais Siruct. Div. ASCE, Febeuary 1966, 20. Ang. A. 1d M. Amin: Reliability of Structures end Structural Systems, J. Eng. Mech, Die. ASCE, April 1968. sal of Safety Criteria and Their Basic Concepts, Prelim. Publy 8th Congr. 68 lambos: Load and Resistance Factor Design for Stel, J. Struct. I: Development 23, Elingwood, B, T. V. Galambos, J. G. MacGregor, and C. A. Cor ty Based Load Criterion for American 577, June 1980. 24, Galambos, T. V. CHAPTER 2 STRUCTURES, METALS, AND FASTENERS The supporting systems used in buildings, bridges, transmission towers, radio towers, observation towers, airplanes, ships, storage vessels, and various other structures may be classified broadly into three categories: (1) framed systems, (2) Suspension systems, and (3) shell systems. Examples of these systems are dis- cussed in following articles. In many cases, structures are supported by com- tions of supporting systems. 2-1 BUILDINGS In addition to their function of partitioning space, w: tion furnish reactions for the beams and joists which s in_bearing-wall construc- mn. Frames may be classified as rigid, and simple, depending upon the stiffness of the beam-to-column con. Rections. In the AISC/ASD specifi ese framing systems are designated Type 1, Type 3, and Type 2, respec ion, rigid frames are designated fully restrained (FR), while semirigid and simple frames are combined in a single system designated partially restrained (PR), Framed build. ings are also called tier buildingsSrRUCTURES METALS AND FasTeNERS 35) 34 esiox oF sree srmucrunes FIGURE 21 Simple frame withFIGURE 23 FIGURE 24 Athkaic and Convocation Center, University of Notre Dame. (American Institute of Steel Construction)38. desiow oF stent sreucruREs Compression ring Tension ring Upper cable y ZA ZE ay PTs >, ” Lower cable: Separotor 240" $$. FIGURE 25 Utica Municipal Auditorium. Gehron & Seltzer, Architects. Lev Zein Associates, Ine, Structural Engineers, New York. cables. These roofs may develop oscillatory m usually called flutter, from the effect of dynamic forces. Flutter may be controlled by tying the roof to the id with cables or by using a system of interconnected cables which are inter- selfdamping. The ipal Auditorium in Utica, New York (Fig. 2-5), is ly selfdamping system. Suspension roofs have been of over 350 ft. Cables may be also used to support cantilever may be built with standard cold-formed steel roof-deck panels. The ue and the hyperbolic paraboloid are suitable for this type of structure. In ‘d-plate roof, flat or inclined light-gage steel panels span between ridge and valley members. The panels and ridge and valley members act together as FIGURE 246 Framing system for TWA hangar, Philadelphia International Airport. STRUCTURES, METALS.AND FaSTewERS 39 FIGURE 27 Fold ‘oot, Marple Township Library, Penasylv beams to support snow or other live loa: doubly curved shells made up of two fami of one set of generators, which requires a slight warping of the individual panel (Fig, 2-8). 2.2 FIXED BRIDGES Most bridges are built for the transportation of highway or railway traffic across natural or artifical obstacles. A deck bridge supports the roadway on its top chords or flanges, while a through bridge supports the floor system at or near the lower chotds ot flanges so that traffic passes through the supporting structure. ‘The rolled-beam bridge supports its roadway direcily on the top flanges of 4 series of rolled beams placed parallel to the direction of traffic and extending from abutment to abutment. It is simple and economical. It may also be used for ‘multiple spans where piers or intermediate bents can be b bridges may be economical for spans up to about 60 ft. A highway trafic is illustrated in Fig, 2.9, For crossings greater than those which can be spanned economi rolled-beam bridge, deck or through plate-gitder bridges may be us40 estan oF sreen staucrunes Riage member ae FIGURE 28 ‘American Aitines supecbay hangar, San Francisco, Lev Zein Associates, Ine, Consulting Enginsers, New York | | | | sreucrorss METALS. AND FasreneRs 4 520m Beorng pore or pad ee “pier or covtment FIGURE 29 Beam bridge. simplest form, a plate girder consists of three plates welded together to form an I Ties and rails for railway bridges rest directly on the top flanges of the deck plate-girder bridge. When clearance below the structure is limited, a through girder span is used. The floor system may consist of a single line of stringers under each rail, supported by floor beams framing into the girders just above lower flanges. If an open floor is objectionable, ballast may be laid on con- or steel-plate decking supported by closely spaced floor beams without 1e¢ braces are used to support the top flanges of through bridges, as, illustrated in Fig. 2-10. Highway plate-girder bridges are usually of the deck type. The floor slab is usually supported directly on the gitders, as in the beam bridge of Fig. 29. Plate girders may also take the form of a box, as shown in Fig. 2-11. Cross sections a and b consist of steel plates for the entire cross section, whereas in c the top flange consists of a reinforced concrete section attached to the steel flange plates to act as a composite section, Such cross sections have substantially greater torsional resistance than the conventional { cross section. Their fabri tion requires careful control, and erection procedures must receive special con- sideration. The latter is especially important for cross section c, which may be very unstable until the concrete deck is in place. Girders of type a and 5 have Anco oroce Seon aa FIGURE 24142. bestow oF sree sraucruRes FIGURE 211 Box gitder cross sections -mployed for continuous girders of highway bridges with spans in excess of t. Type c has been widely used for bridges crossing interstate highways. rections by a system of longitudi- (Fig, 2-12). The deck structure 1s and floor beams. This system makes efficient and economical use of materials, particularly for long-span construction, When the crossing is too long to be spanned economically by plate girders, fa through or deck truss bridge may be used. Deck bridges are more economical than through bridges because the trusses can be placed closer together so that span of the floor beam is shortened. For multiple spans there is also a sa the height of the piers Figure 2-13 shows the common types of simple-span bridge trusses. By varying the depth of a truss throughout its length 30), forces in the chord members can be more nearly equalized and the forces in the web reduced. ig in Wsoring surface FIGURE 2:12 Elements of deck bridges, STRUCTURES. METALS AND FasTENERS 43 Cy ‘Trusses of economical proportions usually result if the angle between diagonals and verticals ranges from 45 to 60°. However, if lng-span trusses are made deep enough for adequate rigidity as well as for economy, a suitable slope of the diago- nals may produce panels too long for an economical floor system. The sub- divided panels of the Baltimore and Petit trusses (Fig. 2-13f and g) solve this, problem. Certain objections to subdivided panels were overcome with the inven- tion of the K truss (Fig, 2-13h). Cantilever bridges, continuous bridges (Fig. 2-14), arch bridges (Fig. 2-15), and suspension bridges are commion types of structures suitable for long spans. A cantilever bridge consists of two shore, or anchor, spans flanked by cantilever arms supporting a suspended simple span, Positive bending moments are decreased because of the shorter simple span, while the cantilever and anchor arms are subjected to negative moments. Positive bending moments in co! uous bridges are reduced because of the negative moments at the piers bridges may be fixed, single-hinged, two-hinged, or three-hinged. The principal supporting elements of the suspension-bridge superstructure are the cables which ass over the towers to be anchored in foundations at each end. Since 1950 a new form for bridge structures called the cable-stayed bridge has been developed (Fig, 2-16). Such structures take advantage of the tensil strength of wire cabl ages which accrue when the roadway le supports with corresponding nin the moments. The towers at the interior supports of forms, and the cables may assume a fan configuration as in configuration as in bigure 244 Rio Grande Gorge Bridge, Taos County, New Mle, New Mexico State Highway Commision (American Insite of Steel Construction) PIGURE 245 Satsop River Bridges, Satsop, Wash, Washington Stale Highway Commission. (American Institute of Stee! Construction) sp46 veston oF sree staucruREs IK : SX JOS ISAZi&y { i FIGURE 2.16 ‘Typlaleable-stayed bridge forms, 2.3 TOWERS Towers are used to support transmissior fon. It can be extruded to produce 8 joints. Because of their lighter ‘optimum shapes for resisting stress and simp! i ely inaccessible } aluminum towers can be flown by helicopter to rela 24 MECHANICAL PROPERTIES OF STRUCTURAL METALS ‘The design of structures which support calculated loads at specified stress: requirements. The most widely ructural materials ate those of the American Society for ‘Testing and M: (ASTM), ASTM specifications for structural steels generally identify the process by STRUCTURES, METALS, AND FasteNeRs 47 Speci Shapes, Sheet ines the procedures strueture, surface and edge condition, repair procedures, testing requirements, dimensional tolerances, and other criteria essential to quality control. Additional may be specified by the purchaser if they are necessary. A370 con- 'd provedures for the mechanical tests used to determine the proper- ties required in the product specification. The yield point, ultimate strength, and elongation specified for structural materials are determined according to the procedures of ASTM A370, Standard specimens, generally called coupons, are used in the tensile test to establish the Properties of the material. Flat specimens of the dimensions shown in Fig, 2-17a ‘or round specimens of the dimensions shown in Fig. 2-17 are the standard speci- ‘mens used for this purpose. The dimensions of the specimens are established to ensure that feifure occurs within the designated gage length. Substandard speci- mens are permitted in special cases, and the results of tests on such specimens hhave been correlated with the corresponding results for the standard coupon, On. different structural sections the location from which the sample or coupon is to be taken is also specified. If the design situation warrants it, the engineer may specify tests in addition to those required by A6. ieee moc pee a FIGURE 2.17 Standard cou G~ gage length; W = width; T= thickness; ngth; A = length of reduced section; B= length of snp Round specimens: G = gage length; D = diameter; R= radius48° vestow oF sree: sraucrunss Sin arrengen Stress, isi 0 Oo08 6005 G00 aos aued ues FIGURE 2418 1 coupons tested in Two typical stress-strain curves for structural tension are shown in Fig. 2-184. The tensile stre is ynal area. ength is reached, is cal by an abrupt decrease in Fig. 2-18, which results in upper and lower yield points. The upper yield point is influenced considerably by the shape of the test specimen and STRUCTURES, METALS. AND FASTENERS 49) by the testing machine itself and is sometimes completely suppressed. The lower yield point is much less sensitive and Stress-strain curves of this type ate t upper and lower t usually have a pronounced behavior which ends 8 to drop and continues to fi ve of Fig. 2-18q). In this case, 2-18). However, yield strength may also be ‘STM Specification A370) as the stress corresponding to a 0.5 percent under load (Fig. 2-185). id stress is commonly used to mean either yield point or yield itis not necessary to make the distinction, in compression has the same modulus of elasticity as in tension. The Tepresents is not a continuous process in the sense that the elongati distributed uniformly over the length of the specimen. Instead, yielding is a di continuous phenomenon, In a tension test, it generally be; ines, flow lines, or yield about 45° to the direction of the tension. They inclined planes, which are planes of maximum shearing these planes of sliding increase sudden! order of 0.02 to 0.04, Flow lines are easily detected on a specimen whose surface hed. They are also revealed by the flaking of steel. They can be detected ipbands are plastic regions which are separated from one another by ly elastic regions. Thus, the elongation which is measured over a region sequence of alternating layers of elastic bands spread and increase in thickness upper curve of Fig, 2-18a. The elongations of the alumin- used for structural purposes are roughly only half that of the structural steels (compare Tables 2-2 and 2-5). The modulus of elasticity isSTEEL STRUCTURES ,000 ksi as compared to 29,000 ksi for steel. Thus, the elastic deforma- tion of an aluminum structure steel structure tthe proportional limit is gener~ low. For some stainless steels the stress-strain curve may be different in tension and compression and the results may also be different in the longitudinal and transverse direction depending upon the method by which they are produe- ed. Consequently the application of these materials in compression requires knowledge of the stressstrain curve, which consists in bending a specimen 180° around a pin. The di in is related to the thickness of the specimen and also varies with the steel. Ductiity is an extremely important property, but there is no generally accepted minimum which is required of steels for structures Other important properties of structural me tance to fracture, and toughness. Fatigue in cracks which may Brittle fractures are usually propagate rapidly at relativel strosses equal to of less than the yield strength and propagate at stresses as low as 20 percent of the yield strength (Act. 2-16), Toughness denotes the capacity of a material to resist fracture under impact loading, The area under the stress-strain curve is a measure of toughness, so both strength and ductility contribute to toughness. Thus, because its modulus of el is only one-third that of steel, an aluminum structural member can absorb three times as much energy, for the same stress, as a steel member of the same dimensions, pr t exceed the proportion: Strain rate and strain aging are sometimes important in assessing the suit- ability of metals for structural purposes. Both yield stress and tensile strength increase with increase in strain rate, These increases can be substantial under tension impact, Strain aging has to do with a change in properties during a rest period following unloading of a specimen which has been stressed into the inelas- tic range. Such an unloading path is parallel to the elastic loading path of the material (AB of Fig. 2- i unloading path to the e specimen is allowed to “age” at room temperature for a few days alter it is unloaded, it may reload along BADE. Aging for 1 or 2 i the yield strength, and the ultimate strength." It will be noted that these increases are accompanied by a loss in ductility (BF compared with BG in Fig, 2-19) STRUCTURES, METALS, AND FasTeseRs SL Steoia FIGURE 2-19 2-5 STRUCTURAL METALS A wide variety of steels is produced for the constr towers, tanks, and other structures basic-oxygen, ot electric furnace process. They may be rimmed, sem led. Rimmed steels are only slightly deoxidized and are characterized by a rim of ‘bon steel which freezes upon initial contact of the molten metal with the ingot mold, a strong ev and a higher-carbon core, In se mn of bridges, buildings, ing agents such as steels are those in which the carbon-oxygen reaction is completely stopped by the a ‘ . ly deoxi- dized by the addition of either silicon or alu od steel provides the Most homogeneous structure in the finished steel. Structural stecls used regularly in bridges and buildings are described in Table 2-1. Tensile properties ate given in Tables 2-2 to 2-4, Producers have given many of these steels brand names, some of which are not covered by ASTM specifications. Only ASTM steels are discussed here, Steels for sntain carbon, which is the most important element except for the fer but carbon steels are generally understood to mean steels whose properties are controlled largely by controlling the carbon content. Most of the high-strength steels are really of intermediate strength com- pared to the quenched and tempered steels. It will be noted that the yield stresses in Table 2-2 range from 36 to 42 ksi for carbon steel, 42 to 65 ksi for the high- strength steels, and 90 to 100 ksi for quenched and tempered steel52. estan oF stent smucrunss TABLE 21 Steel for structural purposes ASTM designation 36 ASS Asis Asoo Aste as Ast Product Carbon-steelchapes, plates, and bare High-strengih low-alloy shapes, plates, and bars shapes, plates, and bars Zinocoated (galvanized) shoes in coils or cut lengths (Coldsformed welded or seamless tubing in round, square, rectanguler, ot special shapes Hootformed welded oF seamless tubing in round, square, rectangular or special shapes ‘Quenched and tempered plates of high yield strengts Carbon-steet pistes end bars Yin thick Hot-rolled carbon-stel sheets and strip in coils or cut legs High-strength lowalloy columbium-vanadivm soot Use Welded, riveted, and bolted construction; bridges, buildings, towers, and general steel; a weathering seo! Welded, sivetd, or bolted other structures; atmospherie-corresion resistance double that of carbon stel CCold-formed structural members for buildings, especially standardized Welded, riveted, or bolted co 5 bridges, buildings, and general structural purposes of building in all grades; welded ‘ridges in grades 42, 45, and $0 only STRUCTURES. METALS AND FasTeNeRs 53. ‘TABLE 2-1—{eontinucd) ASTM Assigeation Use Ass Intended primarily for welded bridges and resistance about four times that of carbon steel; weathering steel A608: Intended for structural and miscellaneous purposes where savings in Weight oF Ast 00s A607 ‘cold or hot-rlled sheets and strip ‘Anos Structural ste! for bridges Structural steel is produced in the form of shapes and flat-rolled products. Flat-rolled steel is called bar, plate, sheet, or strip, depending on its width and thickness, and may be either cold-rolled or hot-rolled. Specified tensile properties of the steels furnished in shapes, plates, and bars are given ia Table 2-2; proper- ties of structural pipe and tubing are given in Table 2-3 and of sheet and strip steels in Table 2-4. All the steels in these three tables are listed in Table 2-1. It will be noted that both yield stress and tensile strength are specified as “minimum values.” Since the specification requires that test specimens conform to these values, rolling mills control the manufacture of the product so that the average yield stress and average tensile strength are well above the specified mini- ‘murs. Figure 2-20 shows the yield-stress distribution of 3974 mill-test specimens taken from 33,000 tons of A7 steel* on nine projects erected between 1938 and * A carbon tcl witha minimum yield stress of 33 kal which was wed extensh 1965. was di in 1967 fom 1936 to about54 peston oF sreet structures STRUCTURES METALS, AND FASTENERS $5 i TABLE 22 Minimum tensile properties of structural stels i asm™ Yo, Seog, HER ‘Minimam tensile properties of sheet and strip steels ' designation ksi unless note) Carbon steels: aera Nils i 436 3% eso 20 designation si A599 2 ws Carbon sets Highacength set 500 A, Aas Grade 30 0 6 » » mn 8 Grade 33 B 2 B 6 7 » Grade 36 ® 3 0 2 6 6 Grade <0 0 3 16 Grade 45 rr) 2 © » Grade 50 0 6 2 o 6 1 att ® % % Grade A 3 8 26 6 om 1s x 8 2 B® 2 To 4in thick nn) 2 a Over dint Sia 6% 6 o 8 + Over Sint Bin 2 6 Qucnched and empered sels 3s 6 roar Ait "To 2f in thick wo oD a8 7 ira 9 nas 0 6 nom 3 ie o 7% ies 8 # ieee m8 ee imum tensile properties of structural pipe and tubing Grade as ss nas Elongation Gra $0 0 © a asT™ Viel, Seng, % in 8in desination a) aes ote Weide or seamless pipe: 35 Grade A wa . Grade 3 @ . Welded or sates tubing ‘30 (eoi-ormed) Round, Grade A Boss 2 Rovnd, Grae B 2 8 Bt Round, Grade C «2 21 ar 25t ar) ar 6 @ a 6S 2 * Varin se speciation odie56. vestan oF steet sravcruss Pecces frger or sme 3) FIGURE 2.29 ‘Yieidstessfequency distribution 1951 (Ref. 2). The figure shows that about 5 percent of the specimens tested less than 35 ksi, with possibly 2 percent below the specified minimum of 33 ksi. More Extensive data on yield stress an¢ of elasticity for different struc- tural materials have been” analyzed st by Galambos and Ravindra? They report mean values of the ratio of mill yield stress to specified yield value from US. mills of approximately 1.20 coefficient of variation of 0.10. Since mill tests are taken from the webs of rolled shapes and the yield stress of the flanges is usually smaller than that of the web, and because are per formed at loading rates higher than occur in most structures, the values are adjusted to reflect these facts. The adjusted ratio of mean yield stress to specified yield stress varies from 1.04 to 1.11. ‘The most important factor affecting mechanical properties of the steels is the chemical composition, Some of the other factors are the total reduction from ingot to finished product, finishing temperatures, and rate of cooling. Because thin plates involve a larger reduction in the ingot than thick plates, which requires more passes through the rolls, they have higher yield stresses. However, Id stress may be kept independent of thickness by varying the chemical com- rolled, This is done with A36 steel for ¢ chemical composition of A242 and is smaller for thicker is lost through erosion by wind and rain in the process. The fil ng in texture. It does not crack or flake, and it retards further corrosion so that painting is unnecessary except where the steel may be exposed industrial fumes or to salt spray and fog. The protective coating 36 months) in an industrial atmosphere. These steels are called weathering steels. They have been used in bridges and in : ; i 4 STRUCTURES METALS. AND FasTENEES 57 buildings with exposed frames. A242 and A588 are weathering steels. They have about four times the atmospheric-corrosion resistance of A36 steel ‘ASTM Specification A36 covers weldable structural carbon steel with a minimum yield point of 36 ksi. It is available in all the standard rolled shapes thick. It is also furnished in plates over 8 in thick, up to 15 1d point of 32 ksi. Copper can be specified as an alloying which doubles the resistance to atmosphs jctural carbon steel. It has a minimum in plates and bars to $ in thick and in the lighter standard shapes. It has the same atmospheric-corrosion resistance as copper-bearing A36 steel It is used principally in standardized steel bui A242, A572, and A588 are called high-strengt feels. A242 and A441 have suitable for riveted, in Tt is not as economical as A441 and is not used Vanadium and silicon are the principal alloying ele- is produced for welded construc Grades 60 and 65 are also of buildings and other structu: bridges. A588 in thicknesses to 4 stresses (Table 2-2) It is suitable for welded, riveted, and bolted construction but is intended primarily for welded bridges and buildings. There are five grades, all with the same yield stress. Each grade is a proprietary steel, and each is produced by a different company. The A709 specification for structural steel for bridges applies to three different yield points and also provides for enhanced corrosion resistance. The strength of this material is comparable to that of A36, A572, AS88, and ASI4, The most significant difference is that this specification requires testing for impact resistance, These requirements vary for fracture-critical and non-fracture-critical members. This concept is discussed in Art, 2-16 The required toughness of these materials is related to the ambient temperature to which the structure will be exposed. Quenched and tempered alloy steei plate is covered by A514. This stcel is furnished in 16 types, each under a proprietary brand name. It has the highest yield stress of ictural steels. It has outstanding toughness and is intended Primarily for use Although all the this does not at identi the ASTM specifi- cation usually cal be suitable for the massed in steel and the intended service. Weldability of many of these steels is, Ref. 4. pipe and tubing is often used in building construction (Table 2-3). AS3 jed and seamless pipe. However, only Grade B of this specification,58 estan oF sreek staucrures TABLE 25 ‘imum tensile properties of aluminum Yield, Streogth, Elongation, ki ksi 2in the same tensile properties as A36 steel (Table 2-2), This pipe is furnished dard sizes ranging in outside diameter from 4 to 24 in. Carbon-steel struc- Square, rectangular, or special shapes may be cold-formed, welded, or seamless (AS00) or hot-formed, welded, or seamless (A501), Properties of these steels are given in Table 2-3, Common sizes of the square tube range from 3 x 3 in to 12 x 12 in, The rectangular tube ranges from 3 x The sheet and strip steels (Table 2-4) are used for cold-form tmembers and find wide app in standardized steel bu range from 30 to 80 ksi. A intended for service requiring greater strength and atmospheric-corrosion resistance equal to or greater than that of plain copper-bearing steel. A446 is produced with eight classes of hot-dip zinc coatings so that sheets with coating consistent with forming hazards and expected service life wil be available. Tae aluminum alloys usually used for structural work are known com mercially as 2014-T6, 6061-T6, 6062-T6, 6063-T5, and 6063-T6 (Table 2-5). The number in these designa lentifies the con tages as ance. 2.6 STRUCTURAL SHAPES A wide variaty of structural-steel shapes is manufactured. Round and square bars are extruded. Flat steel is from classified accordi bearing plates. The common rolled the tube, the channel, and the I. The I is available in two ‘The most widely used is the W shape (formerly WF, meaning The other, once called the American Standatd Beam, is called the STRUCTURES. MEFALS,AND Pasreneas 89 S shape. Miscellaneous column and beam shapes used for lightweight construe- tion are rolled by a few mills. Wide-flange shapes can be produced by passing an assembly of two flange plates and a web plate through submerged-are welding machines which simultaneously weld both flanges to one side of the web. The section is turned over to weld the flanges to the other side. The structural tee is obtained by splitting the web of an I, generally by the use of rotary shears. There is a more or less constant demand for all these shapes, and therefore they are readily procurable. A group of special shapes such as subway columns, special channels, bearing piles, tees, and zecs are rolled only by arrangement with the and should not be used unless the quantity needed is sufficient to warrant a rolling. All these shapes are manufactured to certain tolerances with respect to dimensional variations such as camber, cross section, diameter, squareness, flat ness, length, straightness, sweep, thickness, weight, and width. The specific limi tations are contained in ASTM A6. Rolled section properties which are to be used in structural design caleu- lations are presented in part 1 of the AISC 58 Structural shapes arc ified by a letter designator which indicates the particular cross section, ‘Typical indicators are: W Wide flange beam M Miscellaneous beam s American standard beam C American standard channel MC Miscellaneous channel L Angle WI Structural tee cut from W MT ictural tee cut from M ST _Steuctural tee cut from $ ter designators are followed by numbers which identify the particular n, for example, WI8 x 50, C12 x 30, MC8 « 85, WTI8 x 80. The first numeral in pounds per foot. In the case of W, M, and 8 shapes and tees cut from them the depth designation is the nominal depth, whereas for the C and MC shap icated depth is the exact depth of the section. Angles are designated by the leg sizes and thickness, L6 x 6 x 4 or LT x 5 x $, with the long leg designated first and the ness last. Part of the Manuals also tabulates the properties of many com- ‘monly used b Cross-sectional properties utilized in design calculations are listed in the tables of the Manuals, Particular note should be taken of the fact that for the flange width, flange thickness, and web thickness two values are given, one decimals and the other in fractions. The fractional value is used in detailing. All design calculations employ the decimal values. Aluminum structural shapes are produced by rolling or by extrusion. Stan- dard angles, channels, I's, tees, and zees are available in a wide range of sizes.terlocking shapes, sections with integral backup strips and bevels for welding, structural shapes iffening lips or bulbs to support outstanding anges, and stiffened sheet panels are a few of the sections readily extruded. ‘Cold-formed steel shapes are formed in rolls or brakes from sheet or strip steel, Because of the great variety which can be produced, shapes of this type, unlike hot-rolled shapes, have not been standardized. Although a number of f ricators have developed their he designer may devise special shapes for particular jobs. While shapes up to thicknesses of 4 and even 3 in can be formed, cold-formed st angles in the thinner gages ips or other edge stiffener are made by spot-welding two or more shapes, e.g, two channels or and two angles. Wire rope is made of a m which may be a fiber rope, strand, of a small wire rope. A strand is made of a number of wires around a center wire, Strand itself may ividual load-carrying tension member where flexibility or bending is not a major requirement, Wire rope provides increased flexibili structural purposes it usually consists of six strands plus the central core. Bridge strand is approximately four ti only abo lly around a core, 2-7 RIVETS ‘The components which make up the completed metal member or structure are fastened together by means of rivets, bolts, or welds. Rivets are rarely used in the types of steel is book. They are made from rivet bar stock in a machine which forms one head and shears the rivet to the desired length. Rivet heads are usually of a rounded shape called a buttonhead. The head may be fla imited, When very litle clearance is avail- sunk heads are chipped flush if no formation concerning conventional signs for riveting, Tengihs, and other data are given in the AISC Steel Steel rivets are almost always heated before driving. In the shop the rivet is heated to a miniroum temperature identified by a light cherry-red color. Most shop rivets are driven by pressure-type riveters which complete the riveting oper- ation in one stroke. Riveting guns are portable hand tools, operated by com ¢ rivet by a rapid succession of blows Rivets are made from steel conforming to the specifications for rivet steel, ASTM A502, This specification covers two grades of steel. Grade 1 is a carbon for 1g high- ngth manufac- STRUCTURES, METALS. AND FASTENERS 61 6062-6 are made from alloy 6061-T6 and for structures of 6063-TS and 6063-T6 from alloy 6053-T61, Cold-driven rivets for structures of alloy 2014-T6 are made from alloy 2117-T3. Hot-driven rivets for the five alloys are made from alloy 6061-743. 28 STRUCTURAL BOLTS bolts for steel structures are the unfinished bolt (A307) and the high: (A325, A449, and 4490) The A307 bolt is known by a variety of names—unfnished, rough, common, ordinary, and machine. It is furnished in two grades, A and B, the former for general purposes and the latter for joints in pipe systems. They are made of low-carbon steel with a minimum tensile strength of 60 ksi. They are tightened by using long-handled manual wrenches, so the induced tension is rela- tively small and unpredi types of locknuts can be used to prevent loosening where shock and vibration are consideration. ‘The A325 bolt (Fig. 2-214) is made of medium-carbon steel, It is also used in both hot-rolled and cold-formed steel construction, The tensile strength of this62 ves TABLE 26 Properties of structural bolts ‘Tensile strength ks?” Minimum yield Teng asm - ee Seiten Maloun Maxam = Low-carbon steel: Hl ‘sar GrdeA al @ Gnéeh At © Uihseenst rec bl Maduecatbon te “ats ane Actos 19 Auasand Ae tes a " aio 0 8 3 Alyse eo Haw bo eae Eee eee add “ont wn(0 2 bolt decreases with increase in diameter ofthe bolt ne it, s0 two ranges of diameter are specified (Table 2-6). The A449 bot, also of mediuimcarbon ste is furnished | three ranges of diameter. The A490 b strength grade. It should be noted that the tensile pro “stress area.” This is larger thaa t th 3 thread but smal than the unthreaded area. The tensile strength of the bolt, based on this area, is about the same as the coupon strength, : High-strength bolts can be tightened to large tensions, whi f 0 large tensions, which produce high clamping forces between the coanccted parts. The Research Council on Structr- al Connections (RCSC) Specifications for Structural Joints Using ASTM A325 or ‘A490 Bolts” preseribes four tightening procedutes for control ofthe pretension 1, Turn-of-the-nut method 2. Calibrated-wrench tightening 3. Installation of alternate design bolts 4 ‘ension-indicator tening inary spud wrenches or standard-power impact wrenches can be the turn-of-the-nut method. Bolt elongation, and thus the ‘casion, w controlled is method, After the bolts in a connection have been tightened to the snug- jkt condition, they are given an additional one-third turn if the bolt length is less than four diameters, one-half turn if the bolt length is greater than four STRUCTURES, METALS, AND FASTENERS 63 diameters but less than eight diameters, and two-thirds tur if the bolt length is treater than eight diameters but less than twelve diameters, For lengths greater than twelve diameters special calibration is required. The snug-tight condition is in a joint are in firm contact sponds approximately to the One full turn from a finge one-half turn from the snug ‘Manual torque wrenches or adjustable-power impact wrenches are used in ‘the calibrated-wrench-method of tightening, Wrenches are calibrated by tight- ening, in a hydraulic tension-measuring device, a minimum of three bolts of the same diameter. Impact wrenches are set to stall when the prescribed bolt tension js reached. Manual torque wrenches have a torque-indicating device so that the torque required to produce the initial tension is measured, The calibration should bbe checked at least once a day or when the wrench is to be used on a bolt of difforent size. A hardened washer must be used under the nut or the head, which- ever is turned in tightening, This requirement was added to the specification in 1985 to ensure the development of consistent and reliable pretensioning of the bolt. Without the hardened washer the friction between the nut or head and the connected material is ikely to vary considerably from one connection to another, of even from one bolt to another. ‘Alternate design bolts are fasteners which incorporate design features tended to tension or to automatically provide the required tension. Some are provided with a special appendage beyond the normal thread of the bolt which shears off when the cortect bolt tens obtained. The required by the specifica! ‘Not all the specifications (AISC, AASHTO, and AREA) permit all the tight- ‘ening procedures defined by the RCSC. Each of the specifications has different requirements for the use of washers, which are dependent upon the yield point of the connected material and the tightening procedure employed. The AISC has special requirements in both ASD and LRFD when oversize or slotted holes are ‘ions among the specifications. Other fasteners are used occ y. The high-strength, interference-body bearing bolt (Fig. 2-21) has relatively hard, rolled, serrated ribs which produce a solid bearing fo ickness of the connected parts. This bolt gives the strength and clamping force of the A325 bolt and, because it fills the hole, pre- slip should shear loads exceed the frictional resistance of the joint. Another of interference fastener is the ribbed is made of carbon and has strength equal to or grea the same diameter. It has a standard rivet head, a fluted sha shaped ribs, and a self-locking nut. These fasteners are especially useful where it is impractical to use power tools, ‘A394 galvanized steel bolts and nuts are the common fasteners for towers and similar structures. These bolts are available with hexagonal or square heads64 oestox oF sres. stmucrunes ‘and nuts. A nonloosenin, idea '# connection is provided by lock washers, jam nuts, of im bolts and nuts are fasteners can be used. and aluminum, Alloy 2024-T4 Teor coe) siOY for aluminum bolts, while the nuts are usually of alloy -T6 oF 6062-T6. 29 BOLTED AND RIV CONNECTIONS ae The behavior of joints with mechanical fstners civets o bo -22, where tension T is plotted against joint displacement & a jon of load the forces are transferred from one element 10 th along the contact (faying) surfaces. The load at load at which slip 0 depends upon the magnitude of the clamping force and the secine py revi on the contact surfaces, There will be a slight increase in the ogress because various fasteners wll come into Bearing a of slip as a result of variations in the location ef the f Sip as of the fastener ‘Aft siis complete (,F, oF) the oad is tantred frm on ofter by bearing ofthe fastener onthe hole and by sear on the In riveted connections the ma magnitude of the clamping force is ii glamping force i a tanction ofthe length of the rivet and the magnitace of the shrinkage which takes place after the head is formed. This sire function of the temperature at which the dri tude of the slip depends upon the extent to which th is shown in Fig, FIGURE 2.22 Behavior of mechanically connected joints, : ‘ STRUCTURES, METALS AND FASTENERS. 65 this ease surface is known, the load forces are sufficient to produce a significantly higher case for a similar riveted connection. ‘Two types of connection are recognized. In force which the fastener is per margin of safety the force by fic The condition of the faying surfaces must be taken Slip-critical connections are recommended for joints subjected severo stress fluctuation, impact, or where slip is objectionable. type load is transferred by shear and beating on the bolt. The capacity in shear depends on whether the sheat plane intersects the body of the bolt or the threaded portion. Tests of bearing connections show that the strength of single-shear joints is increased by 40 to 45 percent if threads, are excluded from the shear plane. Of course, this is a consequence of the differ- cence in area. The RCSC specification requires that sti enough to ensure the minimum clamping force given in Table 2-7. In bearing joints it is only required to bring the connectors to the snug-tight condition. Slip-critical connections and connections subjected to direct tensions must be identified on the construction drawings. Both slip-critical and bearing-type connections are proportioned on the basis of the calculated shear on the gross area of the bolt. The permitted force provides a margin of safety against slip in the case of the slip-critical joint and a yhich slip occurs can be calculated. The clamping ip load than would be the al joints be tightened TABLE 27 ‘Minimum installation tension*, ksi, for high-strength bolts in slip- critical connections and connec- tions in direct tension66. oESiGN oF srest. sravcrunss margin of safety against failure of the fastener in the case of the beacing-ype joint, The different types the fasteners in shear are f failure that may occur in bolted or riveted joints with Fig. 2-23, Failure in which the fastener is is illustrated in Fig, 2-23 and 8, The arca sub- ted to shear is the cross-sectional area of the fastener. For the lap joint a, le shear. For the butt joint b, lure occurs on two planes, and the fastener is in double shear. Failure of the plate at 2 hol between the tical surface of fastener, is shown in Fig. 2-23c. This is ion of the compressive stresses around the hole is unknown, For design purposes the stress distribution is assumed uniform over the rectangular area equal to the thickness of the plate times the nominal diameter of the fastener. Failure may also occur in the plate between the hole and the end of the plate by shearing in the direction of the load, as illustra 2-23d, or by transverse tension, as in Fig. ly to occur if the fastener is placed at a distance from the end sufficient to make the shearing th of the fastener. For structural center of the hole to the end of the plate of the fastener. y of the applied forces, the plates of a lap joint tend to bend, as shown in Fig. 2-24a. This bending produces some tensio fastener and nonuniform bearing of the fastener on the plate, as shown in Fig. 2-24b. However, tests indicate that the bearing strength of the fastener is about the same for single-shear and double-shear bearing. Nevertheless, some specifi ‘metals this requires a distance from t of 1.25 to 2 times the diamet Owing to the eccen’ a cy a FIGURE 223 cations allow a larger unit stress for bearing in double shear than for bearing in single shear. : na ss have demonstrated that failure by tearing through the fee edge of the material istance L, measured parallel to th ie jameter of the bi lied force (Fig. 2 than the diamet Ne a of the bearing stress to the tensile strength of Ciera part. Thi ;elationship has been established from tests on finger-tight bolts. veeTne tlationshi between bearing stss and tating Os of the hole can be evaluated by considering the situation shown in Fig. 2-25. If the b« the failure planes will be along the dashed lines shown in the figu ser yative to use the solid straight lines as the failure planes. The force P transmitte by the bolts is given by FIGURE 224 Paf,Dt @ where f, = bearing stress D = diameter of bolt t= plate thickness The force to cause failure along the two shear planes is Pa ae. - Br © taken as where +, is the ultimate shear strength of the plate material, whic FIGURE 2:2568 vesicn STREL STRUCTURES O.7F,, where F, is the tensile strength of the plate material. 7 hhis equation can be rewritten in the form be_ 2 = ‘Le_1) oaa0(‘)pnare, = sane -2) o Equating (a) and (¢) and solving for L/D gives Le bt 1 Og, | 1 @ D” 14D, 32> FR, +2 This equation agrees qui th experimental data,® but there is some diver- gence as L/D increas shearing out of the pl 0 piling it up at the edge of the hole. For values of f,/F, between 1 and 3 (the practical range), Eq, (d) is closely approx- imated by Leh D*F, from which LD_fbt_P = 22. Abt 7 te F, wt Fut en In the AISC/LRED specification this relationship appears in the form P Lae 02) Lea 2) where the resistance factor ¢ is taken to be 0.75. With a factor of safety of 2, Eq. becomes: 2P Lae @3) which is the form used in AISC/ASD specificat these values are increased by one- mn. For spacing between holes the hole diameter d,, 50 Las +o 2-4) in LRED notation and (2-5) is requirement most specifications provide for @ minimum edge distance for all fasteners in any direction toward an edge. The requirements are shown in Table 2-8. The edge distances in this table, which are for standard holes, must be increased if oversized or slotted used, Specifications STRUCTURES, METALS, AND FASTENERS 69° ‘TABLE 28 ‘AISC minimum edge distance (standard hole) Nomiat fastener Seared Rolled edge oF ameter, io edge av-cutedge* i a 2 2 12 x diameter diameter which permit the use of oversized or slotted holes contain provisions to cover this, case, Allowable stresses for high-strength bolts according to the RCSC are given in Tables 2-9 and 2-10. The values in Table 2-9 are based on strength and those in Table 2-10 on prevention of slip. Slip-ctitical connections must be checked for both criteria to provide adequate strength in the event of premature slip. For TABLE 29 : Allowable working stress* on fasteners or connected material (ks!) oad condition A525 A880 “os a B in shear plane eden le no single bots inline of Force in 370 oestox oF sree. sraucrenes ‘TABLE 2410 Allowable load for slip-critical connections (slip load per unit of bolt area, ksi) eo eee eee er Unit of bolt area, ksi) Hole type and direction of load application Ay direction Transverse Parallel Oversize and Standard short slot Longeslots Long slots Contact surface of bolted parts A325 4490 ARIS AM ANS AMD AMS AMO cdeaned surfaces with Clase A coatings* 7 on ss 8 2p bs w 3 Class B (slip coefcient 050) Diast-cleaned surfaces and blast-leaned surfaces with Cass B coatings* BM mw Dp om Mu 7 ow Class C (slip coefcent 040) hotip galvanisd and roughened surfaces connections for Class B conditions are permit- ted to be stressed to 28 ksi (Table 2-10). The strength criterion for this bolt is 21 sii ids are in the shear plane and 30 ksi if they are excluded (Table 2-9). the threads are in the shear planes, the bolts must be proportioned on the basis of 21 ksi (the strength criterion), but if they are excluded from the shear planes, they are proportioned on the basis of 28 ksi. Allowable stresses for high-strength structural bolts specified by AISC, AASHTO, and AREA are presented in Table 2- houid be noted that able tension is independent of the size of the bolt, even though tens vary with the size (Table 2-6), In assessing the factors of safety whi able stresses provide, it will be noted that the ASTM tensile properties are on the stress area, while the allowable stresses in structural spe based for convenience on the unthreaded (“ nominal 4 Lin bolt the stress area is 0.606 in?, while the nominal area is 0.785 in?, The recommended allowable tensile stress of 44 ksi for A325 bolts gives a factor of safety for L-in bolts of 0.606 4 Factors of safety for the sizes listed in Table 2-6 range from 1.76 to 2.35 for buildings. FS. = 120 0.785 = 2.11 STRUCTURES, MEALS. AND FASTENERS TL ‘perfation allowable stresses for high-strength structural bolts ksi) as 90 contin AISCIASD AASHTO AREA AISCIASD AASHTO ARPA Ta «5 4 M&S rd tle wos mmo Overt te B bs i ® i Sher, ning ye tearpine 2S es harps 37} o & Wa} SRE sng akg SR LR rl le. regth connected meal § Special paraison by the eginece equ Results of numerous tests show that the shear strength of a single A325 bolt is 65 to 70 ksi. Thus, the factor of safety for a single bol for a bearing-type connection, 30 ksi, s about 2.3, However, of bolts in the line of stress, as in a long connection, this factor of safety is reduced if the shear is assumed to be divided equally among the bolts, as is customary (Art. 3-10). AISC/ASD, AISC/LRED, and AASHTO require that for members the tabulated values be reduced by red area of the surface. Tests have shown that the strength of the net ber is not affected if the bearing stress is not larger than slip is easily coefficient of friction allowable shear stress, so the factor of safety against uN. AyFy FS.= where z= coefficient of friction N = specified pretension in bolt Ay = cross-sectional area of bolt F, = specified allowable shear stress on bolt.72 vesiox oF sree srmucruRES Assuming « = 0.33 (Class A) and the appropriate values of N, calculated factor of safety for A325 conne A490 bolt connections it varies from 1 Kulak, the factor of safety on the basis of the of slip, method of installation of the bi related to the size of the holes. In any joint there will be variations in the clamp- ing force and the cosfficient of friction, Based on their evaluation there is about a 10 percent probability of slip for A325 connections when the allowable stresses of AISCIASD are used. Table 2-12 gives allowable stresses for AS02 rivets according to AISC/ASD, AASHTO, and AREA. Although there are no tensile requirements for AS02 ENets (hardness is specified instead), Grade 1 corresponds to a former ASTM designat h the specified tensile properties were 28 ksi yield and rength. Thus, it would appear that the allowable tension for Grade 1 rivets, 23 ksi, gives a low factor of safety relative to yield, However, the yield stress of the in-place rivet exceeds that of the undriven work hardening caused by driving. Furthermore, the area of the undriven rivet, which is sm an that because rivet holes are in larger than the nominal stress 20 to 25 percent he allowable tension of 23 ksi gives a Tespect to yield on the order of 1.5. The factor of safety with timate of the undriven rivet is at least 2, The Grade 2 rivet also corresponds to a former ASTM designation, A195, for which the tensile Properties were 38 ksi yield and 68 to 82 ksi tensile si respect to tensile allowable tensions for the two grades, 29/23, and 38/28, differ only slightly. ‘Tensile-strength and shear-strength criteria of the AISC/LRED specifi- fation are presented in Table 2-13. Allowable stresses in bearing recommended by the RCSC, presented in Table 2-9, have been adopted in the AISC/ASD spec ifcation. If deformation around the hole is permitted, they may be increased to ion the ASD values are multiplied by 2. Allowable stresses (ksi) for A502 rivets ED er ASME rete Grade 1 Grade 2 Condon AISC AASHTO AREA AISC AASHTO. Tension 23 8 Shear 1753S BSR ow Beating 12 a 0 36 : oneted pat ie seas, ssraucrones wetats. AND exsrenens 73 AISC/LRED design strength of fasteners a = eee an = eo pores oot ies aoe a cee Steer patrdbeo rae fscmeet tee ors arses 078 oar, Bo percent of he ukimate shear song. Se Ar. “Therm ster os ‘The soil ae seh he ded portion os pet bd pe he ena std dat eal tear an tr cl ey ate od Wee sie es, set se son ese steeper whos gt eed ler fr S0 pene Source: Adapted rom the 1991 AISCILRFD Speciation sample 291 (AISCIASD). Determine the nanber of bls tei: Soyo yoo anomie dead foo ef ia ane foes Bos ate in AS2S(Gandard als) ns beaings¥pe connection freed fom the sent ane Ue the ns of ate actos the Web of te channel Solution Design force = DL + LL = 320 kips4 FIGURE 225 For in bolt, A= 04818 in? Fy=30ksi (Table 2.13) R, = 04418 x 30 = 13.25 kips/shear surface ‘Two shear surfaces per bolt. Number of bolts required is 320 Sor Pea Use 12 Beating. Use Fy = LF, = 1.2 x 58 696 ksi ‘This requires that the end distan« (Table 29) Channel we be Lid and the betweenastener spacing 3d R= dtl, = 3 x 0673 x 69.6 = 35.13 kips ‘There are 24 bearing surfaces, 60 Capacity = 24 x 35.13 = 843 kips > 320 Gusset Rim dF, = 3x 1 x 63.6 = $22 kips ‘There are 12 bearing surfaces, 60 Capacity = 12 x 52.2 = 626 kips > 320 STRUCTURES. METALS, AND easreneas 75 Spacing End distance, sheared edge: If ia minimum. End spacing: Ld = 1.13 in < 1 in Center-to-center spacing for F, = 1.2F, is 34 = 24 in, iderably larger than inimum between- Since R, for both the channel and the gusset is co ‘Then, from Eat 2p oP 2. P= 2A ki eo 1S= =e Sai ps (Ea. (231) wid 2P i regetSaeaaa ty P= 3h7 kis [Ea Capacity a 3 x 244 +9 x 31.71) = 717 ips P = 3625 kips (Eq. (23)) 2 exit? Capacity = 3 x 36.25 +9 x 47.13 = 533 kips Therefore, 12 bolts in three rows of four are adequate with end distance of 1 in and center-to-center spacing of 2 in, ‘Although the calculations show that the minimum spacings may be used, ‘most designers would use the end distance of I4d and betweer-fastener spacing of 3d, which permits a bearing strength of 1.2F,. The minimum spacing would ordi- narily be used only in special situations P= 47.13 kips (Bq. 2) a 2 Example 2.9.2 (AISC/LRED). Determine the number of bolts required, and an appropriate layout, to transmit a dead-load force of 80 kips and a liveload force of 240 kips through two C10 x 30+to a I-in gusset plate (Fig, 2-26). Ail material is A36. Bolts are dein A325 (standard holes) in a bearing-type connection with threads excluded from the shear planes. Use three lines of bolts across the web of the channel, Solution Design force = 1.2DL + LOLL = 1.2 « 80 + 16 x 240 = 480 kips For din bolts, A= 04418 Resistance factor $= 0.75 (Table 2-13) eth = 60ksi (Table ‘Norninal shear76 vestow oF stee1 stmucrunes Bolts are in double shear. Therefore, shear capacity per b Ryo 2 x OAALB x 0.75 x 60 = 39.75 Kips Number of bolts required is 350 121 Use 12 39.76 ie Beariog aod spacing End distance, sheared edge: 1 in minimum (Table 2-8) Bins 1} spacing: 3d = 2.25 in When end distance is 1}d and center-to-center spacing is 3d, the bearing strength is 24F, Beating Channel R= 24deh, = 24 x2 x 0,673 x 58 = 7026 kips 620.75 GR, = 52.7 kips ‘There are 24 bearing surfaces, so Capacity = 24 x $2.7 = 1265 kips > 480 kips R= 24iP, 2A 9x 1 $8 = 1044 kips $075 OR, = 783 kips ‘There ate 12 bearing surfaces, so Capacity = 12 x 78,3 = 940 > 480 kips Spacing between connectors could be reduced by minimum spacing of 2d is used instead of 34, the required Channel. End distance L,, Eq. (2-2), For this case $ is 0.75 ‘ying Ea PoP at sex ie GF” 5 * 58 x 0673 P = 36.59 kips STRUCTURES, METALS. AND FASTENERS 77 Between connec 8 L, Eq. (2-4) Poak feat im P. “075 x 58 x 0673 P= 45.14 kips Capacity = 213 x 36.59 +9 x 45.74) = 1043 kips +i +d Gusset, End distance L,, Eq. (2-2) F Wea geyn age P= 54.38 kips Between connectors L, Ea. 2-4) hee 2 Get P aw tet) P= 6197 kips Capacity = 213 x $438 + 9 67.97) = 775 kips 12 bolts in three rows of four are adequate with end distance of center spacing of 2 in Although the calculations show that the minimum spacings may be used, most designers would use the end distance of Lid and between-fastener spacing of 3d, which permits a bearing strength of 247,. The minimum spacing would ordi- natily be used only in specials Allowable stresses for aluminum rivets according to the Specifications for Aluminum Structures of the Aluminum Association? are given in Table 2-14. It means, of course, that connec 1ed to avoid tension on the rivets as much as possible, of evaluating stce! rivets in terms of thei undriven ameters, tests on um rivets are interpreted in terms of the diameter of the hole. Diameters and areas of rivets and the recommended diameters of the corre- sponding holes are given in Tables 2-15 and 2-16. The allowable shear stresses in Table 2-14 are based on factors of safety, with respect to shear strength, of about 2.25 for buildings and 2.65 for bridges. The allowable bearing stresses are based ‘on factors of safety, with respect to bearing yield strength, of about 165 for buildings and 1.85 for bridges.78 orsicn oF sree sravcroRes TABLE 214 Allowable stresses for aluminum rivets™ les apply for a ratio of edge distance 10 allowable stress by the ra Designation Designation ‘Shes ater Driving of alley — airing procedare fabricated Bridges Duildings Bridges’ Buildings ed _ellgenDaildings Bridges 6061-76 Cold asreceived —O6I-TS 10 fn 30 ey STE Cold, as received 6063-75 73 as t 16 6063-76 8s 2 24 6061-783 Hot, 90-1050"F 6001-76 8 ° 30 rm 6063.75 8 9 t 16 c06t-76 a 9 2 4 TABLE 215 Hot-driven alumisum rivets Rivet diameter, 210 WELDING PROCESSES Most specifications require that the provisions of ANSIAWS D1.1 Structural Welding Code be satisfied. That code contains detailed provisions for design, ‘qualification, workmanshi ing metal parts by means of heat and pressure, which causes fusion of the parts (resi the fusion temperature, with or usually employs either an electric are or an oxyacety- lene fame to heat the metal to the fusion temperature, The electric arc is used for most structural welding, ied according to their type as groove, fillet, plug, and s A groove weld is made in the opening (called a groove) between two parts being STRUCTURES, METALS. AND PasrENERS 79) ae ECE Groove Fires Groove Fillet (0) Flot {61 worizentot oe OL Filer Groove f (a1 ver which is triangular in shape, joins surfaces which are ‘at an angle with one another (Fig. 2-27). A plug weld is made by depositing weld metal in a circular hole in one of two lapped pieces. The hole must be filled completely. A slot weld is similar, the only difference being that the hole is elon- gated. Holes and slots can also be fillet-welded around the circumference, but these are not plug or slot welds. Welds are classified according to the po: flat (also called downhand), horizon position is executed from abot (Fig. 2-274). The horizontal posit (Fig. 2-276). Work in the overhead position is from the underside is the most difficult weld to make (Fig. 2-27¢). The longitudinal axis of the weld is vertical in vertical-position welding (Fig. 2-274) In metal-are welding the aro is a sustained spark between a metallic elec- trode and the work to be welded. At the instant the are is formed, the tem- to be deposited in the molten base metal. Tt is because these globules are actually impelled across the arc that the process can be used in overhead welding. The molten metal, when exposed to the air, combines chemically with oxygen and nitrogen to form oxides and nitrides which tend to embritile it and make it less resistant to corrosion. Tough, ductile welds that are more resistant to corrosion are produced if the molten poo! ‘ded by an inert gas which completely envelops the molten metal and the tip of the electrode. This shielding can be provided in a variety of ways, and the manner in which ined can usually be determined from the name or designator used to indicate the process to be used.‘80. vesion OF sreet srmUCTURES Many types of welding processes are used in arc welding. They are manual or automatic, Other terminology, such as shielded-metal-arc welding (SMAW) or submerged-are welding (SAW), is also used, ‘Most manual welding (stick welding) is performed with the shielded-metal- are process. In this process the electrode is placed in an electrode holder to estab- ing is obtained by the of electrodes heavily coated with a material of su quantities of gas are produced in the heat of the arc. The coating burns at a slower rate than the metal core does, thus directing and concei 1e atmosphere. The metal and protect removed after the are-welding processes produce high- welding speeds. They are commonly use process is called submerged-arc weld hidden-are) a tain a co length. The welding is shielded by a blanket of granular, fusible material which is fed onto the work area by gra submerge the arc completel -29). Some of the granular material idition to protecting the weld from the atmosphere, the covering aids in controlling of cooling of the weld. Multiple-clectrode welding uses two or more small weld wires, instead of the single wire, for increased speed at reduced cost. Submerged-arc welding must be performed downhand. Welding speeds for one-pass groove welds range from 30 in/min in t-in plate to 8 in/min in 14-in plate. The high currents used cause considerable melting of the base metal, so less filler metal is required and the Joint opening may be less than that necessary for other types of welding. In flux-cored are welding (FCAW), shielding is provided by a flux contained within a tubular electrode and may be supplemented by an externally applied gas. In both applications the electrode core material produces a relatively thin slag covering to protect the solidifying weld metal, Flux-cored arc welding is usually a semiautomatic process where the gun, which controls the rate of feed, is held and manipulated by the welder. The process may be used in machine Welding, in which the operator monitors the are during the mechanized travel. ry welds at very high a and fabrication. One FIGURE 2.28 Shieided-metal-are welding, (Linzotn Electric Cam- pany.) — | from atmospheric elements, the welds are stronger, more du sraUcrunes METALS, AND Fastinens SL Electrode can ee Pan dango pot “Bare metal j ricune 229 \ pet mt Sionerpedaeaate wel Inert-gas shielded-are welding is usually performed without flux. The arc id the weld region are shielded from the atmosphere by a gas or gas mixture or rs ot ion to the inert gases argon and helium, carbon is dioxide is popul 's low cost. Since they are com; corrosion-resistant. High welding speeds with thorough pene le distortion are obtained because of the intense arc heat. In gas metal-arc welding (called Mig, for metal-are inert-gas), a bare-wire consumable clectrode is fed automatically into the arc and deposited as weld metal, In the tungsten-are process (Tig, for tungsten-arc, inert-gas), an arc is struck between a virtually nonconsumable tungsten clectrode and the base metal. Filler metal, if required, is added by feeding a welding rod into the weld pool. ; ‘The basic elements of the electroslag welding process are shown in Fig. 2-30. Heat is generated by passing an electric current through molten flux which melts the electrode and the edges of the base metal, Welding is usually done in a Wire feed rolls and oscillating mmechaniem mounted on corri¢ge thot rises ovtomatically 0: eld metal builds up Face of plate #1 Consumable Woter cooled copper slides confine molten Wold metal ‘fag and weld metal Weld metal a Thictness Completed weld of plate Face of platete FIGURE 230 Blecroslag welding.82 pesto oF ses: sreucrurs preparation is simple since only square, edges are necessary. Plates 18 in thick can be welded in a single pass, fab is needed to build up the proper depth of flux to ensure complete fusion of the base metal, A runoff tab is required at the top because the slag depth must be carried beyond the base metal. The slag bath is 13 to 2 in deep. though si s basic elements to the clectroslag process, the electro- ing process uses an open electric arc instead of a molten flux to melt the bbase metal and electrode. Since the are causes instantaneous fusion and no fiux bath is required, starting and runoff tabs are not necessary. Joint thicknesses of $ to 3 in can be produced with a single pass, Electroslag or electrogas welding produces relatively large grain sizes as a result of the slow cooling rate of the weld. This condition also produces carbon portion of the weld which cools first to that part which cools last. As a result, the center section of such welds may consist of material wi relatively latge grain size and increased carbon content. In ad residual stresses may be present as a result of the constraint being joined. Under these conditions the weld could be suscepti tures. This condition is more critical in the presence of a flaw. discussed in Art, 2-16, Electtoslag or electrogas welding of ASTM AS quenched and tempered atter Aluminum alloys can be joined by arc welding, resistance welding, gas iB, oF brazing, The most commonly used procedures are welding. They can be used on metal 75 in or more thick. Tig weldi Used for material thicknesses from 0.05 to 0.25 in, The commonly used structural ints made in aluminum al are usually 100 percent efficient. With alloys in the strain-hardened or heat-treated tempers, however, the heat of welding affects the metal on each side of the weld so that it is not as strong as the parent metal. 241 WELDED JOINTS Welded joints are classified as but ¢, commer, and edge (Fig. 2-31). The butt is fillet-welded. The tee joint can be ‘groove-welded, as shown, or it can be fillt-welded with one fillet on each side Groove-welded joints can be complete-penetration or partial-penetration joints. In some cases the penetration is intention: than the thickness of the part joined, is part welding procedure docs not produce effective penetration in what might appear otherwise to be a complete-penetration joi A number of joints can be used wi tests are required to demonstrate th ut qualification, which means that no adequacy. Some of the manual shielded- FIGURE 2.32 ‘Typical prequalified shiolded-metal-are joints for bu STRUCTURES METALS, AND FASTENERS 83 {oh Bor joo? (eh Lop jorar FIGURE 231 ‘Types of woied joints, shown in Fig. 2-32. Two of them, a and 4, are complete-penetr limited in use to the thicknesses shown. Four, c to fat 7 [rurnon {0} Square ~greove bere Se ABDDOV Ve Dose ree groove ‘one Fela |z-0-£ Tawa Pim io) Square groove devel utr84° enon oF sreet sraucrunes The square-groove joints, a, b, and i, require no preparation of the edges of the parts to be joined. The type of groove for other than square-butt joints depends in part upon the thickness of the material and the position of the weld and whether one side or both sides are accessible for welding. Single grooves, such as, the bevel aud the V, J, and U, are cheaper to form but require more weld metal than the double-grooved joints. For example, the single-V joint requires approx- imately twice as much weld metal as the double-V. The choice between single and the higher cost of preparation FV joints are usu preferred for horizontal welds because to make a good U or is position. The part of a weld which is assumed to be effective in transferring stress is ion welds is welded from one side without backup. In the remaining complete-penetration joints, the root of the weld deposited first must be “gouged” before welding from the other side, which means that it must be cleaned. The fillet weld is quite common in structural connections. The faces of the weld which are in contact with the parts joined are called its legs. The size of an equal-legged fillet weld is given by the length of the side of the largest isosceles right triangle that can be inscribed within the weld cross section. This triangle is called the diagrammatic fillet weld. The throat is the shortest distance from the root of the weld to the hypotenuse. For the equel-legged fillet weld this dimen- sion is 0.7073, where sis the leg size. (In the case of the submerged-are weld, there is an exception to tion in the AISC specification, discussed in Art. 2-13) ‘The legs are usual jut conditions sometimes require unequal legs. Large filet welds made requi or more passes, as indicated in Fig. 2-33. Each pass must cool, and the slag must be removed, before the next pass is made. are those which can be made in one pass. and should not exceed the following: 4 in in the horizontal or overhead position ion in in the flat po 4} in in the vertical a Bificiency of welding is also affected by the amount of filler metal req increases by the square jeg size as the throat dimension For example, a #-in filet weld has a throat only 25 percent greater than that of a d-in weld, but its volume is 56 percent greater. are STRUCTURES, METALS, AND FASTENERS 8S E233 pas let weld ‘The most commonly used filet welds incresse in size by sixteenths of an inch from } to } and by eighths of an inch for sizes greater than § in. To ensure full throat thickness, the size of the weld should be less than the thickness of the edge of the connected part. The maximum size of fillet permitted by the AWS along the edges of material + in or more in thickness is 7 in less than the thickness of the material. When it is impracticable to obtain a sufficient connec with these maximum sizes, the specifications permit a weld of the same thickness as the edge, provided this information is designated on the drawings. ‘Along the edges of material less than } in thick, the weld size may be equal to the thickness of the mat 'A filet weld that is too small compared with the thickness of the material being welded is affected adversely during cooling. The amount of heat required to ceposit a small weld is not sufficient to produce appreciable expansion of the terial, and as the bi is restrained by control this situation, speci deposited on the basis of the maximum thickness of the parts to be joined, as sven in Table 2-17 Groove-welded joints are more elficie because of their greater resistance to repeated st ferred for dynamically loaded members. Groove welds n ‘metal than fillet welds of equal strength, they also frequent for extra metal in the form of connecting plates or other structural shapes. Never- theless, fillet welds arc often used in structural work, partly because many con- TABLE L17 Fillet-weld sizes Size of filet Maximum thickness rel, in of parg in Ted Over deo}86 orsion oF sree. stauerres nections are more easily made with fillet welds and partly because groove welds Fequire the member of a structure to be cut to rather close tolerances, Cold-formed members are usually shop-fabricated by spot welding (resistance welding). In this process the parts to be welded are clamped between {wo electrodes. Resistance of the metal to a strong current passed through the din It is customary to take the force on a fillet weld as a shear on the throat irrespective of the direction of the load relative to the throat. Thus, in Fig. 2-366 the load P produces on the throat of each weld both a shearing component and a tensile component, each equal to PZ. In spite of this fact the stress is determined from Eq. (c). Tests have shown that a much stronger than a fillet w. Forest B%07e wa io "nal weld and that for long welds it may be considerabl STRUCTURES METALS AND FasreNERS 89 f FIGURE 237 pottent, as was shown above. Therefore, it is reasonable to expect the strength of the transverse weld to be intermediate between the shearing and the tensile strengths of the weld met at working loads the shearing stress is not uniform ove the center of the weld. However, since ‘under increasing load, that the stress approaches a uniform ic loading, 7 shows a lap joint transverse ; but it is impossible to determine by statics alone how the load will be divided among the three welds. However, it is customary to assume that the stress is | distributed among the welds in proportion to their lengths, if all welds are the same size. If the weld on the end of the member is different in size from the welds ‘along the sides, the load is assumed to be distributed in proportion to the area of, the welds. ‘Any abrupt discontinuity or change in the section of a member, such as a notch or a sharp reentrant corner, interrupts the transmission of stress along smooth lines. The magnitude of the stress concent ness of the notch or the abruptness of the discontinui eluding those at the ends of longitudinal welds discussed above, are of loads or for cases where only a few thousand repeti- ely to occur. However, they are significant where load to produce a more uniform ‘of Fig. 2-380, Specifications do ress for such a weld. o (01 FIGURE 2.38} 90 bestow oF sree. sreucTuRes : 2:13 SPECIFICATIONS FOR WELDED : CONNECTIONS. Welding electrodes are classified on the basis of the mechanical properties of the weld metal, the welding position, the type of coating, and the type of current required, Electrodes for shielded-metal-are welding (SMAW) are covered by AWS A5.1-81 and AWS AS.5.81, Each electrode is identified by a code number EXXXXX, where E stands for electrode and each X represents a number. The first two (or three) numbers indicate the tensile stcength (kips per square inch) of the weld metal. The next number denotes the position in which the electrode can be used, the number 1 meaning all positions, the number 2 flat and horizontal fillot welds, and the number 3 flat welding only. The last number denotes the type ; of covering, the type of current (alternating or direct), and the polarity (straight ' or reversed), Straight polarity means that the electrode is negative. For example, ' an E7018 electrode has a tensile strength of 70 ksi, the number 1 means that it can be used in all positions, and the number 8 means that it is an iron powder, Jow-hydrogen electrode which can be used with either alternating or direct current but only in reverse polarity. Combinations of flux and electrodes for submerged-arc welding (SAW) are covered by AWS AS.23-80. A flux is designated by the letter F followed by two digits denoting the tensile strength and Charpy V-notch impact strength of the test weld. This is followed by a set of letters and numbers denoting the electrode i used to classify the flux. For example, the letters EL in EL8 signify a low. : ‘manganese electrode, while the number 8 denotes the chemical composition in percent of carbon, manganese, and silicon. Electrodes for gas metal-arc welding are covered by AWS AS.26-78 and are identified by similar notation. Thus, E70S-X and E70-U1 are solid and emissive electrodes, respectively, where the last ‘number identifies the chemical composition, i The mechanical properties of electrodes are given in Table 2-18. For the full-penetration groove weld, most specifications presctibe tensile stresses equal : to those for the base metal and require that “matching” weld metal must be used. Be In compression the allowable stress is the same as for the base metal, but the ‘ weld metal may have a strength equal to or less than the matching weld metal, Shear must be checked on both the base metal and weld metal. Filler-metal requirements are given in Table 2-19. Partial-penetration groove welds and fillet welds may be made with filler metals whose tensile properties are fess than those of the base metal. The per- missible stresses for this sitvation are based on test results evaluated according to i the assumptions as to stress distribution which were discussed in Art. 2-12. Figure 2-39 shows the results of 168 tests on fillet-welded joints.'? Weld sizes wore 2, 8, and } in, base metals A36, A441, and ASI4, and electrode classi fications £60, 70, 90, and 110, The combinations of base metal and electrode tested are given in Table 2-20. Combinations of strong weld metal with weaker i base metal, and vice versa, showed that the effect of dilution upon weld strength was quite small. For example, the strength of din fillet welds made with E110 ‘Mechanical properties 110-125 ESOT eso ent E708-X 80s E1008 E110 FOX-EXXX IXEXXX PEX-EXXX FION-EXXX LIXEXXX only to bide Mechanical properties of electrodes TABLE 2-18 B6OXX BIOXX BADXX EWOOXX ELOXX, 1 pptTABLE 219 Filler-metal requirements for complete-penetration groove welds* ilded me are Suhmerged sre Plux-cored are ‘A36, AS3 Grade B, A500, A501, A529, 'ASTO, A709 Grade 36 EGOXX oF ETOXX*Y POX oF FIX-EXXX BT0S.X of B70U-1 BAOT-X oF ETOT-X |A2#2, AGA, AS72 Grades 42 to 55, ‘AS8K, A709 Grade 50? ETOXXS FIX-EXXX 1E70S-X oF B70U-1 ETORX, A572 Grades 60 and 65 80S FRXCEXXX Grade EXOT [ASI4, A209 Grade 100; 22h in ELIOxxG FUIX-EXX Grade EAT 1S dhin BLOOXX§, FIOX-EXX Grade E1007 * Use same type filer ual having next-higher mechanical prope s penned 1 Low-indcogen erodes must be sed fer welingA36 sel more than tin tik or begs 4 For srcitcturl exposed bare ungsinted applications, te deposited elt metal and the ba etal must have sonar atmospbese-crcoion resistance and coloring characte. Snes. Pollo ste manvlactier esrmmendation. S Low-ydrogn desications Woe eee 3p PeR Uni seam wi aoa Zea] [BER e Ffa os 8 F BRR) JERRY [FF og a i zo OEe ee | 3 et BS Fed SS f s «| lrfes| JE Se SSS S| 8 “li fF BF ¢ i : ect ee it wxx] fo z shite . eee 5 < 2 = 5 Elex[El~ [8g i : ial elle ele | 3 fy. ee aww xx [Eleex lal eS sos Scene e Fl lenlkIS 5 a3 8 ary ~/EE g a 8 BR |e ie ge 8 ee ff i ; fe 8g 7 ~|EZ ' ge 3 7 7 : g 3 ‘gisidclnis ae eee eeu 694 estoy oF srect sraucruses electrodes on Addi steel, whose average tensile strength was 79 ksi, was only 8 Percent less than that of the same electrode on ASi4 steel with an average tensile strength of the strength of 4-in welds of E70 electrodes on ASI4 steel was only 2 percent greater than that of the same electrodes on A441 steel The 36 transverse-shear tests were made only on joints with 4-in welds. It will be noted (Fig, 2-39) that weld strength is considerably greater in transverse shear. For example, the E110 electrode on A514 steel produced a weld which was 32/18.6 = 1.72 times as strong when tested transverse to the load as it was arly, the E70 weld on A36 steel was 22.8/15 times stronger transversely. Specifi ‘The superior penetration of the submerged-are fillet weld justifies a more iberal definition of the thro: throat thickness 7; as the leg size for 3-in and s, and 0.11 in more than the theoreti I throat for larger welds, AISC/ASD. Allowable stresses in welded connections are given in Table 2-21 Shear stress for complete-penetration groove welds, welds, fillet welds, and plug and slot welds tensile strength of the weld met AISC/LRED. Design strengths of welds are given in Table 2-22. The shear strength of complete-penetr penetration groove welds, id plug and slot welds is 0.6 times the nominal tensile strength of ‘or = 0.80 for complete-penetration groove welds, while @ = 0.75 for the other three cases, AASHTO. Allowable stresses in this specification are more conservative than the AISC specifications. The allowable stress on fillet welds is 0.27F,, where F, is the tensile strength of the electrode classification but not greater than the strength of the connected part. For plug welds the allowable shear stress is 12,400 ps AREA. Allowable shear stresses on fillet welds in this specification are given as a function of the base material and the strength of the weld metal. They are 36, Electrode or electrode-ftux combinat 60,000 psi tensile strength 16,500 psi 70,000 psi tensile strength 19.500 psi but not to exceed 12,500 psi shear stress on base materi Ad41, AS72, AS88, Electrode or electrode-flux combinations with: strength 19,000 psi 80,000 psi tensile strength 22,000 psi but not to exceed 0.35F, shear stress on base material, ns with: STRUCTURES, METALS, AND FasreNERS 95 BLE 221 ISCIASD allowable stress on welds “Type of weld and stress Allonable stress Required weld strength ‘Complete penetration groove welds Same as base metal Matching weld ust be used Same as base metal Same as base metal Shear on effective area 0.30 x nominal wed metal ks) o Partial-penetration groove welds “Compression normal io Same as base metal elective azea ‘Tension or compression _ Same as base metal matching be used ath of base metal Fillet welds ‘Shear on effective area matching weld metal may bbe used ‘Tension or compression Samo as ase metal parallel to axis of weld sand slot welds tensile strength Weld metal with a st level equal to Seance: Adapted tom (989 AISC/LRFD Spesigeaon96 estoy oF stesi staucrunes TABLE 222 AISC/LRED design strength of welds x ead Type na Recast ‘on momo Mowisl for Fawr y tel Complete penetration groove welds Base 090 Fr, Compression normal to Base 090 F, fective : metal may be used Tension or compression Base 090 r, parallel to axis of weld ‘Shear on effective area Base 090 oor, Weld electrode 080 060F eax artakpecetation groove welds Comprnion ermal to Bae 090 we effective area, eee cis cae matching wel wt maybe ed Temon competion Bae om parallel to axis of weld a Shest pra toa base ceed We cecvads 075 Fay Teaion somali Base om tex area Weld eecvode 08 00D Fane Filet welds ‘Stress on eetive Base area Weld dectrode 0.75, 060F exe matching Weld metal may be used n or compression Base 990 5, ais of weld faying Base Weld metal with @ (oneffecive Wold electrode 0.75 060i equal to of less than matching weld metal may be used STRUCTURES. METALS, AND FASTENERS 97 Example 2131 (AISC/ASD). Design and detail the welded end connection required to transmit a dead-load force of 80 kips and a live-load force of 300 kips through two C10 x 30 to a {-in gusset plate (Fig. 2-40), All material is A36. Welds are to be deposited manually using B6OXX electrodes. c10x 30 Ay 282 in? Iy20.573 0 Design force = DL + LL = 380 kips or 190 kips per channel Minimum weld size = # in (Table 2-17) Maximum weld size along web = 0.673 ~ 7g = 0.610 in Use in weld, which requires only a single pass. The weld throat is, 0.07 x ye = 0221 in y= 03 x Fyxx = 03 x 60= 18 ks Therefore the capacity of 1 in of zh-in weld is q = 0221 x 18 = 3.98 kips and. the required length Lis 190/398 = 47.8 in. Use 10 in at the end of the channel and 19 in along each flange fora total of 48 in. ‘The shear capacity of the gusset plate must also be checked. The procedure for this check is discussed in Art, 34 and presented in Example 3-4-1 (AISC/ASI ‘The length of the overlap of the channel can be reduced by using Weld, which can be accomplished in a variety of ways. For example, @ 7 would require only 35.2 ia of weld. Overlap could then be reduced to 13 in for a total of 36 in of fillet weld. Example 213-2 (AISC/LRFD). Design and detail the welded end connection - of 300 kips in. gusset plate (Fig. 2-40). All ms ssing EGOXX electrodes.98 estan oF stexL sraucruREs Solution Design force = 1.2L + 1.6LL. = 12 x 80+ 16 x 300 = 576 kips or 288 kips per channel Minimum weld size = in (Table 2-17) Maximum weld size along web = 0673 — 5 = 0.610 in Use fin weld, which requires only a single pass, The weld throat is 0.707 x f= 0221 in F006 x Fayy = 06 ¥ 60 = 36 ksi $= 075 ‘Therefore the capacity of 1 in of yin weld is q = 0.221 x 0.75 x 36 = 597 kips and the required length L is 288/597 = 48.3 in, Use 10 in at the end of the channel and 19% in along each flange for a total of 49 in ‘The shear capacity of the gusset plate must also be checked, The provedare {or this check is discussed in Art, 34 and presented in Example 3-4-2 (AISC/LRFD). ‘The length of the overlap of the channel can be reduced by using a larger weld, which can be accomplished in a variety of ways. For example, a 7-in weld ‘would require only 353 in of weld. Overlap could then be reduced to 13 in for a total of 36 in of filet weld. 214 WELDING QUALITY CONTROL ‘The production of sound welds is governed by many factors. The type of joint, its preparation and fit-up, the root opening, ete., are important, as are the welding n, the welding current and yoltage, the arc length, and the rate of travel for the welding operation is also important, as the qi is determined to a considerable degree by the position of the el fillet weld, the electrode ordinarily should angle between the weld. Furthermore, it must lean about 20° in the din in Fig. 2-41a and b empl favel. The welding procedure was € rest, Shrinkage of the outer portion may cause cracking, as in the weld at the left. The pear-shaped weld at the right in Fig. 2-41d shows a similar crack. The American Welding Society publishes weld qualification procedures. Procedure quali deals with properties of the metals, type of groove and STRUCTURES. METALS, AND FASTENERS 99 “FIGURE 2-41 fect of welding technique on quality of weld. (State of Calornia, Deparonent of Publle Works, electrode type and size, current and voltage, and require the base metal. The operator must also be qualified by ‘which must demonstrate the required strength distribution of weld defe Magnetic-particle magnetic field in a short sec oles develop where ete arc leakages in the field, which occur at discontinuities in the weld, and iron wder sprinkled on the test area migrates to the poles. The resulting pattern tends to outline the discontinuities. In dyc-penetrant inspection a dye is brushed or sprayed on the surface of ‘the Weld. It seeps into surface irregularities. A developer that is sprayed on is ‘tained red by the dye, which rises from the surface defects by capillary action, A100 ossiox oF sree srauerves lar procedure is based on using a fluorescent liquid which detects surface imperfections upon exposure to black light. Radiographic inspection uses shortwave radiations, such as x-rays or gamma rays, to discover surface and subsurface flaws in the weld. The beam ‘encounters less resistance at a defect, and when the radiation is recorded on film, the defect is disclosed. Ultrasonic inspection is also effective in locating subsurface weld defects as well as those on the surface. High-frequency sound waves sent through the atea to be inspected are reflected by discontinuities and density differences. The reflec- ted sound waves are monitored by a receiver, converted to electric energy, and displayed as visual patterns on an oscilloscope sereen. In interpreting the results of inspection of a weld, it is important to assess the severity of a defect as it relates to service requirements. Some defects may be relatively unimportant, while others may be critical in specific service situatic 2.15 FATIGUE Fracture of metals is not always preceded by yielding and the subsequent elon- gation described in Art. 2-4, Instead, it may occur and almost always occurs at a point of stress concentration ‘eld, notch, or even a scratch. These stress raisers are usually on the surface, but they can also be as at a defect in a weld. Because there is xo plastic deformation of the mate igue cracks are hard to detect even on the surface of a member. They usually propagate very slowly and intermittently. ue failure of a laboratory specimen consisting of two plates joined by a shown in Fig. 2-42. The specimen was tested in repeated tension in the longitudinal direction of the weld. Cracking began at an irregularity at the surface of the weld and propagated radially into the plate, This progressive cracking usually results in the “oystershell” or “beach” markings visible in the Photograph. Behavior under repeated load is evaluated in rotating-beam tests, flexure tests, and axiaload tests, In the rotating-beam test a round, polished specimen supported as a simple beam is rotated at constant speed while being subjected to a bending moment so that every fiber of the specimen alternates sinusoidally between tension and compression. A polished specimen is also used in the lure specimens of ASI4 steel for three different ided specimens are subjected to stress eycles ranging from a sraucrunes setaLS.anD Fasreners 101 FIGURE 2.42 pial fatigue factore in 4 welded plat. (Department of Civil Engineering, University of Minis) 17 ot 1 rg oF FIGURE 2-43 SN diagrams for polished specimens af TI stel in ai “BS Steel Corporation, November 1968) se. (From “USS Stee! Design Manel,”102 vestow oF steer Value to a maximum. R denotes the ratio of minimum to maximum R= 0 (upper curve) denotes stress ranging from zero to tension; (middle curve) denotes stress alternating between tension and compr al excursion from the minimum stress the range of stress, A positive value of R lesser and @ maximum tens value of R indicates a reversal of the sign of the stress during the The data in the figure are plotted with the maximum stress in the ‘ordinate. $-N curves tend to become horizontal at large values of sponding strength is called the fatigue limit, The sidered to correspond to a fatigue life of about 2 mi It will be noted that the fatigue limit for one-half the tensile strength of the steel for polished specimens, severe stress conce! may produce drastic reductions, as in Fig. 2-44. This figu reductions can be so large as to make high-strength steels | fatigue than low-strength steels. ‘The data in Fig. 2-43 are presented adapted to design. in Fig. 2-43 is about is is a good approximation 1899. Bach curve is the locus of all the and the diagram covers the full range Of stress ratios —1< R <1. Minimum stresses ate plotted as abscissas and maximum stress inate. Rs ies from the origin correspond to the various stress denotes no reversal of stress, the ray R= 1 corresponds to ie curves for a given steel join at the inter- section of R = | with the ordinate corresponding to the tensile strength of the the yield stress as the limiting useful strength, it will be seen that the curves in Fig. 2-45 can be represented quite closely by the straight lines AB and 120, and ultimate tensile rom Re. 13) _ simpie equations, Allowable-stress formulas are o! | fey 10 ABC to get DEF. This factor of safety can be smaller then the factor "fatigue data. "Data have been obtained on small __ simulate actual structural com STRUCTURES METALS, aNO easrevers 103 ps om hens Bo 85S 0 erensiont ——eamarssson? 2s elded Goodman daa or specimens of Fig 242. From “USS Stel Design Mamta” US Brel Corton, Nene 68) is enables fatigue-design criteria to be put in the form of of Fig. 2-46, ies ined by applying a factor of ‘occurrence of the for static load because of the smaller probability of th lrpr number of eles of Seeceload magnitude ested to case @ fatigue failure, Specifications for fatigue loading prescribe an allowable range of stress, Fe. = Sax ~Snin~ These criteria are based on evaluations of extensive laboratory sigue data, "Dy specimens, specimens which and on actual components, Most of these experiments are conducted under stress amplitudes of constant magnitude, but few structures are subjected to stress cycles of this type. Variations in the loading conditions over the life of a structure will produce stress cycles of varying magni- ide loading under con- FIGURE 246104 sion oF sree srmucruses Specification provisions are presented in the AISC ASD and LRED Manuals and in the AASHTO and AREA specifications for al configurations and loading conditions. Some of these d 2.47. They are defined of specific details (Table 2 lates them with the illustrative examples. It also states the kind of stress and defines the stress category (A, B, C, etc). The allowable stress ranges for four loading conditions, based on the anticipated number of loading cycles, are given in Table 2-24, teri woe hes length FIGURE 247 Fatigue examples STRUCTURES, METALS, AND FASTENERS 105 TABLE 223 - AASHTO fatigue categories ANSHTO fatiene catego Uastatve : Kisd of Stes example ane it eect category Fig. 247) i Situation coviton ‘Puan moteral Base metal with rolled or cleaned suraces;ftame- Tor Rev" A 12 oe ct edges with ASA smoothes of 1000 . Torkey 8 337 uatup members CCaleulated Nesural stress at tor of wansverse Tor Rev C 6 stilfener welds on gitder webs or flanges Base metal at end of partial-lengih welded cover plates having square or tapered ends, with oF [ithout welds across the ends E 7 E orRey B Groove welds Tor Re Torkey B raed Tork C 810-1218 hed by groove welds TorRey B 1 longitudinal loading when the detail is between 2 in and 12 times the plate thickness but less than 4 ia ase tal Torkw E B subject to longi length Zi great thickness oF are | Fitowelded Base metal Tork E ‘sonsetions106 DESIGN oF sree sraucruREs ‘TABLE 2-23-coninued) Genel Kind of condition Situation wee Base metal adjacent to Tor Rev vik length Z in pe EE ee eee Hee TT signin ean + Gast plates sch to rer anges Seure: AASHTO Standard Specs It should be noted that fatigue is not often a factor ings, except for crane runways and the like. For example 100,000 cycles of load ions of maximum service load daily for 25 years, and do not experience such a pattern of load. On the other hand, iat bridges may well experience many more daily cycles of Toad. is obvious ‘maximum ser tigue strength of tension member will ch the surface of the connecting is eliminates the stress concentrations that would arise in the as-welded ‘The following suggestions should be kept in mind when designing struc- tures subject to fatigue:!* 1, Avoid details of design that produce severe stress concentrations of poor sires distribution, 2. Provide gradual changes in the section, and avoid reentrant, notchlike comners. 3. Avoid abrupt changes of section or stiffness in members or components. 4, Align parts to eliminate eccentricities or reduce them to a minimum. 5. Avoid making attachments on parts subjected to severe fatigue loadings. 6, Use continuous welds rather than intermittent welds 7, Avoid details that introduce high, localized constraint GeTable 100,000 500,000 2,000,000 TABLE 22S : ‘AASHTO allowable fatigue stress range Tere qieenn cies oa uae Allable range of stress Fy sit category For For For es _yches—_cychs my 8 jundant load-path structures? 36 m 4 mB 2 » 3 16 10 ps 8 34 38 36 4 m4 ms OB 6 @ 8 10 3 6 10 7 s 12s 3 5 2s 8. Provide suitable inspection to guarantee proper riveting, adequate clamping in high-strength bolts, and the deposition of sound welds. : 9, Provide for suitable inspection during the fabrication and erection of struc- tures. ‘ 10, When fatigue cracks are discovered, take immediate steps to prevent their 2-16 BRITTLE FRACTURE Stel structures sometimes fal suddenly and without warning in the form o ‘excessive deformation, These failures are often catastrophic in the sense that the Structure is more or less completely destroyed. Fatigue failures develop over108. oeston oF steex sraucrunes some time as of repeated tures OCCUE as a reso of some critical combination of stress, temperature, flaw size, and toughness of the metal. Most brite fra under static load at stress levels which are ‘not excessive, but they can result from a dynamic appl overload, manner on January 15, re resulted in the death of 12 people from drowning in molasses or from other injuries. Forty people were injured, and several horses drowned, Damage to houses and a portion of the Boston Elevated Railway was extensive. lures of steel water tanks, ofl tanks, transmission lines, ships, plate-gitder bridges, etc, have occurred. Historical surveys of failures can be found in Shank'* and Parker." Most of these failures occurred under normal ions, rather than overload, and originated at a point of concentra- uch as a defect or a geometrical stress raiser, Furthermore, most of welded and failed at low temperatures. oF no plastic deformation in advance of brittle fracture. In is is another way of saying that the sliding on planes of maximum shear stress described in Art. 2-4, which is the source of the plastic deformation that precedes fracture, is somehow inhil resistance to sliding exceeds its resistance to separation (cleavage) or because sliding is prevented. The rel ing and resistance to cleavage is not constant for a given material but depends on the speed of deformation and on ling increases with increase in speed of defor- tesistance to separation is affected to a smaller degree. Asphalt is a 00d example of se to strain rate; it may flow under its own weight over 4 long period of time, but it is brittle under suddenly applied forces. Again, both kinds of resistance increase with decrease in temperature, ence between the two tends to become smaller and may final ing in a ductile material may be inhibited by local geometry of the member, such as a hole or notch. For example, in a tensile test of a notched specimen, such as that shown in Fig. 2-484, sliding that would or in the notched portion is restrained by the larger, low either side of the notch. The smaller the length d of the no depth and to the size of the specimen, the mor may develop at a small notch in a plate, as in Fig. 2-486. In this ere are tensile stresses f, in the region of the notch, distributed somewhat as shown by the curve abc, in addition to the tensite stresses f, which are distrib- STRUCTURES, WeTALs AND Fasteners 109) ¥ ru ei FIGURE 248 : own, curve de. Therefore, an element such as the one at g is in yielding would normally begin at a, the necessary contraction, which ion of the thickness of the plate (z a) ly stressed portions, in a manner si appearance, in contrast to the silky or fibrous appearance of a shear surface, and usually has a herringbone or chevron appearance, with the apexes of110 esta oF sreet smmucrunes be noted that the restraint of contraction in Fig. 2-48b is accompa- nied by a tensile stress in the direction of the thickness of the plate, so an element in the interior of the plate near the notch isin the tria of stress shown in Fig. 248d. Brittle fracture can also be explained in terms of this condition, (f, ~Jey/2, which acts on the that bisects the angle between the y and z axes, There- fore, if f, approaches f, in magnitude, the shear stress becomes very small, so fracture may be by cleavage rather than by sliding, though failures typical of brittle fracture have been traced back to 1879, the subject did not receive great attention until a large number of occurred in welded ships during World War Il. As a result of inves ducted at that time certain design changes were implemented. For wel these changes involved the use of crack arrestors, improved work qu: sestrictions on chemical composition of ship steels. Des lures were reported in ly 1950s, Although the number of failures has been reduced, they continue to be reported. During the severe cold wave that swept through the midwest during the winter of 1977, several spectacular occurred in bridge structures in Illinois, Minnesota, and Pennsy! January 22, 1988, several brittle failures occurred in’a bridge in Pr Rhode Island. Control of brittle fractures as wel s of fatigue failures has been improved , Whose basic concepts were developed and 1966, Modifications and extensions have continued to the fracture has shown that service temperature, toughness, constraint, residual stress, presence of flaws, e to such failures. Many of these parameters have been sti sing a variety of test specimens and testing conditions. These include the Charpy V-notch (CVN} DT), dynamic tear test, explosion land many others. toughness, crack size, and stress are the primary factors which determine the susceptibility to brittle frac- ture. Fracture mechanics principles provide a methodology to correlate material toughness with the stress and flaw size which would cause brittle fracture. Knowledge of the effect of speed of loading, shown in Figs, 2-50 and 2-51, permits extension of this methodology to various types of structures. Most of the struc tures of interest to the civil engineer are subjected to relatively slow loading con- ditions. A more detailed discussion is given by Barsom and Rolle.!* The Charpy V-not commonly used to evaluate the change in eross section. In this is simply supported as a beam and struck by a pendulum released from a fixed height. The eneray absorbed in fracturing the specimen is determined from the difference in the height of the pendulum before release and the height to which it returns after impact. In the case of steel, it turns out that energy-absorbing capacity in the notched specimen sraucrunes, Metats. AND FASTENERS 111 Notre! agp FIGURE 2.50 Rolfe, Fracture and Fatigue Control in Seal Structures; AISC Eng, J 1et qua other words, a steel that is duct manner at @ notch. Furthermore, ior independently of temperature. Variations in onergy-absorbing capacity with temperature is determined from notched-specimen tests covering a range of temperatures. Figure 2-52 shows a typicu plot of absorbed energy vs, temperature. There is a range of temperature in which fracture is largely by cleavage (separation), another in which it is largely "by shear, and an intermediate zone of transition from one type of fr ther, The lower-temperature boundary of the transition zone is called the duc- ity sransition and the higher-temperature boundary the plastic-fracture tran- jon. The corresponding temperatures are called transition temperatures. A value112 bestow oF sreex sraucrunes seu titre Ductile fate Doctity tegnsivion Energy adsorbed Lsteets Fracture transition FIGURE 252 = ‘Transition ftom duct tob Tempercture behavior of mild structua of 15 fb generally is used to define the lower transition temperature, and 45 felb has sometimes been specified for the higher one.'’ The difference bet 120°F. If the temperature is above the ductility. transition temperature, there will be appreciable plastic flow at the root of a notch before cracking begins. is affected by the chemistry of the steel. St increases in carbon lower the energy-absorbing capacity and raise the transition temperature.'? The size of the piece is also a factor. Thick plates have higher nsition temperatures than in plates, because they requi ress (Art. 3-1) is another fractures have been in ing, in addition to notches. Cold work is also detrimental because of the resultant lower- ing of ductility. Charpy V-notch 15 ftlb transition temperatures are on the order of +30°F for A36 and Addl steels. On the other hand, ASI4 steel has a 15 ft-lb tran le-fracture problems. ‘Therefore, mn temperature is not a necessary Prerequisite to successful, low-temperature applications of these steels in these structures. Thus, the best guide to avoidance of brittle fracture is experience with structures and steels of various types, It must be kept in mind, however, th lower the expected service temperatures, the more important it becomes conditions conducive to much more n, ete), Weld flaws stresses, cold working du ‘The AASHTO and AREA specifications contain provisions developed as a STRUCTURES, METALS AND FasroNeRs 113 fracture-control program. Both specify special criteria for materials to be used in - Gjitical members. The AASHTO specification requices that material used in i isfy supplemental n contains special requirements for - materials, ion, welding, inspection, and testing for fracture-critical ‘members in steel bridges. Fractore-critical members are defined as any tension member or component thereof whose failure would be expected to result in the collapse of the bridge or inability to perform its design function. Impaci-test ‘requirements in the form of Charpy V-notch results are prescribed for three dif- ferent service temperatures "2:17 CORROSION Corrosion may be defined as the deteriora chemical reaction with its environment. In ‘most familiar form it appears as “must on the surface of exposed steel structures and results in the gradual loss of metal. The natural consequence is a reduction of the cross section of the member and a resulting increase in stress in the remaining material. Protection is nor- “mally provided by painting the exposed structure. There are, however, other types of corrosion and other effects of corrosion, which are called direct attack, that may be more serious. The extent of direct attack depends on the solubi the metal from the corrosive agent and reduce the rate of corrosion, in some cases the corrosive product may not adhere, so a protective layer is not maintained, In the case of electrochemical attack, corrosion is accompanied by electric _ currents flowing through the me b it lly separated anodic and cathodic areas. Iron covered with Anode, and the attack can produce serious pitting the scale Corrosion in the presence of stress can produce failure called stress- corrosion cracking. If the crack size, stress, and temperature correspond to the Conditions for brittle fracture discussed in Art. 2-16, failure may occur. The 48.a result of overload and subsequent complete collapse of the structure. __ Stainless steel is subject to a type of corrosion which progresses along the grain boundaries and results in a crack. Failure of a suspended ceiling over a Hmming pool in Switzerland in 1985 resulted from this type of attack. The lorine in the air above the pool produced corrosion of the stainless steelTL4 DEston OF STEEL STRUCTURES hangers supporting the suspended ceiling, Stainless ste! is especially sensitive to this type of attack when halides are present in the corrosive environment. Reference 19 contains a comprehensive discussion of corrosion for a wide range of materials. REFERENCES 1. Chajes, Ay S.. Brtvee, and G. Winter: Bmects of Cold-Straining on Strucural Stet Shets, J ‘Dis ASCE, April 1963 Synopsis of Fist Progress Report of Committee on Factors of Sut, J. Strut. Dt a, Properties of Stel for Use in LRFD, ASCE, Fou, of Sire 78 4th ed, American Welding Society, Miami, 1987. towable Stress Design, 9th ed, American Institut of Steel Con- Construction, 90, 1989. 6 Manual of Steel Construction, Lead and Resistance Factor Devign, American Institute of Stet Construction, Chicago, 1986, ‘on Structural Connections: Specification for Stevetural Joints Using ASTM A325 or A490 Bots, 198%, Fisher, J. W, and Struik, J. H_A.: “Guide to Design Criteria for Bolted and Riveted Joints” 2nd ed, Wiley, New York, 1987, Specifications for Aluminum Structures,” The Aluminim Associati 982 ndard Welding Symbols" American Welding Society, Misr, 1968, AL Higgins, T. R, and F. R, Preece: Proposed Working Stresses for Fillet Welds in Boiling eld. ., October 1968, tural Welding Code-Steel:" American Welding Society, Misti, 198, Memorial Tnstivte: “Prevention of Failure of Metals under Repeated Stress” Wiley, Now York, 1946, 14, Designing and Making Welded Structural Sts! Members for Cyclic Loading, Welding Research Councit Committee on Fatigue of Welded Joins, Weld. vol. 38 August 1984 15. Shank, M. EA Criieal Review of Brite Failure in Carbon Stel Structures Other Thaa Ships, Ship Structure Coramittee Report, Serial No. SSC-65, National Academy of Sciences, National Research Council, Washington, D.C, December £953. 1, Parker, F. R"Britle Behavior of Eogineoring Structures." Wiley, Now York, 1957 17. Murse, W. Hs Fatigue and Brite Fracture, tec. 4 in EH. Gaylord and C. N. Gaylord (eds "Structural Engineering Handbook,” 3d ed, MoGraw-Hil, New Votk, 1990, 18, Barsom, J. M and Rolfe, S. .: “Fracture and Fatigue Control in Structures" 24 e., Prentice |, Englewood Cli, NJ, 1987 19, “Corrosion,” vol 13: “Metals Handbook”: American Society for Metals Inter atk, OH, 19 3. Washington, D.C, Apel nal, Metals CHAPTER 3 TENSION MEMBERS 341 EFFECT OF RESIDUAL STRESSES ‘Tension members are efficient carriers of load and are used in many types of structures. In general, their response to load is much the same as that of the tensile-test coupon ased (0 determine the basic material properties, but itis not identical. The stress-strain curves shown in Fig. 2-18 are typical of coupon results Member behavior may differ from coupon behavior for various reasons, such as slip in bolted and riveted connections, nonlinear behavior of the connec tions, and residual stresses in the member. Residual stresses result principally {rom nonuniform cooling of hot-rolled or welded shapes and from cold straight- ‘ning of bent members. “The rolled I or H shape will be used to explain the manner in which stresses arise from nonuniform cooling after rolling (Fig. 3-1a). Because they have more Strface exposure per unit of volume, the flange tips tend to cool faster than the flange-to-web junctures. Similarly, the central portion of the web tends to cool faster than the junctures. Therefore, the metal at the junctures continues to con- tract as it cools after the flange tips and web interior have cooled to the tem- perature of the surrounding atmosphere. This contraction is partially restrained by the cooler metal, so tensile stresses develop in the regions of the junctures and 3s116 bssioN oF stest srmucrunss rexston mewaens 117 of the 20 cross sections in this investiga- al stresses tend to increase in magi hese values are not representative of W's wi shown in Fig. thicker thar tude with increase in thickness thick flanges and webs. Because of the high concentr tensile residual stresses at the weld in welded members usually equal the itself, which may be es much as 50 percent higher than that of the parent metal. ded shapes are determined by the section geometry and. cated from ow ‘The residual stress in the universal residual stress in compression, may vary from as ‘thin plates) to as much as F, for very thick plates, and f, tension, from 2 or 3 to 15 oF 16 ksi. The residual composed of such plates will be about as shown in Fig, 3-1c, Cooling after ¢ residual tension and the residual compression in the after cutting leaves e, where f may equal jon in a welded H formed from such plates wi ‘approximate that shown in Fig. 3 Large residual tensions develop at the comers of the welded box (Fig. 3-1d), On the other hand, residual stresses in the hot-rolled square box are very | ow and in one investigation averaged less than 5 ksi / i | Because they are quenched and tempered, A514 rolled steel shapes are par- ‘ tially stress-re In one investigation of ASI4 => 10) OM plate § . ‘Thermal residual stresses extend almost the full length of a member. Of el & tom UM plates (AH tam tame cut plotes | course, they must vanish at the ends. However, they build up quite rapi the values indicated in Fig. 3-1 at relatively short distances from the Fabricating operations such as cambering and straightening by cold bending also induce residual stresses. These stresses are superimposed on the thermal residual stresses. They are of about the same mi residual stresses but diffe If the member is zing, jing procedure, the r tion will be changed along the entire length of the member. If gagging, which concentrates the straightening at a few points, thermal residual Hee Ae stresses may remain essentially unchanged over much of the length. ‘Thus, in one The effect of residual stress on the behavior of a tension member can be demonstrated by considering the idealized (webless) H shown in Fig, 3-2b. The ution has also been idealized as linear to simplify the dis- , the assumed residual- fain pattern in d. The f faa eases in the remainder of the cross section. These stresses are called residual stresses. Figure 3-1a shows a typical distribution of residual stress in a standard W shape. These stresses also vary across the thickness, so the pattern shown represents averages of the across-thickness values. Variation across the thickness is discussed in Art. 4-7 (ee Fig. 4 Both magnitude and distribution of thermal residu: ied even more, ranging from 41 means that some W's developed resi compression to 18,2 ksi tension. This tension over the entire web, instead of stress pattern in c, and the corresponding residusLIS pesiow oF sree. sravcrures FIGURE 32 ‘over the cross section reveals that the resultant P is zero, as it must, sinee the internal residual stresses are in equilibrium. If the member is subjected to a grad- ually increasing uniform tensile strain, all fibers will remain elastic for a strait ‘of 0.0008, The strain distribution when a str 0008 has been applic. shown in € and the corresponding stress distribution in f. The load P = 2(24 x 12 x 1) = 576 kips, and the average stress on the cross section is f= He 24 ksi, This gives point A on the stress-strain curve for the member (Fig. If the strain is increased by an additional 0.0004, the resulting strain distribution will be as shown in Fig, 3-2g. The stress distribution is shown in h, where each flange is yielded over a length of 6 in. The load is 36] 9 06008 door cooae FIGURE 33 7 ott ‘enstow Meweens 119 pa2x 36x 6% 1430 x6 x 1 = 792 kips, and the average stress on the cross ection is f= 497 = 33 ksi, This gives point B in Fig. 3-3. If an additional the resulting strain and tribution are as show! flange cross section is now fully yielded, and the average stress is C in Fig. 3-3. Further straining produces no increase in and J. 536 ksi. This gives poi ress until the metal at the juncture of the flange and web begins to strain- hard If this member were free of residual stress, the stress-strain curve in the region would be OAD. Thus, although the residual stresses do not affect ey do lower the proportional fimit (point A) fever, they are of no strength of the member. (They can be impor- 2 other hand, residual stresses have a pro- ‘nounced effect on the strength of columns (Arts, 4-6 and 4-7). ‘Additional discussion of residual stresses is presented in Art. 3-19. 3:2 TYPES OF TENSION MEMBERS ‘The form of a tension member is governed to a large extent by the type of s of which itis a part structure. Some of the more common types are described in the following paragraphs ‘The simplest tension members are made of wire rope or cable, round and square bars, and rectangular bars or plate, Wire rope is used for guy wires, floor suspenders in suspension bridges, hoisting lines, ete. Bridge strand (Art. 2-6) is used for small suspension bridges, suspended roofs, and similar structures. Rope ‘and cable are attached to other members or to anchorages by various kinds of sockets. Main cables of large suspension bridges consist of parallel wires which are strung squeezed together, and wrapped. "Rods and bars are used principally in bracing systems, as in towers and sag ‘ods for purlins in sloping roofs. Round bars may be threaded at the ends and held in place by nuts. Standard clevises are available to fit threaded ends so that | the bar can be pin-conneeted. The end of an unthreaded bar can be bent back along itself and welded to form a loop for a pin connection. Loop rods and rods with clevises are usually made ia two sections joined with a turnbuckle which is ‘adjusted to tighten the member in place. Standard dimensions and details of plain and upset rods, loop rods, and clevises are tabulated in the AISC Manuals. Eyebars are members of rectangular cross section with enlarged heads ‘ach end which are bored for pin connections. They were used extensively in carly bridge trusses and are still used occasionally as hangers. ‘Single shapes, such as an angle, the plate, the W and § shapes, and the tee, thay be used as tension members. However, two or more shapes are often com- dined to form a “built-up” member, of which some of the more common are wath are shown in is to hold the shapes in intermittent connections whose funct120 pesiow OF STEEL sraUcTORES eR TITO MOL a @ FIGURE 34 line are shown in dashed lines. In welded structures, connections to adjoining members can usually be made directly by i riveted or bolted structtres the connection must usually be made to a plate called a gusser plate. A single plate at each joint is sufficient for the lighter roof trusses (sing plane truss). But two parallel gusset plates are required in bridge trusses and large ses (double-plane truss) ingle-angle members are used extensively in towers. Single-angle and double-angle members are common in roof trusses. Single angles, and double angles as in Fig. 3-4a, may be riveted or bolted to a single gusset plate at each end, or they may be welded directly to the webs or flanges of tec or I chord require two gusset plates— the other two legs may be one at each outstanding leg—although occasional connected t0 a single gussct plate. The four-angle section c usually connects to two gusset plates. Two channels may be arranged as in Fig. 3-4d to connect to 2 single gusset Plate which fits between the webs or as in Fig, 3-4e to connect to a gusset plate at each web, Two channels may also have their flanges turned inward, as in Fig. 3-4/ Various shapes may be made up of plates welded together. For example, three plates welded in the form of an I are often used in trusses, and two plates i the form of an angle have been used for transmission towers. juminum shapes are produced same forms as steel and, by extru- sion, in some different forms (Art. 2-6). Cold-formed steel tension members may be of angle, channel, or other shape. The ctoss sections shown in Fig. 3-4g and h are used as chord members in trusses. Cold-formed shapes can also be bi ‘which shows two channels spot-welded together to form an I as 33 DESIGN STRESSES FOR BASE MATERIAL tress is distributed uniformly over 1s of member, however, abrupt changes in cross sec rigid connecting parts at each end. ‘Two hazards must be considered in establishing margins of safe sive elongation at service loads and (2) fracture, Except for unusual ‘elongations less than those at the beginning of yield are tolerable, For example, an A36.stee! tension member 10 ft long which is stressed just short of the yield point elongates 120(36/30,000) = 0.144 in. However, the elongation after yielding Values of U, described somewhat TENSION ewaens 121 may be as much as 12 times the hardening begins. Obvious Id value (1.8 in, say) before strain ere must be a margin of safety with respect _ to yielding at service loads. On the other hand, this margin of safety need not be ‘ag large as the margin with respect to fracture of the member because of the > possible differences in the consequences. Although the stress is assumed to be uniformly distributed on the member tion, there will usually be nonuniform distributions at and near the ends ‘ted to another member or to gusset plates. There are no provi- and AREA specifications to account for this effect. ing of the member is assumed to be over an effective area A,, that is, P= F,A,. The effective area for bolted and jveted connections is given by and for welded connections by A,= UA, Gb) ‘where U = reduction coefficient A, = net area (Art, 3-5) A, = gross area ifferently here than in the specifications, are as follows. 4a, For W, M, or S shapes with flange widths not less than two-thirds the depth and structural tees cut from them, connected by the flanges, and for bolted and riveted connections with at least three fasteners per line in the direction of the stress, U = 0.90, mn are connected, U = 1. load is transmitted by transverse welds to only some of the cross- ‘ectional elements, A, is taken to be the area of the directly connected elements. ‘The AISC specifications also contain a clause which requires the center of gravity of groups of welds or bolts at the end of any axial-force member to coincide with the center of gravity of the member unless the eccent Put which excepts statically loaded single angles, jouble angles, and similar122. estoy oF srest sraucruRes members from the requirement. The exception is based on the fact that this has been the practice for bolted and riveted connections for many years, and that tests show that it applies to welded connections as wel. Both AISC and AASHTO specify requirements based on both F, and F, In AISC/ASD the allowable stress F, is 0.6F, on the gross arca and O.SF, on the effective net area, AISC/LRED specifies a design strength , P,, which for yield. ing of the gross section is 0,Py= OF, Ay = O9F, A, G20) and for fracture of the net section is GP, = 0,F Ag = OTSFA, 6-28) where P, = nominal axial strength and ¢, = resistance factor for tension, For members without holes, fully connected by welds, 4, is the smaller of the gross area of the member and the effective area of the welds. For members with no holes for high-strength bolts or rivets the allowable stresses according to the AASHTO specification are 0.55F, for A36, A572 Grade 50, and AS88, and O.46F, for A514 and ASI7. In members with holes for high- strength bolts or rivets two criteria apply to A36, AS72 Grade 50, and A588: O55F, on the gross section and 0.50F, on the net section, Such members fabri- cated from A314 and A517 are designed on the basis of 0.46F, on the net section, The AREA specifies only 0.55F, on the net section. However, only 36 and 50 ksi materials are applicable under this specification, and for them 0.55F, is always less than 0.46F,, Pactors of safety for structural members of aluminum are suggested in Ref, 2. For buildings and similar structures comparable to those to which the AISC specifications apply, the factors of safety are 1.65 on yield strength and 1.95 on tensile strength. For bridge structures, the recommended factors of safety are 1.85 on yield strength and 2.2 on tensile strength. 3-4 MEMBER CONNECTIONS The proportioning of tension members with welded connections is relatively simple. Welded joints were discussed in Art. 2-11. For butt-welded connections deposited with matching electrode the deposited weld metal is at least as strong as the base metal, so no calculations for the weldment are needed, To allow for the case where a member may have excess cross-sectional atea, the AISC specifi. cations allow pattial-penetration welds. For this case, of course, the capacity of the weldment need only equal, or exceed, the member design load. ‘The principal problem in welded connections is in deciding what form or forms of member are best-suited and how the connections can be arranged t0 assure good welding position. In general, welds should be distributed in such a manner that the center of gravity of the resisting forces coincides with the center of gravity of the member. Computations to determine the required welds and their arrangement for concentricity are usually simple. For symmetrical members ‘TENSION wewsens 123 © goncentricity is easily achieved by distributing the required welds symmetrical Ju the case of an T-shaped member connected to gusset plates or to another member by welds om the flanges, it is only necessary to divide the required length {weld equally along the four flange tips. The design of tension members with bolted or riveted connections is more complicated than that of members with welded connections because of the loss of © ross section due to the holes for the fasteners. There may be several net sections ‘on which fracture may occur, and the problem is to make sure that the critical section has been found (Art. 3-9). "Connecting elements (gusset plates, splice plates, etc) may determine the ‘usable strength of @ tension member. In Fig. 3-Sa the gusset plate may fail in tension on the net area of section a-a, and in Fig. 3-5e by tension on the gross ‘area of section a-a, The angle member in Fig. 3-Sa may also separate from the | gusset plate by shear on the net area 1-2 combined with tension on the net area 22, a8 shown in Fig, 3-5b. A similar fracture of the welded connection of Fig, _4.5e is shown in Fig, 3-5d. Fracture of a gusset plate for a double-angle member,124 esion oF sree: sreucrunes ot of one of the gusset plates for a W shape (Fig. 3-5 gusset plate in Fig. 3-Sd may also fail on the net sect in Fig, 3-5b, d, and fare called block-shear failures. is shown in Fig. 3-5. The mn a-a. The failures shown, AISC/ASD. Connecting-element allowable stresses where failure may be by shear ‘on a plane through the fasteners, or by a combination of such a shear with tension on a perpendicular plane, are the following (sec. 15.2. (On the net shear area, F, = 030F, 6-39) and on the net tension area, F,=050F, 6-38) Allowable block shear is the sum of the total shear on the net shear area and the ‘on the periphery of welded connect jorhovde® have shown that this evaluation of at the AISC/LRFD evaluation (Eqs. test results, ms to be checked. Hardash and lock shear is conservative and in better agreement with available AISC/LRED. For coane section a-a of Fi For yielding on the gross section, OR, = 04, F, = 094, and for fracture of the net section, OR, = OA, F, = 0.154, Fy (3-40) where A, = gross area of connecting element ‘4, = net area of connecting element, but not to be taken larger than 03854, ly loaded in tension (for example, 1 (see, 55.2). (3-4a) Equations (3-4) are the same as Eqs 2) for 6, P, except for the the net area. The reason for the ing value is discussed in Art. 3-6. If the design of a connecting element involves shear on the element, the design strength R, is given by @R, = HOTA, F,) = 0.80 x 0.74, F, 65) If block shear is involved, the specification states (sec. J5.2) that the re ing value of jeular path, Furthermore, when in shear (or tension) js taken to be the = 0.75 for both. ‘on the net section, the resistance of the perpendicular se vyield strength in tension (or shear) on the gross section, wi enstion Memvers 125 ‘This leads to the following formulas for the design strength #R, ew [02S08F Ae + Re = 19 75(0.6F, Ave + Fy Ay) where ,, = gross area in shear ‘Ay, = gross area in tension ‘Ans = net area in shear ‘Ay, = net area in tension ‘The factor 0.6F, in Eq, (3-6a) is a rounded value of the yield stress in shear according to the maximum-strain-energy theory, namely, F,/\/3 = 0.577F,. The iutimate shear stress is assumed in Eq, (3-6b) to be F,/\/3 = 0.6F ‘Only one of Eqs. (3-6) applies to a given connection, namely, the one in which the fracture term is larger than the yield term. This is explained by n force area 2-2 in Fig, 3-5b approaches zero, Eq. (3-64) would fing on the gross area 1-2 rather than on shear fort ‘A similar argument shows that for @ conn Single row of fasteners perpendicular to the member force, Eq. ‘apply. Thus the correct formula in a given situation is that one in which the facture term is larger than the yield term. If the ftacture plane is not readily identified, the correct value can be determined by using both equations and choosing the larger value of R. ‘No net areas are involved in the block-shear failure of the welded connec. wn of Fig, 3-Se. Therefore, in applying Eq is case the second term should be F, A, and the first term in Eq, (3-66) should be O.6F, 4,. Example 3-1 (AISC/ASD). Design an A36-steel double-angle tension member 40 kips dead load and 110 kips live load, Connection to & (elds deposited with E7OXX electrodes. Solution. Allowable stresses are OF, = 0.6 x 36 ~ 22 ksi on the gross area and OF, = 05 x 58 = 29 ksi on the effective net area, Since the connection is welded, the het area equals the gross area, The value of U is 0.85 for this connection, so A, = 0884, Therefore, T=2xA, T= 29 x 085A, = 2465 x A, ‘The smaller of the allowables determines che required gross area 40.4110 Load per angle = “37° = 15 kips Ae Haat Fora 5 x 34 x ck angle, d= 23.53 in, Use two angles with long legs back to back Fig 6,126 estan oF steex srmuctunes From tables ia the AISC Manual, r, = 1.59 in and, with a j-in gusset plate, r= LAT in, L_1sx12 2 = BAT = 125.5300 (The speciation prefered upper is welded at the heel (weld L, in lengths for concentricity with the center of er by taking moments. For weld L,,, with moments about the heel, 5 x AGL, = SST = 1.63 « 162 % 464 1, = $28 in ‘Subtraction gives the length of weld along the hee! Ly 162-53 =109in Ifeach angle is also welded along its end may be reduced by balf the length of the weld the moment about the heel: end. This is proved by taking 464 x SL = 468 «16.2 x 163-464 x 5 x 25 1, =53-25=28 in welds must be checked. The allow- fansfer the shear along \d shear per inch gives able shear is 1L, and L,. Equating the gusset-plate shear per inch to the 1x 144, Lx 464 2 = 0.64 in ‘The required thickness to prevent a block-shear failure (Fig. 35d) is found by att a} pum 2 Fe i FIGURE 3.6 ‘TENSION MEMBERS R47 equating the member load adjacent tothe transverse tudinal welds, The allovat tvhich is greater than the allowable shear 144 for, block shear need not be cheeks Based on the ewo computed values off, 064 and 0706 i, use a -in gusset 13 and the shears on the sections adjacer 74 plate Example 34-2 (AISC/LRFD). Design an A3 15 ft long to transmit 40 kips dead load and 110 kips li _quset plate is by Billet welds deposited with E7OXX electrodes 1m, The factored load is 1.2DL + L6LL = 12 x 40+ 16 Tor members of this type U = 085, so 4, = 0.854, From Eas. 0.90 x 36a, = 224 A, = 691 in? = 346 075 x 58 x 085A, = 224 A, = 604 in? = 302 in* per angle = 224 kips. per angle For a $34 x dangle, A = 3.53 in?, Use two angles with Tong legs back to beck Leas

You might also like

• A Unified Design of Steel Structures

  

  100% (5)

  A Unified Design of Steel Structures

  458 pages
• Steel Structures Design and Behavior 4th Edition by Charles G. Salmon and John E. Johnson PDF

  

  100% (2)

  Steel Structures Design and Behavior 4th Edition by Charles G. Salmon and John E. Johnson PDF

  1,045 pages
• Steel Structures Third Edition

  

  33% (3)

  Steel Structures Third Edition

  406 pages
• Answer To Home Work 2

  

  No ratings yet

  Answer To Home Work 2

  7 pages
• Steel Structure-Gaylord (ASCE)

  

  92% (13)

  Steel Structure-Gaylord (ASCE)

  818 pages
• Flexural Buckling Example

  

  No ratings yet

  Flexural Buckling Example

  8 pages
• Charles G. Salmon, John E. Johnson, Faris A. Malhas-Steel Structures - Design and Behavior (5th Edition) - Prentice Hall (2008)

  

  80% (5)

  Charles G. Salmon, John E. Johnson, Faris A. Malhas-Steel Structures - Design and Behavior (5th Edition) - Prentice Hall (2008)

  888 pages
• Instructor's Solutions Manual: Structural Analysis

  

  0% (1)

  Instructor's Solutions Manual: Structural Analysis

  4 pages
• Battened Strut Design

  

  No ratings yet

  Battened Strut Design

  4 pages
• Solutions Manual of Structural Analysis 7th Edition

  

  100% (1)

  Solutions Manual of Structural Analysis 7th Edition

  403 pages
• Structural Steel Design-Chapter 4

  

  100% (1)

  Structural Steel Design-Chapter 4

  54 pages
• The Behaviour and Design of Steel Structures To BS5950

  

  100% (5)

  The Behaviour and Design of Steel Structures To BS5950

  237 pages
• Steel Design, Segui PDF

  

  No ratings yet

  Steel Design, Segui PDF

  706 pages
• RCD Beam Analysis and Design

  

  No ratings yet

  RCD Beam Analysis and Design

  33 pages
• Design of Steel Structures B C Punmia Jain PDF

  

  No ratings yet

  Design of Steel Structures B C Punmia Jain PDF

  14 pages
• M1-Ce TRK1-103

  

  100% (1)

  M1-Ce TRK1-103

  9 pages
• Thermal Treatment

  

  No ratings yet

  Thermal Treatment

  23 pages
• Strength of Materials 4th Edition by Pytel and Singer

  

  No ratings yet

  Strength of Materials 4th Edition by Pytel and Singer

  2 pages
• Calculation of Torsion Constant PDF

  

  No ratings yet

  Calculation of Torsion Constant PDF

  12 pages
• Design of Connections

  

  No ratings yet

  Design of Connections

  16 pages
• Design of Compression Member

  

  No ratings yet

  Design of Compression Member

  29 pages
• Segui-Steel Design 4th

  

  100% (2)

  Segui-Steel Design 4th

  418 pages
• Flexural Members

  

  No ratings yet

  Flexural Members

  32 pages
• IS 800 Amendment No:1

  

  No ratings yet

  IS 800 Amendment No:1

  5 pages
• Chapter 07 - Composite Beams

  

  No ratings yet

  Chapter 07 - Composite Beams

  21 pages
• CE122 Reader

  

  100% (1)

  CE122 Reader

  375 pages
• Sample - Solution Manual For Structural Steel Design 2nd Edition Abi Aghayere

  

  100% (1)

  Sample - Solution Manual For Structural Steel Design 2nd Edition Abi Aghayere

  10 pages
• H2.4.10-Response To Step and Ramp Forces

  

  No ratings yet

  H2.4.10-Response To Step and Ramp Forces

  4 pages
• Bending and Shear Stresses in Beams

  

  100% (1)

  Bending and Shear Stresses in Beams

  28 pages
• Stability of Structures PDF

  

  No ratings yet

  Stability of Structures PDF

  11 pages
• BNBC - 2017 Part6 - Chap 2 PDF

  

  0% (1)

  BNBC - 2017 Part6 - Chap 2 PDF

  118 pages
• Chevron Brace Connection

  

  No ratings yet

  Chevron Brace Connection

  5 pages
• SteelDesign Tension Fu

  

  No ratings yet

  SteelDesign Tension Fu

  24 pages
• Analysis of Structures

  

  No ratings yet

  Analysis of Structures

  264 pages
• Steel - 10-Prof. Zahid Siddiqi PDF

  

  No ratings yet

  Steel - 10-Prof. Zahid Siddiqi PDF

  71 pages
• Theory of Structures by S Ramamruthampdf

  

  No ratings yet

  Theory of Structures by S Ramamruthampdf

  220 pages
• Applied Structural Steel Design 4 Edition

  

  100% (5)

  Applied Structural Steel Design 4 Edition

  460 pages
• CE 333-001 - Reinforced Concrete Design

  

  No ratings yet

  CE 333-001 - Reinforced Concrete Design

  6 pages
• Escaleras - Hassoun

  

  No ratings yet

  Escaleras - Hassoun

  28 pages
• Week 6 Lecture Material - Watermark

  

  No ratings yet

  Week 6 Lecture Material - Watermark

  59 pages
• Statically Determinant Structure PDF Free

  

  No ratings yet

  Statically Determinant Structure PDF Free

  21 pages
• Applied Structural Steel Design - by L Spiegel

  

  100% (1)

  Applied Structural Steel Design - by L Spiegel

  460 pages
• Plastic Design of Continuous Beams & Portal Frames

  

  No ratings yet

  Plastic Design of Continuous Beams & Portal Frames

  9 pages
• Interaction Diagram Column

  

  No ratings yet

  Interaction Diagram Column

  12 pages
• Theory of Structure PDF

  

  No ratings yet

  Theory of Structure PDF

  547 pages
• Sample - Solution Manual For Design of Concrete Structures 15th Edition Nilson & Darwin

  

  No ratings yet

  Sample - Solution Manual For Design of Concrete Structures 15th Edition Nilson & Darwin

  10 pages
• AISC Section Table PDF

  

  94% (16)

  AISC Section Table PDF

  133 pages
• Design of Steel Structures - Subramaniam PDF

  

  100% (8)

  Design of Steel Structures - Subramaniam PDF

  626 pages
• Column Base Plate Design PDF

  

  No ratings yet

  Column Base Plate Design PDF

  4 pages
• ST7203 Steel Structures

  

  No ratings yet

  ST7203 Steel Structures

  20 pages
• Charles Head Norris - Elementary Struc

  

  No ratings yet

  Charles Head Norris - Elementary Struc

  536 pages
• Determine

  

  No ratings yet

  Determine

  3 pages
• Chapter 2 Approximate Frame Analysis

  

  No ratings yet

  Chapter 2 Approximate Frame Analysis

  21 pages
• DSS Notes Up To 21st January - 2019 PDF

  

  No ratings yet

  DSS Notes Up To 21st January - 2019 PDF

  25 pages
• Steel Structures Design As Per AISC 360 (ASD & LRFD)

  

  100% (3)

  Steel Structures Design As Per AISC 360 (ASD & LRFD)

  888 pages
• Steel PowerPoint I (2018-2019)

  

  No ratings yet

  Steel PowerPoint I (2018-2019)

  89 pages
• Structural Steel Design Bowles PDF

  

  100% (2)

  Structural Steel Design Bowles PDF

  276 pages
• Steel Mto 01

  

  No ratings yet

  Steel Mto 01

  5 pages
• Toc

  

  No ratings yet

  Toc

  9 pages
• Module 0 - Topics in Advanced Steel Design

  

  No ratings yet

  Module 0 - Topics in Advanced Steel Design

  3 pages
• Brochure - Flange Insulation Products

  

  No ratings yet

  Brochure - Flange Insulation Products

  4 pages
• Design Class Detail Plant Layout Project Date Expandability Design Item Design Class Class 1 Class 2 Class 3

  

  No ratings yet

  Design Class Detail Plant Layout Project Date Expandability Design Item Design Class Class 1 Class 2 Class 3

  2 pages
• Steam Desuperheating: Reliability - High End Technology - Simplicity

  

  No ratings yet

  Steam Desuperheating: Reliability - High End Technology - Simplicity

  12 pages
• Steam Desuperheaters: Bulletin 6D

  

  No ratings yet

  Steam Desuperheaters: Bulletin 6D

  16 pages
• Accurate Steam Temperature Control: Desuperheating For

  

  No ratings yet

  Accurate Steam Temperature Control: Desuperheating For

  4 pages
• Desuperheater Brochure

  

  No ratings yet

  Desuperheater Brochure

  8 pages
• Fittings & Flanges For Pipe Series 2410 and 3410 Using The TaperTaper Adhesive-Bonded Joint FP657-10 0898

  

  No ratings yet

  Fittings & Flanges For Pipe Series 2410 and 3410 Using The TaperTaper Adhesive-Bonded Joint FP657-10 0898

  40 pages
• Design Class Detail Paco Project Date: Pump/Fin Fan Control Shall Be Local Only

  

  No ratings yet

  Design Class Detail Paco Project Date: Pump/Fin Fan Control Shall Be Local Only

  9 pages
• Datasheet CBRN-CAP-1-Canister HS 6574A 0913

  

  No ratings yet

  Datasheet CBRN-CAP-1-Canister HS 6574A 0913

  2 pages
• DBL 818702 00 KDS13-KDL13 en

  

  No ratings yet

  DBL 818702 00 KDS13-KDL13 en

  4 pages
• ASTM Index PDF

  

  No ratings yet

  ASTM Index PDF

  34 pages
• Display Multiple Items in Pivot Table Filter

  

  No ratings yet

  Display Multiple Items in Pivot Table Filter

  42 pages
• Design Class Detail Overpressure Protection Project Date Capacity Design Item Design Class Class 1 Class 2 Class 3

  

  No ratings yet

  Design Class Detail Overpressure Protection Project Date Capacity Design Item Design Class Class 1 Class 2 Class 3

  1 page
• Eil Manual Vol 1 PDF

  

  100% (1)

  Eil Manual Vol 1 PDF

  601 pages
• COVID 19 KEF Reference Manual Rev1 PDF

  

  No ratings yet

  COVID 19 KEF Reference Manual Rev1 PDF

  13 pages
• Design Class Detail Piping Project Date Capacity Design Item Design Class Class 1 Class 2 Class 3

  

  No ratings yet

  Design Class Detail Piping Project Date Capacity Design Item Design Class Class 1 Class 2 Class 3

  9 pages
• Kuwait Bus Routes

  

  No ratings yet

  Kuwait Bus Routes

  3 pages
• Pipe Material Strength Calculation

  

  No ratings yet

  Pipe Material Strength Calculation

  8 pages
• Approval Standard For Flame Arresters: Class Number 6061

  

  No ratings yet

  Approval Standard For Flame Arresters: Class Number 6061

  31 pages
• Flame Arrester (Plate Element)

  

  No ratings yet

  Flame Arrester (Plate Element)

  4 pages
• Kuwait Bus Routes

  

  No ratings yet

  Kuwait Bus Routes

  3 pages
• PROFIS Anchor 2.6.1: WWW - Hilti.us

  

  No ratings yet

  PROFIS Anchor 2.6.1: WWW - Hilti.us

  5 pages
• Stellite - Wikipedia, The Free Encyclopedia

  

  No ratings yet

  Stellite - Wikipedia, The Free Encyclopedia

  3 pages